the fly-by-wire "throttle" project thread
the fly-by-wire "throttle" project thread
01-14-2010, 03:10 PM
#1
I'll edit this post to maintain it as a summary of the project.
engine: OM617
turbo: garret GT2256V, electronic vane actuator
major system components:
- accelerator pedal with TPS (throttle position sensor)
- electronic turbo vane actuator
- servo controlled injection pump rack
- servo controlled kickdown lever
inputs:
- TPS
- boost pressure, temperature
- rpm
outputs:
- rack position
- vane position
i was discussing engine control methodology with a friend and he introduced me to notion of calculating engine "load".
I'll define 100% load as the point at which the amount of fuel being injected entirely consumes the available air (ie, past this point you make black smoke). so the load has to be calculated on the fly, based on rpm, boost pressure and temperature, and rack position.
for most efficient operation, you'd want to set the turbo vane position to produce the minimum boost required to keep load below 100%.
so the control algorithm would basically be a feedback loop which tries to maintain a load of, say, 90%. as load falls below 90%, reposition the vanes to make less boost. as it climbs above 90%, reposition the vanes to create more boost.
the user input to the system is the TPS.
I'll describe my understanding of the stock mercedes accelerator, rack, and ALDA.
the stock mercedes accelerator pedal controls rack position, but its input is limited by the ALDA to prevent load from exceeding 100%.
that is, normally, if the pedal position is at 75%, the rack position would be at 75%. however, if the engine reaches 100% load at a rack position of only 50% (due to lack of boost), the ALDA limits the rack position to 50%. so any user input past 50% is ignored.
what I don't like about the mercedes system is that the user has no feedback to indicate that his (pedal) input has effectively been truncated.
eventually, what I'd like to do is develop a force-feedback accelerator pedal. that is, i'd still tie TPS to rack position, but a force feedback mechanism would physically resist pushing the pedal past a load point of 100%.
with the stock mercedes system, if the user depresses the pedal all of the way, he feels a very non-linear response from the car. that is, it feels sluggish until boost starts to build, at which point it "takes off" very abruptly. this is very different from the typical naturally aspriated gasoline accelerator pedal experience.
with the force-feedback system, the user would depress the pedal, but would be met with resistance at, say 50% of pedal travel. if the user keeps pressure on the pedal, he would then feel the pedal start to "fall away" as the boost builds. this would give a much better indication of the power available.
additionally, the force feedback would be weak enough that if the user needed to (ie, in case of emergency), he could override it and depress the pedal further, pushing the IP past the 100% load point (ie, if you really mash the pedal, you'll make black smoke).
This post was last modified: 01-14-2010, 03:38 PM by cell.
I'm starting this thread to document my attempt at creating a fully fly-by-wire "throttle" setup.
I'll edit this post to maintain it as a summary of the project.
engine: OM617
turbo: garret GT2256V, electronic vane actuator
major system components:
- accelerator pedal with TPS (throttle position sensor)
- electronic turbo vane actuator
- servo controlled injection pump rack
- servo controlled kickdown lever
inputs:
- TPS
- boost pressure, temperature
- rpm
outputs:
- rack position
- vane position
i was discussing engine control methodology with a friend and he introduced me to notion of calculating engine "load".
I'll define 100% load as the point at which the amount of fuel being injected entirely consumes the available air (ie, past this point you make black smoke). so the load has to be calculated on the fly, based on rpm, boost pressure and temperature, and rack position.
for most efficient operation, you'd want to set the turbo vane position to produce the minimum boost required to keep load below 100%.
so the control algorithm would basically be a feedback loop which tries to maintain a load of, say, 90%. as load falls below 90%, reposition the vanes to make less boost. as it climbs above 90%, reposition the vanes to create more boost.
the user input to the system is the TPS.
I'll describe my understanding of the stock mercedes accelerator, rack, and ALDA.
the stock mercedes accelerator pedal controls rack position, but its input is limited by the ALDA to prevent load from exceeding 100%.
that is, normally, if the pedal position is at 75%, the rack position would be at 75%. however, if the engine reaches 100% load at a rack position of only 50% (due to lack of boost), the ALDA limits the rack position to 50%. so any user input past 50% is ignored.
what I don't like about the mercedes system is that the user has no feedback to indicate that his (pedal) input has effectively been truncated.
eventually, what I'd like to do is develop a force-feedback accelerator pedal. that is, i'd still tie TPS to rack position, but a force feedback mechanism would physically resist pushing the pedal past a load point of 100%.
with the stock mercedes system, if the user depresses the pedal all of the way, he feels a very non-linear response from the car. that is, it feels sluggish until boost starts to build, at which point it "takes off" very abruptly. this is very different from the typical naturally aspriated gasoline accelerator pedal experience.
with the force-feedback system, the user would depress the pedal, but would be met with resistance at, say 50% of pedal travel. if the user keeps pressure on the pedal, he would then feel the pedal start to "fall away" as the boost builds. this would give a much better indication of the power available.
additionally, the force feedback would be weak enough that if the user needed to (ie, in case of emergency), he could override it and depress the pedal further, pushing the IP past the 100% load point (ie, if you really mash the pedal, you'll make black smoke).
01-14-2010, 07:42 PM
#2
You might also want to add rack position and vane position to the feedback data. You can add subroutines to analyze and compare the requested rack and vane position to the actual position and set "codes" if they are out of sync.
As far as the force feedback is concerned, it should not be needed if the fuel curve is designed properly to add fuel based on boost and the turbo vane operation is properly designed to provide linear boost.
The acceleration delay is a side effect of a compromise in the design of the induction system.
As for the rest of the parts, you should able to eliminate the alda and the governor and hard link the electronic servo to the rack.
This post was last modified: 01-14-2010, 07:44 PM by E300TSC.
1998 MB E300 - 194,000 miles. Current/future mods: 7.5mm pump elements, Holset HX40 Turbo feeding Eaton M90 supercharger through custom fabbed intake system. Aeroturbine muffler, scratch-built 4" stainless exhaust. EGR eliminated, ECM recalibrated, modified rack position feedback circuit. Porterfield RS brake pads, Bilstein sport shocks, VDO EGT, oil and boost gauges.
2000 Ford F250 - 150,000 miles. Current/future mods: DP Tuner 3 map chip, scratch-built high flow intake, high output IDM, open circuit fuel system, ball bearing turbo, South Bend full metal clutch, slotted and drilled rotors. Autometer Sport Comp EGT and boost gauges.
Just a couple comments:
You might also want to add rack position and vane position to the feedback data. You can add subroutines to analyze and compare the requested rack and vane position to the actual position and set "codes" if they are out of sync.
As far as the force feedback is concerned, it should not be needed if the fuel curve is designed properly to add fuel based on boost and the turbo vane operation is properly designed to provide linear boost.
The acceleration delay is a side effect of a compromise in the design of the induction system.
As for the rest of the parts, you should able to eliminate the alda and the governor and hard link the electronic servo to the rack.
1998 MB E300 - 194,000 miles. Current/future mods: 7.5mm pump elements, Holset HX40 Turbo feeding Eaton M90 supercharger through custom fabbed intake system. Aeroturbine muffler, scratch-built 4" stainless exhaust. EGR eliminated, ECM recalibrated, modified rack position feedback circuit. Porterfield RS brake pads, Bilstein sport shocks, VDO EGT, oil and boost gauges.
2000 Ford F250 - 150,000 miles. Current/future mods: DP Tuner 3 map chip, scratch-built high flow intake, high output IDM, open circuit fuel system, ball bearing turbo, South Bend full metal clutch, slotted and drilled rotors. Autometer Sport Comp EGT and boost gauges.
01-14-2010, 09:41 PM
#3
(01-14-2010, 07:42 PM)E300TSC You might also want to add rack position and vane position to the feedback data. You can add subroutines to analyze and compare the requested rack and vane position to the actual position and set "codes" if they are out of sync.
that's a neat idea, but for version 1.0 of this system I'm going to do without this to keep the complexity down. with servos, you don't have to worry about them being out of position unless they are under-spec'ed on torque.
however, that gives me an idea: a poor man's method of providing a "servo out of position" alarm would be to simply monitor the current used by the servo. if the current maxes out and stays that way, you know that the servo is unable to reach the requested position. that would at least give you an alarm-style indication, without the added complexity of adding some sort of shaft encoder.
(01-14-2010, 07:42 PM)E300TSC As far as the force feedback is concerned, it should not be needed if the fuel curve is designed properly to add fuel based on boost and the turbo vane operation is properly designed to provide linear boost.
well, that's more of a matter of "I want it" rather than "I need it"
(01-14-2010, 07:42 PM)E300TSC As for the rest of the parts, you should able to eliminate the alda and the governor and hard link the electronic servo to the rack.
yup!
my plan is to "digitize" each component of the system by linking it to an arduino or perhaps a stand-alone atmel chip, and then link them all together on a i2c bus.
the other night I got a proof-of-concept i2c demo up and running with two arduinos.
![[Image: Blinki2c.jpg]](http://jason.pepas.com/wiki/images/8/8b/Blinki2c.jpg)
forum thread: http://www.arduino.cc/cgi-bin/yabb2/YaBB...1263425676
webpage: http://jason.pepas.com/wiki/index.php?ti...nkMeetsI2C
This post was last modified: 01-14-2010, 09:44 PM by cell.
(01-14-2010, 07:42 PM)E300TSC You might also want to add rack position and vane position to the feedback data. You can add subroutines to analyze and compare the requested rack and vane position to the actual position and set "codes" if they are out of sync.
that's a neat idea, but for version 1.0 of this system I'm going to do without this to keep the complexity down. with servos, you don't have to worry about them being out of position unless they are under-spec'ed on torque.
however, that gives me an idea: a poor man's method of providing a "servo out of position" alarm would be to simply monitor the current used by the servo. if the current maxes out and stays that way, you know that the servo is unable to reach the requested position. that would at least give you an alarm-style indication, without the added complexity of adding some sort of shaft encoder.
(01-14-2010, 07:42 PM)E300TSC As far as the force feedback is concerned, it should not be needed if the fuel curve is designed properly to add fuel based on boost and the turbo vane operation is properly designed to provide linear boost.
well, that's more of a matter of "I want it" rather than "I need it"
(01-14-2010, 07:42 PM)E300TSC As for the rest of the parts, you should able to eliminate the alda and the governor and hard link the electronic servo to the rack.
yup!
my plan is to "digitize" each component of the system by linking it to an arduino or perhaps a stand-alone atmel chip, and then link them all together on a i2c bus.
the other night I got a proof-of-concept i2c demo up and running with two arduinos.
forum thread: http://www.arduino.cc/cgi-bin/yabb2/YaBB...1263425676
webpage: http://jason.pepas.com/wiki/index.php?ti...nkMeetsI2C
01-15-2010, 01:41 PM
#4
I had the cruise control system fail on my 1982 300cd and set WOT while I was in traffic not FUN
I shortly removed all of it!
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-15-2010, 01:41 PM #4whats the advantage to this over stock linkages? I think Toyota just recalled all of there fly by wire car to be retrofired back to cable
I had the cruise control system fail on my 1982 300cd and set WOT while I was in traffic not FUN
I shortly removed all of it!
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-15-2010, 01:54 PM
#5
Picture it this way: maxed out TC, but no return to idle issues, maxed out ALDA but no smoking, instantaneous high-rpm cut out at whichever rpm you choose, no gradual loss of power in the top end due to governor slowly cutting fuel. If you have a VNT, it is now very easy to tune electronically, if you have a standard wastegate turbo you can hook up an actuator to the wastegate lever and have very much better boost response and control. The list goes on.
GREASY_BEAST
01-15-2010, 01:54 PM #5The advantage is that you can remove the mechanical governor and have much more control over how much fuel goes in and when it goes in. Also, you have boost, road speed and therefore transmission gear (if you have an electronic speedometer), rack position, and EGT inputs into your electronic governor which allows very precise control over the engine's performance.
Picture it this way: maxed out TC, but no return to idle issues, maxed out ALDA but no smoking, instantaneous high-rpm cut out at whichever rpm you choose, no gradual loss of power in the top end due to governor slowly cutting fuel. If you have a VNT, it is now very easy to tune electronically, if you have a standard wastegate turbo you can hook up an actuator to the wastegate lever and have very much better boost response and control. The list goes on.
01-15-2010, 02:18 PM
#6
(01-15-2010, 01:41 PM)willbhere4u whats the advantage to this over stock linkages? I think Toyota just recalled all of there fly by wire car to be retrofired back to cable
I had the cruise control system fail on my 1982 300cd and set WOT while I was in traffic not FUN
I shortly removed all of it!
the only advantage is that its fun!
if you are a budding crackpot inventor who's looking for a design challenge, its a great idea. if you just want reliable transportation, its probably a horrible idea.its funny, I started out in the middle, swung to one extreme, and am now swinging back over to the other.
I started out driving relatively modern cars with electronics, etc. Then I moved out on my own after college and was hit by my first repair bill. Half of it was labor, and I just had to trust that his diagnosis was correct (and honest) . I thought "this is ridiculous. if this guy can figure this out, so can I". But I realized that with modern cars, it isn't that simple. There are so many electronic systems that there are limits to what you can fix yourself.
So I started on a mission to learn how cars work, and to become self-sufficient with respect to them. I went on ebay and bought a 1967 chevy pickup. Carbureted. Points ignition. Manual transmission. Manual door locks, windows, etc. No A/C. In fact, I'm pretty sure the only piece of silicon on the car was the voltage regulator.
I wanted to reduce the complexity of the system to an absolute minimum. start off as simple as possible, master that much, and then work my way up from there. The stated goal of this project was that the car would never be touched by a mechanic.
Then I fell in love with diesels (after sitting in on a thermodynamics course at UT), and I swapped in a mercedes OM617. You could say I actually reduced the complexity of the system even further by doing this -- I drove it without an alternator for the first few weeks! Once you had the engine started, you could even remove the battery. You think zero electronics is an ideal to strive for? That's nothing! You could operate this car without any electrical system at all!
anyway, once I felt I had mastered this level of complexity, I had the desire to improve on the design. add features. I realized it wasn't complexity I hated, it was black boxes that I hated. it turns out I have no problem with electrical, electronic, or even digital and software systems being part of my car, as long as I understand them, can debug them, and modify or improve them.
so now, I've taken something that I hated (electronic black boxes) and turned into learning exercise, and something that's a lot of fun! the key in doing that is the ability to "make it your own".
This post was last modified: 01-15-2010, 02:19 PM by cell.
(01-15-2010, 01:41 PM)willbhere4u whats the advantage to this over stock linkages? I think Toyota just recalled all of there fly by wire car to be retrofired back to cable
I had the cruise control system fail on my 1982 300cd and set WOT while I was in traffic not FUN
I shortly removed all of it!
the only advantage is that its fun!
if you are a budding crackpot inventor who's looking for a design challenge, its a great idea. if you just want reliable transportation, its probably a horrible idea.its funny, I started out in the middle, swung to one extreme, and am now swinging back over to the other.
I started out driving relatively modern cars with electronics, etc. Then I moved out on my own after college and was hit by my first repair bill. Half of it was labor, and I just had to trust that his diagnosis was correct (and honest) . I thought "this is ridiculous. if this guy can figure this out, so can I". But I realized that with modern cars, it isn't that simple. There are so many electronic systems that there are limits to what you can fix yourself.
So I started on a mission to learn how cars work, and to become self-sufficient with respect to them. I went on ebay and bought a 1967 chevy pickup. Carbureted. Points ignition. Manual transmission. Manual door locks, windows, etc. No A/C. In fact, I'm pretty sure the only piece of silicon on the car was the voltage regulator.
I wanted to reduce the complexity of the system to an absolute minimum. start off as simple as possible, master that much, and then work my way up from there. The stated goal of this project was that the car would never be touched by a mechanic.
Then I fell in love with diesels (after sitting in on a thermodynamics course at UT), and I swapped in a mercedes OM617. You could say I actually reduced the complexity of the system even further by doing this -- I drove it without an alternator for the first few weeks! Once you had the engine started, you could even remove the battery. You think zero electronics is an ideal to strive for? That's nothing! You could operate this car without any electrical system at all!
anyway, once I felt I had mastered this level of complexity, I had the desire to improve on the design. add features. I realized it wasn't complexity I hated, it was black boxes that I hated. it turns out I have no problem with electrical, electronic, or even digital and software systems being part of my car, as long as I understand them, can debug them, and modify or improve them.
so now, I've taken something that I hated (electronic black boxes) and turned into learning exercise, and something that's a lot of fun! the key in doing that is the ability to "make it your own".
01-15-2010, 03:58 PM
#8
How fast do you plan on running your I2C? I had a couple boards running over I2C before switching to ethernet because I got errors at higher data rates/longer cables.
Issues I had:
CAN: Difficulty debugging without expensive scope
I2C: Errors with high speed and long cables
SPI: Limited number of nodes + speed/cable issues
Ethernet: Complex board routing
Ich liebe meine Autos!
1981 Mercedes-Benz 300SD | 156K Miles | 2nd Owner | EGR Disabled [Removal Pending] | ALDA Removed | Straight Pipes | GT2256V??? | Laser Interceptor | Engine swap over summer, hopefully with GT2256V attached...
1981 Mercedes-Benz 300SD | Odo Stopped at 160K (at least 50K more) | EGR Disabled | ALDA All The Way Out | Straight pipes | FM-870 Remote Start Alarm System | B100 Biodiesel | AC Fixed x2 | Trunk crushed in
| Retired to garage.Excessive speeding? It ain't excessive till I redline!
Thats basically the same reason I'm doing my system. I hate working on newer cars because I can't get into any of the computer systems.. But I have no issues putting my own computer controls into an older car..
How fast do you plan on running your I2C? I had a couple boards running over I2C before switching to ethernet because I got errors at higher data rates/longer cables.
Issues I had:
CAN: Difficulty debugging without expensive scope
I2C: Errors with high speed and long cables
SPI: Limited number of nodes + speed/cable issues
Ethernet: Complex board routing
Ich liebe meine Autos!
1981 Mercedes-Benz 300SD | 156K Miles | 2nd Owner | EGR Disabled [Removal Pending] | ALDA Removed | Straight Pipes | GT2256V??? | Laser Interceptor | Engine swap over summer, hopefully with GT2256V attached...
1981 Mercedes-Benz 300SD | Odo Stopped at 160K (at least 50K more) | EGR Disabled | ALDA All The Way Out | Straight pipes | FM-870 Remote Start Alarm System | B100 Biodiesel | AC Fixed x2 | Trunk crushed in
| Retired to garage.Excessive speeding? It ain't excessive till I redline!
01-15-2010, 04:48 PM
#9
This post was last modified: 01-15-2010, 04:48 PM by GREASY_BEAST.
GREASY_BEAST
01-15-2010, 04:48 PM #9You can also have "sport mode" and "efficient mode" with the flip of a switch. This would make the car very amenable to the addition of big IP elements, and would make the operation less expensive because the injection shop would not have to do any governor tuning, just balancing the elements to eachother.
01-15-2010, 05:15 PM
#10
I would NOT use your standard RC servo for this. What happens if you lose 12V power? The rack position is now stuck wherever it was, and you have to figure out some way to turn off the engine without wrecking, over-revving, etc. Servo's are definitely easy to come by and control, and I think even I could mount one to the IP rack. However, for safety's sake please use a solenoid or something that has a failsafe. Maybe adding a plate or something to completely shut off airflow to the engine would work. Be sure to test it with the mechanical pump though
I'm going to need something like this for my 606 electronic pump as well. At first it is going to be a 2x6" manually put in front of the turbo. 
(01-14-2010, 03:10 PM)cell for most efficient operation, you'd want to set the turbo vane position to produce the minimum boost required to keep load below 100%.
Yes!!! A lot of people don't figure that out and think that more boost is always better. Light years down the road, you can also make an engine brake with a VNT.
Quote:...force-feedback system...
No offense, but I think that's pretty silly. By no means am I trying to discourage you, but I personally want to just mash the pedal to the floor and let the ECU figure out what the engine can do. If nothing else, one less thing to distract me during an evasive maneuver or something.
(01-15-2010, 01:54 PM)GREASY_BEAST Picture it this way: maxed out TC, but no return to idle issues, maxed out ALDA but no smoking, instantaneous high-rpm cut out at whichever rpm you choose, no gradual loss of power in the top end due to governor slowly cutting fuel.
Exactly!!!
(01-15-2010, 04:48 PM)GREASY_BEAST You can also have "sport mode" and "efficient mode" with the flip of a switch.
I'm not quite sure what we could do to add an efficient mode. In the common rail engines I think they adjust timing, etc. to improve economy, but that isn't something we can pull off. Now a drag and street mode....
(01-15-2010, 03:58 PM)300SD81 Issues I had:
CAN: Difficulty debugging without expensive scope
I2C: Errors with high speed and long cables
SPI: Limited number of nodes + speed/cable issues
Ethernet: Complex board routing
I would use CAN and only CAN. There isn't any way to debug it electrically with a scope by any reasonable means; however, using a pre-built CAN interface will remove the basic electrical problems. After that you just have to make sure your sending the properly formatted message. The first time it will be a *huge* PITA, but after that you'll be fine. It is designed to work in automotive environments, and there is a reason it was developed as you've found out!
I think I2C and SPI will definitely have problems with wire length.
John Robbins
'05 E320 CDI - 118k - Faaaaaast!!
'87 300TD - 317k - Cracked head... but an OM606 is on the way!
'79 300SD - 295k - Bad engine = project car!
I like the project and I hope you can get it going! I've been wanting to do this for years, and I've never been able to kick myself into actually working on it.
I would NOT use your standard RC servo for this. What happens if you lose 12V power? The rack position is now stuck wherever it was, and you have to figure out some way to turn off the engine without wrecking, over-revving, etc. Servo's are definitely easy to come by and control, and I think even I could mount one to the IP rack. However, for safety's sake please use a solenoid or something that has a failsafe. Maybe adding a plate or something to completely shut off airflow to the engine would work. Be sure to test it with the mechanical pump though I'm going to need something like this for my 606 electronic pump as well. At first it is going to be a 2x6" manually put in front of the turbo.
(01-14-2010, 03:10 PM)cell for most efficient operation, you'd want to set the turbo vane position to produce the minimum boost required to keep load below 100%.
Yes!!! A lot of people don't figure that out and think that more boost is always better. Light years down the road, you can also make an engine brake with a VNT.
Quote:...force-feedback system...
No offense, but I think that's pretty silly. By no means am I trying to discourage you, but I personally want to just mash the pedal to the floor and let the ECU figure out what the engine can do. If nothing else, one less thing to distract me during an evasive maneuver or something.
(01-15-2010, 01:54 PM)GREASY_BEAST Picture it this way: maxed out TC, but no return to idle issues, maxed out ALDA but no smoking, instantaneous high-rpm cut out at whichever rpm you choose, no gradual loss of power in the top end due to governor slowly cutting fuel.
Exactly!!!
(01-15-2010, 04:48 PM)GREASY_BEAST You can also have "sport mode" and "efficient mode" with the flip of a switch.
I'm not quite sure what we could do to add an efficient mode. In the common rail engines I think they adjust timing, etc. to improve economy, but that isn't something we can pull off. Now a drag and street mode....
(01-15-2010, 03:58 PM)300SD81 Issues I had:
CAN: Difficulty debugging without expensive scope
I2C: Errors with high speed and long cables
SPI: Limited number of nodes + speed/cable issues
Ethernet: Complex board routing
I would use CAN and only CAN. There isn't any way to debug it electrically with a scope by any reasonable means; however, using a pre-built CAN interface will remove the basic electrical problems. After that you just have to make sure your sending the properly formatted message. The first time it will be a *huge* PITA, but after that you'll be fine. It is designed to work in automotive environments, and there is a reason it was developed as you've found out!
I think I2C and SPI will definitely have problems with wire length.
John Robbins
'05 E320 CDI - 118k - Faaaaaast!!
'87 300TD - 317k - Cracked head... but an OM606 is on the way!
'79 300SD - 295k - Bad engine = project car!
01-15-2010, 05:42 PM
#11
Picture this: W123 or W126 + 617 with big elements and modern controls, VNT, etc. Heavy sound insulation. Seats and interior parts from a late-model Benz. Modern, refined, yet classic. I've seen similar things done to a MKI Rabbit, but the effect would be far more pronounced in these cars.
In any case I'll just shut up now because I don't have the time or money to put into this stuff at the moment (although I really want to). Good luck guys, can't wait to see what comes out of this!
This post was last modified: 01-15-2010, 05:50 PM by GREASY_BEAST.
GREASY_BEAST
01-15-2010, 05:42 PM #11I guess what I meant by "efficient" mode was more along the lines of "smoke limiting", and with a VNT "restriction limiting".. It would be nice to integrate all these electronics into the transmission as well.. I guess you could rig up an actuator to pull on the bowden cable and integrate the kickdown servo. That would truly "modernize" the old W123. Having a little more control over downshifts would improve the performance characteristics of these cars significantly.
Picture this: W123 or W126 + 617 with big elements and modern controls, VNT, etc. Heavy sound insulation. Seats and interior parts from a late-model Benz. Modern, refined, yet classic. I've seen similar things done to a MKI Rabbit, but the effect would be far more pronounced in these cars.
In any case I'll just shut up now because I don't have the time or money to put into this stuff at the moment (although I really want to). Good luck guys, can't wait to see what comes out of this!
01-15-2010, 05:44 PM
#12
(01-15-2010, 03:58 PM)300SD81 How fast do you plan on running your I2C? I had a couple boards running over I2C before switching to ethernet because I got errors at higher data rates/longer cables.
Issues I had:
CAN: Difficulty debugging without expensive scope
I2C: Errors with high speed and long cables
SPI: Limited number of nodes + speed/cable issues
Ethernet: Complex board routing
hey, that's great experience -- I'm sure your input on that will become vital as I get further along.
I don't have any set plans yet for the speed.
let's, see, a round-trip system speed of something like 100 control loops per second would probably be more than adequate. if that were three messages (master to TPS, master to IP, master to turbo), and each were a single byte (assuming 256 steps on each device is enough resolution), we'd need to initiate communication, send a byte, and terminate communication 3 times for each "round" of control. so I guess that's probably equivalent to a single pair of devices sending 300 single-byte messages per second.
I haven't looked at the i2c spec enough to know what sort of wire speed that translates into.
I suppose I can cobble together a test case to measure the current transfer speed. something like "toggle an LED every thousand rounds of communication" would probably give me a pretty good idea, and should only take a few minutes to code up.
I was thinking it would be neat to make a modified version of i2c which used balanced signals, so it would be four wires instead of two (i'm planning on stringing everything together with ethernet cables anyway, so I've got 8 wires to play with). I haven't put a lot of thought into it yet, but it would seem like it shouldn't be that hard to have an "instrumentation" op-amp at each end of the cables which converts from normal i2c to "balanced" i2c. I'll have to give that some more thought.
(01-15-2010, 04:48 PM)GREASY_BEAST You can also have "sport mode" and "efficient mode" with the flip of a switch. This would make the car very amenable to the addition of big IP elements, and would make the operation less expensive because the injection shop would not have to do any governor tuning, just balancing the elements to eachother.
I've had similar ideas, but with an axis labeled "economy" at one end and "emissions" at the other. I was thinking about labeling it "country" and "city"
This post was last modified: 01-15-2010, 05:48 PM by cell.
(01-15-2010, 03:58 PM)300SD81 How fast do you plan on running your I2C? I had a couple boards running over I2C before switching to ethernet because I got errors at higher data rates/longer cables.
Issues I had:
CAN: Difficulty debugging without expensive scope
I2C: Errors with high speed and long cables
SPI: Limited number of nodes + speed/cable issues
Ethernet: Complex board routing
hey, that's great experience -- I'm sure your input on that will become vital as I get further along.
I don't have any set plans yet for the speed.
let's, see, a round-trip system speed of something like 100 control loops per second would probably be more than adequate. if that were three messages (master to TPS, master to IP, master to turbo), and each were a single byte (assuming 256 steps on each device is enough resolution), we'd need to initiate communication, send a byte, and terminate communication 3 times for each "round" of control. so I guess that's probably equivalent to a single pair of devices sending 300 single-byte messages per second.
I haven't looked at the i2c spec enough to know what sort of wire speed that translates into.
I suppose I can cobble together a test case to measure the current transfer speed. something like "toggle an LED every thousand rounds of communication" would probably give me a pretty good idea, and should only take a few minutes to code up.
I was thinking it would be neat to make a modified version of i2c which used balanced signals, so it would be four wires instead of two (i'm planning on stringing everything together with ethernet cables anyway, so I've got 8 wires to play with). I haven't put a lot of thought into it yet, but it would seem like it shouldn't be that hard to have an "instrumentation" op-amp at each end of the cables which converts from normal i2c to "balanced" i2c. I'll have to give that some more thought.
(01-15-2010, 04:48 PM)GREASY_BEAST You can also have "sport mode" and "efficient mode" with the flip of a switch. This would make the car very amenable to the addition of big IP elements, and would make the operation less expensive because the injection shop would not have to do any governor tuning, just balancing the elements to eachother.
I've had similar ideas, but with an axis labeled "economy" at one end and "emissions" at the other. I was thinking about labeling it "country" and "city"
01-15-2010, 06:01 PM
#13
BTW load is easy to figure out and no one mentions it here.... O2 sensor!
Personally I would not want to do rack control but since I am sure have read the Hella thread then you know what I am looking for.
1987 300D Sturmmachine
1991 300D Nearly Perfect
1985 300D Weekend/Camping/Dog car
1974 L508D Motoroam Monarch "NightMare"
OBK #42
Man if you can tell me what I am doing wrong with the PWM control of the hell actuator I will get on my knees. Without bitbanging it.....
BTW load is easy to figure out and no one mentions it here.... O2 sensor!
Personally I would not want to do rack control but since I am sure have read the Hella thread then you know what I am looking for.
1987 300D Sturmmachine
1991 300D Nearly Perfect
1985 300D Weekend/Camping/Dog car
1974 L508D Motoroam Monarch "NightMare"
OBK #42
01-15-2010, 06:02 PM
#14
(01-15-2010, 05:15 PM)Tymbrymi I would NOT use your standard RC servo for this. What happens if you lose 12V power? The rack position is now stuck wherever it was, and you have to figure out some way to turn off the engine without wrecking, over-revving, etc. Servo's are definitely easy to come by and control, and I think even I could mount one to the IP rack. However, for safety's sake please use a solenoid or something that has a failsafe. Maybe adding a plate or something to completely shut off airflow to the engine would work. Be sure to test it with the mechanical pump though
yeah, i'd definitely like a manual override.
ideas:
* take apart the vacuum diaphram shutoff mechanism, and see if I can substitute it with a lever and pull-cord. run the pull-cord into the cab where the "choke" knob currently sits. this avoids interaction with the IP servo entirely (I hope? or do they interact inside of the IP such that you'd still have to overcome the friction of a dead servo?)
* the 2x6 isn't such a bad idea -- I was thinking about using the cheapest wide-mouth throttle body I could find on ebay, and either running a cable to it, or having it held open by a solenoid which would spring it shut if it loses power, and run the power cable through the cab to a "big red button" on the dash.
(01-15-2010, 05:15 PM)Tymbrymi Yes!!! A lot of people don't figure that out and think that more boost is always better.
you know, it occured to me this morning that the concept I was describing as "load" also happens to be the concept of "lambda", or fuel-air ratio, where 1.0 is stoichiometric. so basically I'd be attempting to maintain an lambda of 0.9 (or would that be 1.1?).
(01-15-2010, 05:15 PM)Tymbrymi No offense, but I think that's pretty silly. By no means am I trying to discourage you, but I personally want to just mash the pedal to the floor and let the ECU figure out what the engine can do. If nothing else, one less thing to distract me during an evasive maneuver or something.
hehe, that'll probably have to come later anyway
(01-15-2010, 05:15 PM)Tymbrymi engines I think they adjust timing, etc. to improve economy, but that isn't something we can pull off.
actually I was wondering about this. I haven't yet had the IP off the engine, so I don't know how exactly it is mounted. How crazy would it be to have the IP mounted on something like a hydraulic piston which would give you a little bit of real-time IP timing ability?
(01-15-2010, 05:15 PM)Tymbrymi I would use CAN and only CAN. There isn't any way to debug it electrically with a scope by any reasonable means; however, using a pre-built CAN interface will remove the basic electrical problems. After that you just have to make sure your sending the properly formatted message. The first time it will be a *huge* PITA, but after that you'll be fine. It is designed to work in automotive environments, and there is a reason it was developed as you've found out!
I think I2C and SPI will definitely have problems with wire length.
I'll have to read more about CAN. One thing I remember seeing is that there are atmel chips which have CAN capabilities as a hardware peripheral. So it might turn out to be pretty easy to make a CANduino variant.
There is already a guy who is working on making a variant of the arduino environment for some of the smaller chips (ie, the attiny series), so if we decide to go the CAN route we can probably leverage his work to see what needs to be changed.
here's atmel's selection of CAN compatible chips:
http://www.atmel.com/dyn/products/device...d=607#1609
looks like almost of dozen of them, some of them with more flash and ram than the chips in the arduino's.
I tried searching for some of those on octopart, but not all of them come up, and the prices are pretty variable, but they all appear to be within $5 to $15. so, more expensive than the regular chips, but not show-stoppingly so.
(01-15-2010, 05:42 PM)GREASY_BEAST It would be nice to integrate all these electronics into the transmission as well.. I guess you could rig up an actuator to pull on the bowden cable and integrate the kickdown servo. That would truly "modernize" the old W123. Having a little more control over downshifts would improve the performance characteristics of these cars significantly.
definitely!
(01-15-2010, 06:01 PM)winmutt BTW load is easy to figure out and no one mentions it here.... O2 sensor!
good idea!
man, that sucks you've been having trouble with the PWM. I remember reading that thread, especially the part about potentially bricking it with 12 volts? yikes, I probably would have learned that one the hard way.
as soon as I get my turbo arrives I'll get on it!
This post was last modified: 01-15-2010, 06:19 PM by cell.
(01-15-2010, 05:15 PM)Tymbrymi I would NOT use your standard RC servo for this. What happens if you lose 12V power? The rack position is now stuck wherever it was, and you have to figure out some way to turn off the engine without wrecking, over-revving, etc. Servo's are definitely easy to come by and control, and I think even I could mount one to the IP rack. However, for safety's sake please use a solenoid or something that has a failsafe. Maybe adding a plate or something to completely shut off airflow to the engine would work. Be sure to test it with the mechanical pump though
yeah, i'd definitely like a manual override.
ideas:
* take apart the vacuum diaphram shutoff mechanism, and see if I can substitute it with a lever and pull-cord. run the pull-cord into the cab where the "choke" knob currently sits. this avoids interaction with the IP servo entirely (I hope? or do they interact inside of the IP such that you'd still have to overcome the friction of a dead servo?)
* the 2x6 isn't such a bad idea -- I was thinking about using the cheapest wide-mouth throttle body I could find on ebay, and either running a cable to it, or having it held open by a solenoid which would spring it shut if it loses power, and run the power cable through the cab to a "big red button" on the dash.
(01-15-2010, 05:15 PM)Tymbrymi Yes!!! A lot of people don't figure that out and think that more boost is always better.
you know, it occured to me this morning that the concept I was describing as "load" also happens to be the concept of "lambda", or fuel-air ratio, where 1.0 is stoichiometric. so basically I'd be attempting to maintain an lambda of 0.9 (or would that be 1.1?).
(01-15-2010, 05:15 PM)Tymbrymi No offense, but I think that's pretty silly. By no means am I trying to discourage you, but I personally want to just mash the pedal to the floor and let the ECU figure out what the engine can do. If nothing else, one less thing to distract me during an evasive maneuver or something.
hehe, that'll probably have to come later anyway
(01-15-2010, 05:15 PM)Tymbrymi engines I think they adjust timing, etc. to improve economy, but that isn't something we can pull off.
actually I was wondering about this. I haven't yet had the IP off the engine, so I don't know how exactly it is mounted. How crazy would it be to have the IP mounted on something like a hydraulic piston which would give you a little bit of real-time IP timing ability?
(01-15-2010, 05:15 PM)Tymbrymi I would use CAN and only CAN. There isn't any way to debug it electrically with a scope by any reasonable means; however, using a pre-built CAN interface will remove the basic electrical problems. After that you just have to make sure your sending the properly formatted message. The first time it will be a *huge* PITA, but after that you'll be fine. It is designed to work in automotive environments, and there is a reason it was developed as you've found out!
I think I2C and SPI will definitely have problems with wire length.
I'll have to read more about CAN. One thing I remember seeing is that there are atmel chips which have CAN capabilities as a hardware peripheral. So it might turn out to be pretty easy to make a CANduino variant.
There is already a guy who is working on making a variant of the arduino environment for some of the smaller chips (ie, the attiny series), so if we decide to go the CAN route we can probably leverage his work to see what needs to be changed.
here's atmel's selection of CAN compatible chips:
http://www.atmel.com/dyn/products/device...d=607#1609
looks like almost of dozen of them, some of them with more flash and ram than the chips in the arduino's.
I tried searching for some of those on octopart, but not all of them come up, and the prices are pretty variable, but they all appear to be within $5 to $15. so, more expensive than the regular chips, but not show-stoppingly so.
(01-15-2010, 05:42 PM)GREASY_BEAST It would be nice to integrate all these electronics into the transmission as well.. I guess you could rig up an actuator to pull on the bowden cable and integrate the kickdown servo. That would truly "modernize" the old W123. Having a little more control over downshifts would improve the performance characteristics of these cars significantly.
definitely!
(01-15-2010, 06:01 PM)winmutt BTW load is easy to figure out and no one mentions it here.... O2 sensor!
good idea!
man, that sucks you've been having trouble with the PWM. I remember reading that thread, especially the part about potentially bricking it with 12 volts? yikes, I probably would have learned that one the hard way.
as soon as I get my turbo arrives I'll get on it!
01-15-2010, 06:20 PM
#15
(01-15-2010, 06:02 PM)cell I'll have to read more about CAN. One thing I remember seeing is that there are atmel chips which have CAN capabilities as a hardware peripheral. So it might turn out to be pretty easy to make a CANduino variant.
There is already a guy who is working on making a variant of the arduino environment for some of the smaller chips (ie, the attiny series), so if we decide to go the CAN route we can probably leverage his work to see what needs to be changed.
here's atmel's selection of CAN compatible chips:
http://www.atmel.com/dyn/products/device...d=607#1609
looks like almost of dozen of them, some of them with more flash and ram than the chips in the arduino's.
You can always go with a standalone SPI CAN controller as well.
Heres what I used in an early attempt at controlling the VNT...
http://www.microchip.com/wwwproducts/Dev...e=en010406
Ich liebe meine Autos!
1981 Mercedes-Benz 300SD | 156K Miles | 2nd Owner | EGR Disabled [Removal Pending] | ALDA Removed | Straight Pipes | GT2256V??? | Laser Interceptor | Engine swap over summer, hopefully with GT2256V attached...
1981 Mercedes-Benz 300SD | Odo Stopped at 160K (at least 50K more) | EGR Disabled | ALDA All The Way Out | Straight pipes | FM-870 Remote Start Alarm System | B100 Biodiesel | AC Fixed x2 | Trunk crushed in
| Retired to garage.Excessive speeding? It ain't excessive till I redline!
(01-15-2010, 06:02 PM)cell I'll have to read more about CAN. One thing I remember seeing is that there are atmel chips which have CAN capabilities as a hardware peripheral. So it might turn out to be pretty easy to make a CANduino variant.
There is already a guy who is working on making a variant of the arduino environment for some of the smaller chips (ie, the attiny series), so if we decide to go the CAN route we can probably leverage his work to see what needs to be changed.
here's atmel's selection of CAN compatible chips:
http://www.atmel.com/dyn/products/device...d=607#1609
looks like almost of dozen of them, some of them with more flash and ram than the chips in the arduino's.
You can always go with a standalone SPI CAN controller as well.
Heres what I used in an early attempt at controlling the VNT...
http://www.microchip.com/wwwproducts/Dev...e=en010406
Ich liebe meine Autos!
1981 Mercedes-Benz 300SD | 156K Miles | 2nd Owner | EGR Disabled [Removal Pending] | ALDA Removed | Straight Pipes | GT2256V??? | Laser Interceptor | Engine swap over summer, hopefully with GT2256V attached...
1981 Mercedes-Benz 300SD | Odo Stopped at 160K (at least 50K more) | EGR Disabled | ALDA All The Way Out | Straight pipes | FM-870 Remote Start Alarm System | B100 Biodiesel | AC Fixed x2 | Trunk crushed in
| Retired to garage.Excessive speeding? It ain't excessive till I redline!
01-15-2010, 08:36 PM
#16
You should add what Ford calls an IVS or idle validation switch. This switch is active when the throttle pedal is released to activate the idle speed regulating subroutine (governor).
You may also want to have a brake switch and a clutch switch (if so equipped) to develop a cruise control subroutine.
1998 MB E300 - 194,000 miles. Current/future mods: 7.5mm pump elements, Holset HX40 Turbo feeding Eaton M90 supercharger through custom fabbed intake system. Aeroturbine muffler, scratch-built 4" stainless exhaust. EGR eliminated, ECM recalibrated, modified rack position feedback circuit. Porterfield RS brake pads, Bilstein sport shocks, VDO EGT, oil and boost gauges.
2000 Ford F250 - 150,000 miles. Current/future mods: DP Tuner 3 map chip, scratch-built high flow intake, high output IDM, open circuit fuel system, ball bearing turbo, South Bend full metal clutch, slotted and drilled rotors. Autometer Sport Comp EGT and boost gauges.
I thought of a couple more things to toss out there:
You should add what Ford calls an IVS or idle validation switch. This switch is active when the throttle pedal is released to activate the idle speed regulating subroutine (governor).
You may also want to have a brake switch and a clutch switch (if so equipped) to develop a cruise control subroutine.
1998 MB E300 - 194,000 miles. Current/future mods: 7.5mm pump elements, Holset HX40 Turbo feeding Eaton M90 supercharger through custom fabbed intake system. Aeroturbine muffler, scratch-built 4" stainless exhaust. EGR eliminated, ECM recalibrated, modified rack position feedback circuit. Porterfield RS brake pads, Bilstein sport shocks, VDO EGT, oil and boost gauges.
2000 Ford F250 - 150,000 miles. Current/future mods: DP Tuner 3 map chip, scratch-built high flow intake, high output IDM, open circuit fuel system, ball bearing turbo, South Bend full metal clutch, slotted and drilled rotors. Autometer Sport Comp EGT and boost gauges.
01-15-2010, 08:39 PM
#17
(01-15-2010, 08:36 PM)E300TSC I thought of a couple more things to toss out there:
You should add what Ford calls an IVS or idle validation switch. This switch is active when the throttle pedal is released to activate the idle speed regulating subroutine (governor).
You may also want to have a brake switch and a clutch switch (if so equipped) to develop a cruise control subroutine.
good ideas. cruise control is definitely a must. not only for driving, but eventually I have plans to create a alternator-based welding power supply. something very similar to cruise control will be used to regulate the welding amperage.
(01-15-2010, 08:36 PM)E300TSC I thought of a couple more things to toss out there:
You should add what Ford calls an IVS or idle validation switch. This switch is active when the throttle pedal is released to activate the idle speed regulating subroutine (governor).
You may also want to have a brake switch and a clutch switch (if so equipped) to develop a cruise control subroutine.
good ideas. cruise control is definitely a must. not only for driving, but eventually I have plans to create a alternator-based welding power supply. something very similar to cruise control will be used to regulate the welding amperage.
01-15-2010, 09:29 PM
#18
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-15-2010, 09:29 PM #18On the late model Toyota's I worked on there was a back up throttle cable that did not inter fear in the flyby wire deal unless it failed it would actuate the throttle %20 at full throttle so if it failed you could still drive like a limp mode
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-15-2010, 10:53 PM
#19
video: http://www.youtube.com/watch?v=kKYJDkni5U4
code, etc: http://jason.pepas.com/wiki/index.php?ti...orAndGauge
the next step is to make it a three member exchange -- have the i2c master request data from an i2c slave sensor and then send the data to an i2c slave gauge.
the next step after that is to send the data to a servo instead of a gauge, and to replace the sensor with a pot (ie, a TPS).
got another i2c demo coded up. this one is an i2c master which generates a test pattern (a sawtooth) and sends it to an i2c slave gauge.
video: http://www.youtube.com/watch?v=kKYJDkni5U4
code, etc: http://jason.pepas.com/wiki/index.php?ti...orAndGauge
the next step is to make it a three member exchange -- have the i2c master request data from an i2c slave sensor and then send the data to an i2c slave gauge.
the next step after that is to send the data to a servo instead of a gauge, and to replace the sensor with a pot (ie, a TPS).
01-16-2010, 12:43 AM
#20
I've now have a three part i2c network -- sensor, orchestrator, and gauge.
video: http://www.youtube.com/watch?v=4gK2eJJHS-U
code, etc: http://jason.pepas.com/wiki/index.php?ti...sorNetwork
next step is to replace the gauge with a servo.
I'm on a roll! Man, I am really starting to like this arduino development environment. Its so quick to go from idea to reality.
I've now have a three part i2c network -- sensor, orchestrator, and gauge.
video: http://www.youtube.com/watch?v=4gK2eJJHS-U
code, etc: http://jason.pepas.com/wiki/index.php?ti...sorNetwork
next step is to replace the gauge with a servo.
01-16-2010, 01:05 AM
#21
(01-14-2010, 03:10 PM)cell I'll define 100% load as the point at which the amount of fuel being injected entirely consumes the available air (ie, past this point you make black smoke). so the load has to be calculated on the fly, based on rpm, boost pressure and temperature, and rack position.
for most efficient operation, you'd want to set the turbo vane position to produce the minimum boost required to keep load below 100%.
I suggest some more background reading on the A/F ratios that diesels are happy with.
Stoich (what you define at 100%) is about 14.5:1 and will melt your diesel in a big puff of black smoke.
Diesels always run lean, how rich you can get that lean-ness depends on how how your charge temps are. But you're going to want to keep it leaner than 20:1 under sustained load and probably 18:1 in transients.
You'll need an airflow meter to acheive this, calculate fuel volumes off rack position and probably generate some VE tables too. Because in transient operation you won't be able to calculate from the rapidly changing airflow, you'll just have to set max rack position based on current rpm
(01-15-2010, 01:41 PM)willbhere4u whats the advantage to this over stock linkages? I think Toyota just recalled all of there fly by wire car to be retrofired back to cable
No they didn't.
Toyota has been selling fly by wire diesels worldwide since about 1997. None of them have been recalled and retrofitted with cables.
My 2000 nissan diesel is fly by wire and every other VP37 or VP44 pumped car on up is.
For petrols it depends on the brand. There's a 99 subaru in the family that is fly by wire, my 98 petrol nissan still has a throttle cable though.
This post was last modified: 01-16-2010, 01:12 AM by Kiwibacon.
(01-14-2010, 03:10 PM)cell I'll define 100% load as the point at which the amount of fuel being injected entirely consumes the available air (ie, past this point you make black smoke). so the load has to be calculated on the fly, based on rpm, boost pressure and temperature, and rack position.
for most efficient operation, you'd want to set the turbo vane position to produce the minimum boost required to keep load below 100%.
I suggest some more background reading on the A/F ratios that diesels are happy with.
Stoich (what you define at 100%) is about 14.5:1 and will melt your diesel in a big puff of black smoke.
Diesels always run lean, how rich you can get that lean-ness depends on how how your charge temps are. But you're going to want to keep it leaner than 20:1 under sustained load and probably 18:1 in transients.
You'll need an airflow meter to acheive this, calculate fuel volumes off rack position and probably generate some VE tables too. Because in transient operation you won't be able to calculate from the rapidly changing airflow, you'll just have to set max rack position based on current rpm
(01-15-2010, 01:41 PM)willbhere4u whats the advantage to this over stock linkages? I think Toyota just recalled all of there fly by wire car to be retrofired back to cable
No they didn't.
Toyota has been selling fly by wire diesels worldwide since about 1997. None of them have been recalled and retrofitted with cables.
My 2000 nissan diesel is fly by wire and every other VP37 or VP44 pumped car on up is.
For petrols it depends on the brand. There's a 99 subaru in the family that is fly by wire, my 98 petrol nissan still has a throttle cable though.
01-16-2010, 08:07 PM
#22
BMW by far has the best throttle by wire system for the gas engine with out even having a throttle body just using engine management/valve timing to set its idle! Unlike Toyota using that stupid electric throttle body!!!
http://www.ennislaw.com/toyota_lexus_thr...02009.html
http://kansascity.injuryboard.com/automo...eid=275138
This post was last modified: 01-16-2010, 08:09 PM by willbhere4u.
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-16-2010, 08:07 PM #22I was talking about gasers not diesels Toyota hasen't sold diesels here in the US ever! And what Subaru in 99 has throttle by wire? I would highly doubt that! all of the Subarus I have sever worked on have throttle cable's Impreza Foresters and Legacys! here in the US toyota recalled all of there gas cars with throttle by wire they claim it's the floor mats but the fix is installing a throttle cable and disabling throttle by wire!!!
BMW by far has the best throttle by wire system for the gas engine with out even having a throttle body just using engine management/valve timing to set its idle! Unlike Toyota using that stupid electric throttle body!!!
http://www.ennislaw.com/toyota_lexus_thr...02009.html
http://kansascity.injuryboard.com/automo...eid=275138
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-17-2010, 12:49 AM
#23
(01-16-2010, 08:07 PM)willbhere4u I was talking about gasers not diesels Toyota hasen't sold diesels here in the US ever! And what Subaru in 99 has throttle by wire? I would highly doubt that! all of the Subarus I have sever worked on have throttle cable's Impreza Foresters and Legacys! here in the US toyota recalled all of there gas cars with throttle by wire they claim it's the floor mats but the fix is installing a throttle cable and disabling throttle by wire!!!
I was talking petrols and diesels. Which subaru has fly by wire? Well the specific one I mentioned is an RSK-B4. Twin turbocharged legacy. Just like this one:
http://www.carandclassic.co.uk/car/C127786/
Since 05 every subaru sold here has been fly-by-wire. Not sure exactly when they started.
http://www.subaru.co.nz/Technical_Lab/News/
You can doubt all you like, it's been done for over a decade now.
The 100 series landcruiser used it in both petrol and diesel. It was launched in 99, here's a press release from 2002
http://www.pressroom.com.au/press_kit_de...tID=33#737
The link you provided claims toyota was using fly by wire in 2002.
Quote:Toyota first installed electronic throttles in 2002 model year Lexus ES and Camry sedans.
This post was last modified: 01-17-2010, 12:55 AM by Kiwibacon.
(01-16-2010, 08:07 PM)willbhere4u I was talking about gasers not diesels Toyota hasen't sold diesels here in the US ever! And what Subaru in 99 has throttle by wire? I would highly doubt that! all of the Subarus I have sever worked on have throttle cable's Impreza Foresters and Legacys! here in the US toyota recalled all of there gas cars with throttle by wire they claim it's the floor mats but the fix is installing a throttle cable and disabling throttle by wire!!!
I was talking petrols and diesels. Which subaru has fly by wire? Well the specific one I mentioned is an RSK-B4. Twin turbocharged legacy. Just like this one:
http://www.carandclassic.co.uk/car/C127786/
Since 05 every subaru sold here has been fly-by-wire. Not sure exactly when they started.
http://www.subaru.co.nz/Technical_Lab/News/
You can doubt all you like, it's been done for over a decade now.
The 100 series landcruiser used it in both petrol and diesel. It was launched in 99, here's a press release from 2002
http://www.pressroom.com.au/press_kit_de...tID=33#737
The link you provided claims toyota was using fly by wire in 2002.
Quote:Toyota first installed electronic throttles in 2002 model year Lexus ES and Camry sedans.
01-17-2010, 02:43 PM
#24
But remember mechanical throttle linkage has been used for 125 years now on automobiles and works pretty darn well on a gas or diesel engine!
like I said I like BMW's system seams to work the best on a petrol engine because the engine was designed for it to work! with no throttle body at all like a diesel!
It makes a lot more seance on a diesel than on a petrol engine in IMO
here in the us we didn't get turbo legacy's either until 05! and all Subaru's in the US are single turbo with a throttle cable even to this day! we don't get the cool ones!!! I'm a big Subaru fan myself and would kill to put one of those engines in my 96 legacy outback wagon ej22 5spd
and I like knowing my 240d turbo 4spd will be the only car running if we ever get hit by an EMP blast as its completely mechanical in every way! all I need is a long enough hill and a warm day!
sorry for hijacking you thread and keep up the good work!
This post was last modified: 01-17-2010, 02:48 PM by willbhere4u.
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-17-2010, 02:43 PM #24I'm not doubting they used it at all!!! I just pointed out it didn't work well! I have heard from several Porsche and Toyota owners it doesn't have the same throttle responce as the cable operated ones and that there is a delay between when you push the throttle and when it accelerates we have had throttle by wire for over a decade over here as well
But remember mechanical throttle linkage has been used for 125 years now on automobiles and works pretty darn well on a gas or diesel engine!
like I said I like BMW's system seams to work the best on a petrol engine because the engine was designed for it to work! with no throttle body at all like a diesel!
It makes a lot more seance on a diesel than on a petrol engine in IMO
here in the us we didn't get turbo legacy's either until 05! and all Subaru's in the US are single turbo with a throttle cable even to this day! we don't get the cool ones!!! I'm a big Subaru fan myself and would kill to put one of those engines in my 96 legacy outback wagon ej22 5spd
and I like knowing my 240d turbo 4spd will be the only car running if we ever get hit by an EMP blast as its completely mechanical in every way! all I need is a long enough hill and a warm day!
sorry for hijacking you thread and keep up the good work!
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-17-2010, 03:19 PM
#25
(01-17-2010, 02:43 PM)willbhere4u I'm not doubting they used it at all!!! I just pointed out it didn't work well! I have heard from several Porsche and Toyota owners it doesn't have the same throttle responce as the cable operated ones and that there is a delay between when you push the throttle and when it accelerates we have had throttle by wire for over a decade over here as well
I think the "it feels laggy" call is BS. In a blind test they couldn't tell you which was which. The transmission programming has by far the biggest impact on acceleration.
I've pretty sure the 996 porsche I've driven was fly-by-wire and I know the half a dozen toyotas, that subaru, my work car, several audi's and other cars I've driven are fly-by-wire.
Only one of them has an awful pedal feel and that's a 99 landcruiser. Lag isn't the problem there, the pedal just needs more spring behind it.
(01-17-2010, 02:43 PM)willbhere4u I'm not doubting they used it at all!!! I just pointed out it didn't work well! I have heard from several Porsche and Toyota owners it doesn't have the same throttle responce as the cable operated ones and that there is a delay between when you push the throttle and when it accelerates we have had throttle by wire for over a decade over here as well
I think the "it feels laggy" call is BS. In a blind test they couldn't tell you which was which. The transmission programming has by far the biggest impact on acceleration.
I've pretty sure the 996 porsche I've driven was fly-by-wire and I know the half a dozen toyotas, that subaru, my work car, several audi's and other cars I've driven are fly-by-wire.
Only one of them has an awful pedal feel and that's a 99 landcruiser. Lag isn't the problem there, the pedal just needs more spring behind it.
01-17-2010, 06:16 PM
#26
(01-16-2010, 01:05 AM)Kiwibacon I suggest some more background reading on the A/F ratios that diesels are happy with.
Stoich (what you define at 100%) is about 14.5:1 and will melt your diesel in a big puff of black smoke.
Diesels always run lean, how rich you can get that lean-ness depends on how how your charge temps are. But you're going to want to keep it leaner than 20:1 under sustained load and probably 18:1 in transients.
thanks, I'll have to do some more reading.
(01-16-2010, 01:05 AM)Kiwibacon I suggest some more background reading on the A/F ratios that diesels are happy with.
Stoich (what you define at 100%) is about 14.5:1 and will melt your diesel in a big puff of black smoke.
Diesels always run lean, how rich you can get that lean-ness depends on how how your charge temps are. But you're going to want to keep it leaner than 20:1 under sustained load and probably 18:1 in transients.
thanks, I'll have to do some more reading.
01-17-2010, 07:21 PM
#27
Evan Fuel injection can't snap throttle response like a carburetor I had an mg with a 1275cc and duel SU carbs and it would snap to redline with a blip of the throttle so quick you couldn't stop it from smacking the limiter and back to idle so fast you wouldn't believe it!
Even my 94 Miata with a supercharger if you snap off the throttle will only drop to 2000k rpm and then slowly lower to idle
In a gas car there is no power to be gained from throttle by wire and no use for it! unless you remove the throttle body from the equation
Why not just eliminate the driver from the equation and let a computer do all of the driving!!!
This post was last modified: 01-18-2010, 10:06 PM by willbhere4u.
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-17-2010, 07:21 PM #27An electric motor cant snap a throttle open and closed like a cable and spring!!! Floor your car and snap off the throttle it will pull for a second before it comes back to normal that's the majority of the complaints I hear day in and day out about those systems!
Evan Fuel injection can't snap throttle response like a carburetor I had an mg with a 1275cc and duel SU carbs and it would snap to redline with a blip of the throttle so quick you couldn't stop it from smacking the limiter and back to idle so fast you wouldn't believe it!
Even my 94 Miata with a supercharger if you snap off the throttle will only drop to 2000k rpm and then slowly lower to idle
In a gas car there is no power to be gained from throttle by wire and no use for it! unless you remove the throttle body from the equation
Why not just eliminate the driver from the equation and let a computer do all of the driving!!!
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-17-2010, 07:42 PM
#28
(01-17-2010, 07:21 PM)willbhere4u An electric motor cant snap a throttle open and closed like a cable and spring!!!
Oh yes it can.
I've worked in automation, we can make actuators move faster than the eye can see.
I've worked with robots which can move far faster than people can.
There's no 1 second lagin the cars I've driven.
(01-17-2010, 07:21 PM)willbhere4u An electric motor cant snap a throttle open and closed like a cable and spring!!!
Oh yes it can.
I've worked in automation, we can make actuators move faster than the eye can see.
I've worked with robots which can move far faster than people can.
There's no 1 second lagin the cars I've driven.
01-17-2010, 08:22 PM
#29
we used electric actuators for shifting on gsxr 1000 and huyabusa powered scca f 1000 cars they are quick but that isnt the techoligey being used by manufactures!
again what is the advantage? It dosent help hp? If you think of the power it takes to do all of this it loads the alternator and looses hp maybe not much!
but there is still no advantage just because you can doesn't mean you should Its just taking something simple and complicating it why not make it hydraulic???
This post was last modified: 01-17-2010, 08:23 PM by willbhere4u.
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-17-2010, 08:22 PM #29Yea and the valves in an F1 engine are driven pneumatically what difference dose that make when the throttle body are drive with an electric motor and gears? actuators use one plunger in an electric field! gears can only turn so fast! and pneumatic actuators are even faster!
we used electric actuators for shifting on gsxr 1000 and huyabusa powered scca f 1000 cars they are quick but that isnt the techoligey being used by manufactures!
again what is the advantage? It dosent help hp? If you think of the power it takes to do all of this it loads the alternator and looses hp maybe not much!
but there is still no advantage just because you can doesn't mean you should Its just taking something simple and complicating it why not make it hydraulic???
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-17-2010, 08:26 PM
#30
(01-17-2010, 07:21 PM)willbhere4u no use for it!
One exceptionally important "use" is that you can design failures to be safe. That is, if any component fails the engine will default to idle. With a cable you have the possibility, however remote, of binding. Without, there is no such possibility.
I agree that "fly-by-wire" control of a carb is probably does not lend itself to utility in the cost-benefit sense, but it would give you much more precise control over the throttle body. This combined with the actuator speed that kiwi was referring to means that the governor responds faster, much like in the case with a diesel IP, so there could be marginal gains in top-end.
Another real benefit is that computer control of the throttle gives, coupled with some sensors in the trans, the ability to auto-match engine revs for shifts allowing you to shift faster and more efficiently. If you have a really big servo actuating the shift lever and the clutch you could turn a manual transmission into an automatic that has shift times which are faster than you can blink. This provides an ENORMOUS benefit in acceleration and driveability, not to mention the hardware is protected from human error and always operates within its limitations.
Quote:Why not just eliminate the driver from the equation and let a computer do all of the driving!!!
Ludd-ism holds very little sway in a forum dedicated to technology.
GREASY_BEAST
01-17-2010, 08:26 PM #30(01-17-2010, 07:21 PM)willbhere4u no use for it!
One exceptionally important "use" is that you can design failures to be safe. That is, if any component fails the engine will default to idle. With a cable you have the possibility, however remote, of binding. Without, there is no such possibility.
I agree that "fly-by-wire" control of a carb is probably does not lend itself to utility in the cost-benefit sense, but it would give you much more precise control over the throttle body. This combined with the actuator speed that kiwi was referring to means that the governor responds faster, much like in the case with a diesel IP, so there could be marginal gains in top-end.
Another real benefit is that computer control of the throttle gives, coupled with some sensors in the trans, the ability to auto-match engine revs for shifts allowing you to shift faster and more efficiently. If you have a really big servo actuating the shift lever and the clutch you could turn a manual transmission into an automatic that has shift times which are faster than you can blink. This provides an ENORMOUS benefit in acceleration and driveability, not to mention the hardware is protected from human error and always operates within its limitations.
Quote:Why not just eliminate the driver from the equation and let a computer do all of the driving!!!
Ludd-ism holds very little sway in a forum dedicated to technology.
01-17-2010, 08:39 PM
#31
(01-17-2010, 08:22 PM)willbhere4u Yea and the valves in an F1 engine are driven pneumatically what difference dose that make when the throttle body are drive with an electric motor and gears? actuators use one plunger in an electric field! gears can only turn so fast! and pneumatic actuators are even faster!
we used electric actuators for shifting on gsxr 1000 and huyabusa powered scca f 1000 cars they are quick but that isnt the techoligey being used by manufactures!
again what is the advantage? It dosent help hp? If you think of the power it takes to do all of this it loads the alternator and looses hp maybe not much!
but there is still no advantage just because you can doesn't mean you should Its just taking something simple and complicating it why not make it hydraulic???
Maybe you've never seeen how fast an electric motor and gears can go. That robot I mentioned earlier, that's exactly what they run. Steppers or servo motors (depends on brand and size) with reduction gears.
Why?
Many reasons.
It gives the computer full throttle control for starting, cruise control, limp-mode, softening gear changes etc.
It allows the throttle pedal to be programmed to feel how the driver expects it to feel. Landrover do this to provide a short travel pedal on-road for the expected feel but providing a long travel pedal offroad to minimise throttle errors due to bumpy terrain.
Packaging, gets rid of a cable with a dictated path and minimum bend radius.
NVH, one less connection between the car and chassis
Safety, no frayed cables jamming on (yes I've had this, but being a diesel it simply meant a 1500rpm idle, not half throttle to redline like a petrol).
Feel, less friction in a pedal than a metre or so of cable.
Performance, faster reaction in both directions than a cable and return spring is capable of.
Lower maintenance, the upper and lower limits of motion aren't dependent on cable stretch, wear and how good the guy who adjusted it was.
Cables are dead. It's interesting that the US markets get the new technology last.
(01-17-2010, 08:26 PM)GREASY_BEAST Another real benefit is that computer control of the throttle gives, coupled with some sensors in the trans, the ability to auto-match engine revs for shifts allowing you to shift faster and more efficiently. If you have a really big servo actuating the shift lever and the clutch you could turn a manual transmission into an automatic that has shift times which are faster than you can blink. This provides an ENORMOUS benefit in acceleration and driveability, not to mention the hardware is protected from human error and always operates within its limitations.
An ex-bus mechanic tells me such transmissions on buses are quite common now. They have 5 times the clutch life of an equivalent manual transmission in city service.
This post was last modified: 01-17-2010, 08:43 PM by Kiwibacon.
(01-17-2010, 08:22 PM)willbhere4u Yea and the valves in an F1 engine are driven pneumatically what difference dose that make when the throttle body are drive with an electric motor and gears? actuators use one plunger in an electric field! gears can only turn so fast! and pneumatic actuators are even faster!
we used electric actuators for shifting on gsxr 1000 and huyabusa powered scca f 1000 cars they are quick but that isnt the techoligey being used by manufactures!
again what is the advantage? It dosent help hp? If you think of the power it takes to do all of this it loads the alternator and looses hp maybe not much!
but there is still no advantage just because you can doesn't mean you should Its just taking something simple and complicating it why not make it hydraulic???
Maybe you've never seeen how fast an electric motor and gears can go. That robot I mentioned earlier, that's exactly what they run. Steppers or servo motors (depends on brand and size) with reduction gears.
Why?
Many reasons.
It gives the computer full throttle control for starting, cruise control, limp-mode, softening gear changes etc.
It allows the throttle pedal to be programmed to feel how the driver expects it to feel. Landrover do this to provide a short travel pedal on-road for the expected feel but providing a long travel pedal offroad to minimise throttle errors due to bumpy terrain.
Packaging, gets rid of a cable with a dictated path and minimum bend radius.
NVH, one less connection between the car and chassis
Safety, no frayed cables jamming on (yes I've had this, but being a diesel it simply meant a 1500rpm idle, not half throttle to redline like a petrol).
Feel, less friction in a pedal than a metre or so of cable.
Performance, faster reaction in both directions than a cable and return spring is capable of.
Lower maintenance, the upper and lower limits of motion aren't dependent on cable stretch, wear and how good the guy who adjusted it was.
Cables are dead. It's interesting that the US markets get the new technology last.
(01-17-2010, 08:26 PM)GREASY_BEAST Another real benefit is that computer control of the throttle gives, coupled with some sensors in the trans, the ability to auto-match engine revs for shifts allowing you to shift faster and more efficiently. If you have a really big servo actuating the shift lever and the clutch you could turn a manual transmission into an automatic that has shift times which are faster than you can blink. This provides an ENORMOUS benefit in acceleration and driveability, not to mention the hardware is protected from human error and always operates within its limitations.
An ex-bus mechanic tells me such transmissions on buses are quite common now. They have 5 times the clutch life of an equivalent manual transmission in city service.
01-17-2010, 09:43 PM
#32
Of course, my experience on this is limited solely to working on a fleet of workhorse P vans, (very shitty gathering of random GM parts, spackle, left over aircraft aluminum, duct tape, and a childs tears)
We went through a round where we had to replace all the GM drive by wire systems where we could because they all started to malfunction in daily service with the slightest corrosion. It got to the point where trucks A-F were designated "fair weather platforms"
There was a rash of shorting connections causing the trucks to A, shut off without warning, B enter limp mode without warning, and C, display general psychotic behavior without input from the driver, most of the time on the highway with high water spray.
these were all 05 and 06 trucks I believe, using somewhat earlier GM technology. I found it incredibly unreliable after only a couple of years, BUT, again, a workhorse van is not well designed, and the system was in constant contact with weather and road salt/sand. Should work better with a little consideration.
Just to chime in on this, I appreciate the idea of drive by wire, but there is a lot to be said to an honest mechanical connection to the throttle for reliability, especially if your deal with salt and weather.
Of course, my experience on this is limited solely to working on a fleet of workhorse P vans, (very shitty gathering of random GM parts, spackle, left over aircraft aluminum, duct tape, and a childs tears)
We went through a round where we had to replace all the GM drive by wire systems where we could because they all started to malfunction in daily service with the slightest corrosion. It got to the point where trucks A-F were designated "fair weather platforms"
There was a rash of shorting connections causing the trucks to A, shut off without warning, B enter limp mode without warning, and C, display general psychotic behavior without input from the driver, most of the time on the highway with high water spray.
these were all 05 and 06 trucks I believe, using somewhat earlier GM technology. I found it incredibly unreliable after only a couple of years, BUT, again, a workhorse van is not well designed, and the system was in constant contact with weather and road salt/sand. Should work better with a little consideration.
01-18-2010, 12:20 AM
#33
And a manual transmissions don't need any computer input it's just a box full of gears and levers???
There is no performance benefit unless you remove the throttle body all together like BMW did
Keep it simple stupid!!! The more complicated it is the more there is to break or fail
I should know I owned a Citroen DS hydromatic with a hydraulic computer to control the clutch and gear changes using break and throttle inputs from the carburetor and an electric shifter it's nothing new its all been around 60 years mechanical/hydraulic linkages can do any thing a computer can!
personally I prefer the price control over my throttle that a cable/linkage gives me!
I would hate the pause between when you tell it to do something and when the computer decides to do it or not! And there is that not so remote possibility that the throttle body will fail/bind at wot or die just when you need to out accelerate that bus that is out of control because of it's failed throttle by wire system
P.S. my mercedes 240d dose not have a cable it has a direct linkage from the pedal to the injector pump! works great even 30 years later!
and my 1928 Buick throttle linkage still works great with no maintenance 82 years later! for that mater our ten year old 2000 Subaru Forester throttle CABLE still works great even with 235,000 miles and no adjustments still his full throttle
This post was last modified: 01-18-2010, 12:44 AM by willbhere4u.
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-18-2010, 12:20 AM #33How is throttle by wire safer? Maybe for the engine but not for the driver! What if the electronics short or fail at WOT like they have been doing! This is why a bunch of the auto manufactures have recalled and retrofitted these systems!!!
And a manual transmissions don't need any computer input it's just a box full of gears and levers???
There is no performance benefit unless you remove the throttle body all together like BMW did
Keep it simple stupid!!! The more complicated it is the more there is to break or fail
I should know I owned a Citroen DS hydromatic with a hydraulic computer to control the clutch and gear changes using break and throttle inputs from the carburetor and an electric shifter it's nothing new its all been around 60 years mechanical/hydraulic linkages can do any thing a computer can!
personally I prefer the price control over my throttle that a cable/linkage gives me!
I would hate the pause between when you tell it to do something and when the computer decides to do it or not! And there is that not so remote possibility that the throttle body will fail/bind at wot or die just when you need to out accelerate that bus that is out of control because of it's failed throttle by wire system
P.S. my mercedes 240d dose not have a cable it has a direct linkage from the pedal to the injector pump! works great even 30 years later!
and my 1928 Buick throttle linkage still works great with no maintenance 82 years later! for that mater our ten year old 2000 Subaru Forester throttle CABLE still works great even with 235,000 miles and no adjustments still his full throttle
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-18-2010, 12:37 AM
#34
(01-18-2010, 12:20 AM)willbhere4u How is throttle by wire safer? Maybe for the engine but not for the driver! What if the electronics short or fail at WOT like they have been doing! This is why a bunch of the auto manufactures have recalled and retrofitted these systems!!!
"Fail-safe" means that something fails in a mode which is safe. Take out a napkin and a pencil and draw an electronic-actuator system that fails safe. Now draw a mechanical one that fails safe. You'll notice that its a lot easier to draw the electronic one. I have no idea if/why there have been recalls, however, I know of plenty of VW TDi's that have gone for 300K+ without incident...
Quote:And a manual transmissions don't need any computer input it's just a box full of gears and levers???
Did you not read? Of course a manual transmission doesn't need electronics, but electronic controls make it far more reliable and a LOT more economical, not to mention more enjoyable to drive. A transmission is a hell of a lot more than just "a box full of gears" btw... On your napkin draw a picture of a transmission, and somewhere write down the loads on each part for a given input torque and rotational speed.
Quote:There is no performance benefit unless you remove the throttle body all together like BMW did
I have no idea what you are talking about. There have been plenty of performance benefits listed earlier. Smoother governing, for one, more efficient shifting characteristics (even if you don't touch the tranny), more efficient idle, etc.
Quote:Keep it simple stupid!!! The more complicated it is the more there is to break or fail
Now we are talking religion. Electronics are simple. That's the whole point. The reason they exist is to simplify things that would otherwise be devastatingly complex. Having this simplified complexity at ones fingertips allows vast performance gains to be made in all areas.
Quote:I should know I owned a Citroen DS hydromatic with a hydraulic computer to control the clutch and gear changes using break and throttle inputs from the carburetor and an electric shifter it's nothing new its all been around 60 years mechanical/hydraulic linkages can do any thing a computer can!
Not a chance. Draw the pictures! How the heck are you going to modify the RW governor to approach the abilities of a simple algorithm controlling off-the-shelf electronic parts? In short, you can't. That's why all these new-fangled eelectricity sparkin' doo-dads exist.
This thread was about hooking up electrics to the MW. Discussion of 19th century religious extremism should be kept to the open section. Mods please delete liberally here!!!
GREASY_BEAST
01-18-2010, 12:37 AM #34(01-18-2010, 12:20 AM)willbhere4u How is throttle by wire safer? Maybe for the engine but not for the driver! What if the electronics short or fail at WOT like they have been doing! This is why a bunch of the auto manufactures have recalled and retrofitted these systems!!!
"Fail-safe" means that something fails in a mode which is safe. Take out a napkin and a pencil and draw an electronic-actuator system that fails safe. Now draw a mechanical one that fails safe. You'll notice that its a lot easier to draw the electronic one. I have no idea if/why there have been recalls, however, I know of plenty of VW TDi's that have gone for 300K+ without incident...
Quote:And a manual transmissions don't need any computer input it's just a box full of gears and levers???
Did you not read? Of course a manual transmission doesn't need electronics, but electronic controls make it far more reliable and a LOT more economical, not to mention more enjoyable to drive. A transmission is a hell of a lot more than just "a box full of gears" btw... On your napkin draw a picture of a transmission, and somewhere write down the loads on each part for a given input torque and rotational speed.
Quote:There is no performance benefit unless you remove the throttle body all together like BMW did
I have no idea what you are talking about. There have been plenty of performance benefits listed earlier. Smoother governing, for one, more efficient shifting characteristics (even if you don't touch the tranny), more efficient idle, etc.
Quote:Keep it simple stupid!!! The more complicated it is the more there is to break or fail
Now we are talking religion. Electronics are simple. That's the whole point. The reason they exist is to simplify things that would otherwise be devastatingly complex. Having this simplified complexity at ones fingertips allows vast performance gains to be made in all areas.
Quote:I should know I owned a Citroen DS hydromatic with a hydraulic computer to control the clutch and gear changes using break and throttle inputs from the carburetor and an electric shifter it's nothing new its all been around 60 years mechanical/hydraulic linkages can do any thing a computer can!
Not a chance. Draw the pictures! How the heck are you going to modify the RW governor to approach the abilities of a simple algorithm controlling off-the-shelf electronic parts? In short, you can't. That's why all these new-fangled eelectricity sparkin' doo-dads exist.
This thread was about hooking up electrics to the MW. Discussion of 19th century religious extremism should be kept to the open section. Mods please delete liberally here!!!
01-18-2010, 12:53 AM
#35
(01-18-2010, 12:20 AM)willbhere4u How is throttle by wire safer? Maybe for the engine but not for the driver! What if the electronics short or fail at WOT like they have been doing! This is why a bunch of the auto manufactures have recalled and retrofitted these systems!!!
So you keep saying. But I've not found a single reference other than yours. Unintended acceleration is very rare, but the solution is pretty obvious. Put it in neutral.
Mechanical cables and springs fail too, chances of them failing at WOT are quite good. All you need is a broken return spring.
(01-18-2010, 12:20 AM)willbhere4u And a manual transmissions don't need any computer input it's just a box full of gears and levers???
How many vehicles are sold new these days with a manual transmission?
(01-18-2010, 12:20 AM)willbhere4u How is throttle by wire safer? Maybe for the engine but not for the driver! What if the electronics short or fail at WOT like they have been doing! This is why a bunch of the auto manufactures have recalled and retrofitted these systems!!!
So you keep saying. But I've not found a single reference other than yours. Unintended acceleration is very rare, but the solution is pretty obvious. Put it in neutral.
Mechanical cables and springs fail too, chances of them failing at WOT are quite good. All you need is a broken return spring.
(01-18-2010, 12:20 AM)willbhere4u And a manual transmissions don't need any computer input it's just a box full of gears and levers???
How many vehicles are sold new these days with a manual transmission?
01-18-2010, 02:16 AM
#36
(01-17-2010, 07:21 PM)willbhere4u Evan Fuel injection can't snap throttle response like a carburetorFuel injection has big advantages over a carburetor and most g@s engines can be easily converted to it. Deceleration fuel shutoff, accurate and even fuel distribution, real time A/F ratio control, RPM limiting without washing the cylinders, no fuel condensation, no back firing, anti-lag programs, etc.
Thats the big difference. A mechanical Diesel already works very well, adding an electric motor to control the mechanical throttle and governor is a huge evolutionary step backwards. If you want an electronic Diesel, you should get one that started that way. Either with an electronic IP, common rail or unit injectors.
(01-18-2010, 12:53 AM)Kiwibacon How many vehicles are sold new these days with a manual transmission?
Quite a lot, but most are so computerized they are fully automatic (as in DSG).
This post was last modified: 01-18-2010, 02:21 AM by ForcedInduction.
ForcedInduction
01-18-2010, 02:16 AM
#36
(01-17-2010, 07:21 PM)willbhere4u Evan Fuel injection can't snap throttle response like a carburetorFuel injection has big advantages over a carburetor and most g@s engines can be easily converted to it. Deceleration fuel shutoff, accurate and even fuel distribution, real time A/F ratio control, RPM limiting without washing the cylinders, no fuel condensation, no back firing, anti-lag programs, etc.
Thats the big difference. A mechanical Diesel already works very well, adding an electric motor to control the mechanical throttle and governor is a huge evolutionary step backwards. If you want an electronic Diesel, you should get one that started that way. Either with an electronic IP, common rail or unit injectors.
(01-18-2010, 12:53 AM)Kiwibacon How many vehicles are sold new these days with a manual transmission?
Quite a lot, but most are so computerized they are fully automatic (as in DSG).
01-18-2010, 02:25 AM
#37
here's my progress for the evening:
![[Image: I2cthrottle.jpg]](http://jason.pepas.com/wiki/images/3/34/I2cthrottle.jpg)
video: http://www.youtube.com/watch?v=tciLgdMQsOE
code, etc: http://jason.pepas.com/wiki/index.php?ti...ePrototype
This post was last modified: 01-18-2010, 02:26 AM by cell.
boy this thread is a hot one
here's my progress for the evening:
video: http://www.youtube.com/watch?v=tciLgdMQsOE
code, etc: http://jason.pepas.com/wiki/index.php?ti...ePrototype
01-18-2010, 05:05 AM
#38
(01-18-2010, 02:16 AM)ForcedInduction Thats the big difference. A mechanical Diesel already works very well, adding an electric motor to control the mechanical throttle and governor is a huge evolutionary step backwards. If you want an electronic Diesel, you should get one that started that way. Either with an electronic IP, common rail or unit injectors.
I'd have to agree. While I won't discourage you from experimenting it certainly won't be anything I'll look to duplicate. Fly by wire has to be one of the more useless 'innovations' I've seen in the automotive marketplace. One of the big complaints about BMW's is that you get the normal mode and it never gives you enough throttle opening, and then you engage sport mode and it opens too fast. A linear 1:1 ratio is best but then you're using servos, motors, etc to duplicate the function of a cable, which is silly. Throttle return springs do break, sure, sometimes.
Closest thing I've seen to that in real life was a body shop customer with a 94 Vette, as a BS move to sway the judge the guy's lawyer started a dummy lawsuit against GM for their 'broken throttle return spring' that 'caused' the guy to do a few miles flat out on the LIE. It worked, and the guy has a real car guy relic - a speeding ticket for 165 in a 55 that he got out of! I've only gotten an 85 in a 35 and I'm still dealing with that! Almost triple, but not quite
. The only reason for FBW, in my opinion, is for electronic nannies, and I'm a big boy, thanks. I think we need better drivers instead of 'safer' cars, myself.
If you're this comfy with automotive electronics I'd love to see you tinker with a TDI or other modern diesel. Say you accomplished all you wanted with this on a 617, you're still stuck with a two valve IDI head, which besides the injection pump is the biggest hurdle to making power on this engine, and unless you're a machining wizard you're pretty much stuck with that. You'd end up with a fly by wire'd mechanical IDI diesel, which would be like using a late model OBDII ECU to control a Holley four barrel. Worst of both worlds.
Like I said though, don't let me discourage you, but I bought my 240D for a reason (it was cheap!) but I'd by another in a heartbeat for it's simple nature. While electronics can be made to be as or more reliable as a mechanical connection the manufacturers have little interest in that. You do, though. I'd like to take some late model EFI motor and strip it down to basics, get rid of the factory cockblocking (mostly software) and basically build the system I'd want. I think the software is just as important, the thing that makes the later cars have problems is the computer getting 'confused' and doing weird stuff.
Speaking of which, be careful testing that in the real world. That has been a problem with some factory systems, lurching out into traffic for no apparent reason and things like that. Factories get sued for that kind of thing, too, so I'm assuming they're going to be more rigorous in their design than you might. Just don't want everyone going 'where's cell? We never did find out what happened to that FBW setup he was working on!'
CID Vicious
01-18-2010, 05:05 AM #38(01-18-2010, 02:16 AM)ForcedInduction Thats the big difference. A mechanical Diesel already works very well, adding an electric motor to control the mechanical throttle and governor is a huge evolutionary step backwards. If you want an electronic Diesel, you should get one that started that way. Either with an electronic IP, common rail or unit injectors.
I'd have to agree. While I won't discourage you from experimenting it certainly won't be anything I'll look to duplicate. Fly by wire has to be one of the more useless 'innovations' I've seen in the automotive marketplace. One of the big complaints about BMW's is that you get the normal mode and it never gives you enough throttle opening, and then you engage sport mode and it opens too fast. A linear 1:1 ratio is best but then you're using servos, motors, etc to duplicate the function of a cable, which is silly. Throttle return springs do break, sure, sometimes.
Closest thing I've seen to that in real life was a body shop customer with a 94 Vette, as a BS move to sway the judge the guy's lawyer started a dummy lawsuit against GM for their 'broken throttle return spring' that 'caused' the guy to do a few miles flat out on the LIE. It worked, and the guy has a real car guy relic - a speeding ticket for 165 in a 55 that he got out of! I've only gotten an 85 in a 35 and I'm still dealing with that! Almost triple, but not quite
. The only reason for FBW, in my opinion, is for electronic nannies, and I'm a big boy, thanks. I think we need better drivers instead of 'safer' cars, myself.
If you're this comfy with automotive electronics I'd love to see you tinker with a TDI or other modern diesel. Say you accomplished all you wanted with this on a 617, you're still stuck with a two valve IDI head, which besides the injection pump is the biggest hurdle to making power on this engine, and unless you're a machining wizard you're pretty much stuck with that. You'd end up with a fly by wire'd mechanical IDI diesel, which would be like using a late model OBDII ECU to control a Holley four barrel. Worst of both worlds.
Like I said though, don't let me discourage you, but I bought my 240D for a reason (it was cheap!) but I'd by another in a heartbeat for it's simple nature. While electronics can be made to be as or more reliable as a mechanical connection the manufacturers have little interest in that. You do, though. I'd like to take some late model EFI motor and strip it down to basics, get rid of the factory cockblocking (mostly software) and basically build the system I'd want. I think the software is just as important, the thing that makes the later cars have problems is the computer getting 'confused' and doing weird stuff.
Speaking of which, be careful testing that in the real world. That has been a problem with some factory systems, lurching out into traffic for no apparent reason and things like that. Factories get sued for that kind of thing, too, so I'm assuming they're going to be more rigorous in their design than you might. Just don't want everyone going 'where's cell? We never did find out what happened to that FBW setup he was working on!'
01-18-2010, 05:40 AM
#39
(01-18-2010, 05:05 AM)CID Vicious A linear 1:1 ratio is bestA linear throttle is rare. Look at the throttle cam on most g@ssers and you'll see a steep first half with a shallow second half. Its what helps automakers fool customers into thinking their engine has lots of torque. As a side note, thats also the exact same effect adjusting the ALDA returns...
ForcedInduction
01-18-2010, 05:40 AM
#39
(01-18-2010, 05:05 AM)CID Vicious A linear 1:1 ratio is bestA linear throttle is rare. Look at the throttle cam on most g@ssers and you'll see a steep first half with a shallow second half. Its what helps automakers fool customers into thinking their engine has lots of torque. As a side note, thats also the exact same effect adjusting the ALDA returns...
01-18-2010, 05:52 AM
#40
None of that splashes on what cell is trying to accomplish, and he may well be enjoying the truck for years set up like that while we're all scratching our heads over 'why?' He has the option of having the benefit without a bunch of factory engineers (and the legal and accounting departments that control them) mucking up a good idea. VW's, or Infinity's or whatever might have goofy gremlins in their throttle but his doesn't have to unless he wants it that way.
CID Vicious
01-18-2010, 05:52 AM #40Well, you just said it - even if the linear reaction weren't desirable the only benefit would be a bi-mode over a linkage or cable. The idea that the computer knows what throttle input is needed at a particular time is hilarious, like someone riding in the trunk giving you directions while blind. Manual transmissions would have gone bye-bye a long, long time ago if X algorithm was capable of being 'smarter' than a decent driver. However most drivers out there are horrible and have essentially no idea what they're doing, so these electronic nannies get put in so these lemmings won't kill themselves. So far the only thing I like about modern cars is EFI and ABS.
None of that splashes on what cell is trying to accomplish, and he may well be enjoying the truck for years set up like that while we're all scratching our heads over 'why?' He has the option of having the benefit without a bunch of factory engineers (and the legal and accounting departments that control them) mucking up a good idea. VW's, or Infinity's or whatever might have goofy gremlins in their throttle but his doesn't have to unless he wants it that way.
01-18-2010, 10:34 AM
#41
(01-18-2010, 02:16 AM)ForcedInduction Thats the big difference. A mechanical Diesel already works very well, adding an electric motor to control the mechanical throttle and governor is a huge evolutionary step backwards. If you want an electronic Diesel, you should get one that started that way. Either with an electronic IP, common rail or unit injectors.
It would be pointless to control the mechanical governor with electronics. I don't think anyone is suggesting that. What would make sense is to pull all the mechanical parts out of the governor and replace them with an electronic actuator and a bit of software. That would simplify things tremendously, provide better control characteristics, and would surely be an evolutionary step forward, no?
GREASY_BEAST
01-18-2010, 10:34 AM #41(01-18-2010, 02:16 AM)ForcedInduction Thats the big difference. A mechanical Diesel already works very well, adding an electric motor to control the mechanical throttle and governor is a huge evolutionary step backwards. If you want an electronic Diesel, you should get one that started that way. Either with an electronic IP, common rail or unit injectors.
It would be pointless to control the mechanical governor with electronics. I don't think anyone is suggesting that. What would make sense is to pull all the mechanical parts out of the governor and replace them with an electronic actuator and a bit of software. That would simplify things tremendously, provide better control characteristics, and would surely be an evolutionary step forward, no?
01-18-2010, 11:43 AM
#42
(01-18-2010, 05:05 AM)CID Vicious If you're this comfy with automotive electronics I'd love to see you tinker with a TDI or other modern diesel. Say you accomplished all you wanted with this on a 617, you're still stuck with a two valve IDI head,
oh I agree entirely. the plan for the next truck is to take a TDI, hook it up to a generator, and power the truck with an electric motor -- a diesel-electric truck. version 1.0 would have them simply coupled together, version 2.0 would add a bunch of batteries, effectively making a diesel-series hybrid.
however, that level of modification really requires a shop, which requires a cheap chunk of land outside of the city, which is definitely part of my plan, but is still about 5 years out. at that point I'll be able to get into machining, metal casting, I'll have a car lift... life will be good
but for now, I have to be a bit less ambitious and work within my limits (I live in a 650 sq ft condo with a single car garage.)
(01-18-2010, 05:05 AM)CID Vicious If you're this comfy with automotive electronics I'd love to see you tinker with a TDI or other modern diesel. Say you accomplished all you wanted with this on a 617, you're still stuck with a two valve IDI head,
oh I agree entirely. the plan for the next truck is to take a TDI, hook it up to a generator, and power the truck with an electric motor -- a diesel-electric truck. version 1.0 would have them simply coupled together, version 2.0 would add a bunch of batteries, effectively making a diesel-series hybrid.
however, that level of modification really requires a shop, which requires a cheap chunk of land outside of the city, which is definitely part of my plan, but is still about 5 years out. at that point I'll be able to get into machining, metal casting, I'll have a car lift... life will be good
but for now, I have to be a bit less ambitious and work within my limits (I live in a 650 sq ft condo with a single car garage.)
01-18-2010, 02:44 PM
#43
Throttle by wire makes no seance on a gas car unless you remove the throttle body ! no benefit's over a cable you will find out on your own!
If you want to put your family's safety in the control of a little black box do it by all means! I personally like having 100% control over my car in every aspect! starting throttle response transmissions gear and clutch
There isn't an automatic worth driving! unless you are drag racing!
automatic's are for pussy's and people that cant shift or use a clutch
This post was last modified: 01-18-2010, 02:59 PM by willbhere4u.
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-18-2010, 02:44 PM #43Throttle by wire makes seance on an electric car or diesel electric hybrid personally if you don't like mechanical linkages take all of the moving parts out of the equation rods pistons and all and go electric!
Throttle by wire makes no seance on a gas car unless you remove the throttle body ! no benefit's over a cable you will find out on your own!
If you want to put your family's safety in the control of a little black box do it by all means! I personally like having 100% control over my car in every aspect! starting throttle response transmissions gear and clutch
There isn't an automatic worth driving! unless you are drag racing!
automatic's are for pussy's and people that cant shift or use a clutch
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
![[Image: gt2256v.jpg]](http://jason.pepas.com/images/gt2256v.jpg)
01-18-2010, 06:42 PM
#45
(01-18-2010, 02:44 PM)willbhere4u If you want to put your family's safety in the control of a little black box do it by all means! I personally like having 100% control over my car in every aspect! starting throttle response transmissions gear and clutch
That's the point! Software does what it should, unless someone screwed up. Just like mechanical components. It's okay if you don't understand it, by all means keep your stuff mechanical, but please don't project your insecurity about electronics onto the problem by making (exaggerated) claims that just aren't true.
(01-18-2010, 02:44 PM)willbhere4u There isn't an automatic worth driving! unless you are drag racing!
automatic's are for pussy's and people that cant shift or use a clutch
wtf.
cell It's here!
That is one ungainly monster!
This post was last modified: 01-18-2010, 06:45 PM by GREASY_BEAST.
GREASY_BEAST
01-18-2010, 06:42 PM #45(01-18-2010, 02:44 PM)willbhere4u If you want to put your family's safety in the control of a little black box do it by all means! I personally like having 100% control over my car in every aspect! starting throttle response transmissions gear and clutch
That's the point! Software does what it should, unless someone screwed up. Just like mechanical components. It's okay if you don't understand it, by all means keep your stuff mechanical, but please don't project your insecurity about electronics onto the problem by making (exaggerated) claims that just aren't true.
(01-18-2010, 02:44 PM)willbhere4u There isn't an automatic worth driving! unless you are drag racing!
automatic's are for pussy's and people that cant shift or use a clutch
wtf.
cell It's here!
That is one ungainly monster!
01-18-2010, 09:11 PM
#46
I have no problem with electronics where they make seance! Like I said before it makes total seance when you eliminate the throttle body all together and let the engine management control the engine speed with valve timing and ignition timing and fueling
but when adding an actuator to an already existing throttle body whats the point a cable can open/close a THB with out any problems!!!
I have no real problems with automatics they just aren't any fun to drive like a minivan! Automatics are for soccer moms only real men can drive stick!
This post was last modified: 01-18-2010, 10:21 PM by willbhere4u.
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
willbhere4u
01-18-2010, 09:11 PM #46Nice turbo cell! keep up the good work sorry for hijacking the thread!!! Your progress is very interesting are you going to tie the VNT part of the turbo into the throttle by wire system?
I have no problem with electronics where they make seance! Like I said before it makes total seance when you eliminate the throttle body all together and let the engine management control the engine speed with valve timing and ignition timing and fueling
but when adding an actuator to an already existing throttle body whats the point a cable can open/close a THB with out any problems!!!
I have no real problems with automatics they just aren't any fun to drive like a minivan! Automatics are for soccer moms only real men can drive stick!
1987 300SDL 6spd manual om606.962 swap project
1985 300td euro 5spd wagon running
01-20-2010, 09:11 PM
#47
You can feed a PCM signal through a power transistor to control it. Also, having a feedback circuit will allow you to compensate for different conditions, loads, temperatures, etc...
1998 MB E300 - 194,000 miles. Current/future mods: 7.5mm pump elements, Holset HX40 Turbo feeding Eaton M90 supercharger through custom fabbed intake system. Aeroturbine muffler, scratch-built 4" stainless exhaust. EGR eliminated, ECM recalibrated, modified rack position feedback circuit. Porterfield RS brake pads, Bilstein sport shocks, VDO EGT, oil and boost gauges.
2000 Ford F250 - 150,000 miles. Current/future mods: DP Tuner 3 map chip, scratch-built high flow intake, high output IDM, open circuit fuel system, ball bearing turbo, South Bend full metal clutch, slotted and drilled rotors. Autometer Sport Comp EGT and boost gauges.
After looking at the video, you have a great start. You should really look into using a long stroke solenoid with a spring return like the Bosch electronic pumps have.
You can feed a PCM signal through a power transistor to control it. Also, having a feedback circuit will allow you to compensate for different conditions, loads, temperatures, etc...
1998 MB E300 - 194,000 miles. Current/future mods: 7.5mm pump elements, Holset HX40 Turbo feeding Eaton M90 supercharger through custom fabbed intake system. Aeroturbine muffler, scratch-built 4" stainless exhaust. EGR eliminated, ECM recalibrated, modified rack position feedback circuit. Porterfield RS brake pads, Bilstein sport shocks, VDO EGT, oil and boost gauges.
2000 Ford F250 - 150,000 miles. Current/future mods: DP Tuner 3 map chip, scratch-built high flow intake, high output IDM, open circuit fuel system, ball bearing turbo, South Bend full metal clutch, slotted and drilled rotors. Autometer Sport Comp EGT and boost gauges.
01-23-2010, 08:10 PM
#48
http://www.superturbodiesel.com/std/garr...l#pid10096
I got the Hella actuator on my GT2256V working, which is related to this project.
http://www.superturbodiesel.com/std/garr...l#pid10096
01-30-2010, 01:34 PM
#49
Cell, or anyone else, for that matter: have you thought of employing the cruise actuator for this task? We know its more than ample power-wise. I know very little about electronics, however I have some programming experience. I have little doubts that I can write all the software, its the hardware that I will need to learn something about. What type of circuitry would one require to convert all the signals to something the Arduino can deal with? Where should I look to learn about this stuff? I will order an Arduino kit later this month. Which one should I get? I have an extra pump and a few cruise servos to play with, so I would really like to hack something together!
This post was last modified: 01-30-2010, 01:40 PM by GREASY_BEAST.
GREASY_BEAST
01-30-2010, 01:34 PM #49See this thread for an argument as to why an electronic governor would be useful with no other pump modifications (i.e. stock elements). The standard governor is incapable of providing maximum fueling in the high part of the rpm range. Electronic rack control will allow full fueling at any rpm. This will yield more power.
Cell, or anyone else, for that matter: have you thought of employing the cruise actuator for this task? We know its more than ample power-wise. I know very little about electronics, however I have some programming experience. I have little doubts that I can write all the software, its the hardware that I will need to learn something about. What type of circuitry would one require to convert all the signals to something the Arduino can deal with? Where should I look to learn about this stuff? I will order an Arduino kit later this month. Which one should I get? I have an extra pump and a few cruise servos to play with, so I would really like to hack something together!
01-30-2010, 10:50 PM
#50
(01-30-2010, 01:34 PM)GREASY_BEAST See this thread for an argument as to why an electronic governor would be useful with no other pump modifications (i.e. stock elements). The standard governor is incapable of providing maximum fueling in the high part of the rpm range. Electronic rack control will allow full fueling at any rpm. This will yield more power.
I was just thinking that after reading that thread
(01-30-2010, 01:34 PM)GREASY_BEAST Cell, or anyone else, for that matter: have you thought of employing the cruise actuator for this task? We know its more than ample power-wise.
cool. my engine came out of a '79 300SD, and it had a cruise actuator which appeared to be a big vacuum can. Is that what you are talking about?
(01-30-2010, 01:34 PM)GREASY_BEAST I know very little about electronics, however I have some programming experience. I have little doubts that I can write all the software, its the hardware that I will need to learn something about. What type of circuitry would one require to convert all the signals to something the Arduino can deal with? Where should I look to learn about this stuff? I will order an Arduino kit later this month. Which one should I get? I have an extra pump and a few cruise servos to play with, so I would really like to hack something together!
sweet! I came from the same place you are: I had a programming background, but knew nothing about electronics. I think microcontrollers (ie, the arduino) is the perfect way to branch from programming into electronics.
I'd start off with lady ada's boarduino, because it makes breadboard prototyping easier:
http://www.adafruit.com/index.php?main_p...3f891228e9
if you get that version of the boarduino (the non-usb one), you will probably also want an ftdi cable as well as a programmer:
http://www.adafruit.com/index.php?main_p...3f891228e9
http://www.adafruit.com/index.php?main_p...b552bbef1f
also pick up a power supply:
http://www.adafruit.com/index.php?main_p...3f891228e9
ladyada is really cool in my book -- pretty much everything about her business is open source and hobbyist oriented. you can even download the gerber files for her boards, tweak them, and make you own custom variant of her stuff. slick! I look at the way she does business and then daydream about how incredible it would be if people like mercedes offered downloadable CAD files of all the parts in my engine...
you'll want at least one breadboard. I don't recommend the small ones ladyada has (the connection points are too stiff). I use these from electronics goldmine:
http://www.goldmine-elec-products.com/pr...mber=G8534
as for sensors, a good place to start would be temperature and pressure. for temperature, you can grab an LM335A from sparkfun:
http://www.sparkfun.com/commerce/product...ts_id=9438
(I'm using several LM35's in my truck, which are almost the same thing but with a slightly different scale).
for pressure, grab a motorola sensor:
http://www.futureelectronics.com/en/Tech...ALE9341934
note: I found the above by doing a search for mpx4115ap on http://octopart.com/
at some point you'll want a bunch of small parts on hand so you don't have to order individual resistors, etc each time you do a project. to stock up on parts:
try futurlec's value packs:
http://www.futurlec.com/ValuePacks.shtml
search for "kit", "assortment" and "grab bag" on http://www.jameco.com
also, hit up the assortments and "goldpaks" at goldmine:
http://www.goldmine-elec-products.com/de...?dept=1409
http://www.goldmine-elec-products.com/pr...?dept=1053
also, i've gotten in the habit that everytime I order something from goldmine, I order one of their "surprise" boxes:
http://www.goldmine-elec-products.com/pr...mber=G2200
half of it will be unidentifiable or junk, but its kind of fun, and they're cheap
if you don't have any tools, ladyada has some good tool guides:
http://www.ladyada.net/library/equipt/kits.html
another great site to browse is http://www.sparkfun.com/ they started out offering "break-out" boards (small pcb's which take a surface mount component and break it out to the standard 0.1" breadboard pinout spacing), but they've really grown their lineup over the years.
I'll be happy to help when you need it
(01-30-2010, 01:34 PM)GREASY_BEAST See this thread for an argument as to why an electronic governor would be useful with no other pump modifications (i.e. stock elements). The standard governor is incapable of providing maximum fueling in the high part of the rpm range. Electronic rack control will allow full fueling at any rpm. This will yield more power.
I was just thinking that after reading that thread
(01-30-2010, 01:34 PM)GREASY_BEAST Cell, or anyone else, for that matter: have you thought of employing the cruise actuator for this task? We know its more than ample power-wise.
cool. my engine came out of a '79 300SD, and it had a cruise actuator which appeared to be a big vacuum can. Is that what you are talking about?
(01-30-2010, 01:34 PM)GREASY_BEAST I know very little about electronics, however I have some programming experience. I have little doubts that I can write all the software, its the hardware that I will need to learn something about. What type of circuitry would one require to convert all the signals to something the Arduino can deal with? Where should I look to learn about this stuff? I will order an Arduino kit later this month. Which one should I get? I have an extra pump and a few cruise servos to play with, so I would really like to hack something together!
sweet! I came from the same place you are: I had a programming background, but knew nothing about electronics. I think microcontrollers (ie, the arduino) is the perfect way to branch from programming into electronics.
I'd start off with lady ada's boarduino, because it makes breadboard prototyping easier:
http://www.adafruit.com/index.php?main_p...3f891228e9
if you get that version of the boarduino (the non-usb one), you will probably also want an ftdi cable as well as a programmer:
http://www.adafruit.com/index.php?main_p...3f891228e9
http://www.adafruit.com/index.php?main_p...b552bbef1f
also pick up a power supply:
http://www.adafruit.com/index.php?main_p...3f891228e9
ladyada is really cool in my book -- pretty much everything about her business is open source and hobbyist oriented. you can even download the gerber files for her boards, tweak them, and make you own custom variant of her stuff. slick! I look at the way she does business and then daydream about how incredible it would be if people like mercedes offered downloadable CAD files of all the parts in my engine...
you'll want at least one breadboard. I don't recommend the small ones ladyada has (the connection points are too stiff). I use these from electronics goldmine:
http://www.goldmine-elec-products.com/pr...mber=G8534
as for sensors, a good place to start would be temperature and pressure. for temperature, you can grab an LM335A from sparkfun:
http://www.sparkfun.com/commerce/product...ts_id=9438
(I'm using several LM35's in my truck, which are almost the same thing but with a slightly different scale).
for pressure, grab a motorola sensor:
http://www.futureelectronics.com/en/Tech...ALE9341934
note: I found the above by doing a search for mpx4115ap on http://octopart.com/
at some point you'll want a bunch of small parts on hand so you don't have to order individual resistors, etc each time you do a project. to stock up on parts:
try futurlec's value packs:
http://www.futurlec.com/ValuePacks.shtml
search for "kit", "assortment" and "grab bag" on http://www.jameco.com
also, hit up the assortments and "goldpaks" at goldmine:
http://www.goldmine-elec-products.com/de...?dept=1409
http://www.goldmine-elec-products.com/pr...?dept=1053
also, i've gotten in the habit that everytime I order something from goldmine, I order one of their "surprise" boxes:
http://www.goldmine-elec-products.com/pr...mber=G2200
half of it will be unidentifiable or junk, but its kind of fun, and they're cheap
if you don't have any tools, ladyada has some good tool guides:
http://www.ladyada.net/library/equipt/kits.html
another great site to browse is http://www.sparkfun.com/ they started out offering "break-out" boards (small pcb's which take a surface mount component and break it out to the standard 0.1" breadboard pinout spacing), but they've really grown their lineup over the years.
I'll be happy to help when you need it
Users browsing this thread:
29 Guest(s)
29 Guest(s)
Users browsing this thread:
29 Guest(s)
29 Guest(s)