A friend of mine (my son, actually) is trying to get his ECU to talk to the dash of a Dodge SRT4. Apparently it uses some weird comms transport mode that isn't quite NRZ and isn't quite anything else. It's too old for CAN.

His main problem is that it seems to have a check byte at the end of the packet, but he can't find out what to write into it so that the dash accepts it.

I told him that I'd ask the nerds here. There just might be someone who knows...

Does anyone have any idea?

Last edited by tonyw on 28-Nov-2014 at 08:56 PM.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

@tonyw

What Dodge model is your Son's SRT4? Is it the Caliber? If yes, what year?

My 2007 Dodge Nitro uses CAN bus for communications (I am pretty sure), and believe that your Son's SRT4 probably uses the same, but I am a novice about such things. You are in Australia (or was it New Zealand), so the model names of Dodge products will be different than they are here, but the SRT4 does not mean anything to me, but I will look it up.

I would be interested in finding out what the correct answer is, when you finally find out, so please let me know. I don't know that much about automobile computer communications, but would like to learn more. I plan on keeping my Dodge Nitro for a very long time and would like to be able to modify and upgrade some of it's features when things fail and need replacement. It will be a great project to learn on and the size and features it has, fit my needs perfectly, which is why I want to keep it long term (5,000lb towing capability, 4-Wheel Drive, fold flat rear seats which make a perfect place to sleep on certain trips, or camping by myself, has a DVD/CD player that holds up to 6 discs to play movies on the flip-down LCD for the rear seat passengers, plus additional AV inputs and Satellite capable radio, TIPM tire pressure monitoring, and many other electronic monitoring features I can't remember at the moment). My Nitro is the RT version with the High Output 4.0L V6 engine & 5-speed automatic transmission, which actually gets better gas mileage that the crappy 3.7L V6 standard engine & 4-speed automatic, or 6-speed manual transmission provide.

I would love to eventually install a Hybrid Gasoline/Electric, (or even better, Hybrid turbocharged Diesel/Electric) engine, like the V6 Hybrid Gasoline/Electric used in the Toyota Highlander & Lexus R400, which are about the same size SUV's as my Nitro and are also available in 4-Wheel Drive models. I could probably use many of the Toyota/Lexus parts in my Nitro, to complete a transformation to a Hybrid vehicle, but would have to be lucky and find them at the right price from a wrecking yard to make the project feasible. If/When I ever attempt to do such a conversion, I will need to learn much more about the electronic systems and computers that control both vehicles, to make it all work without losing the features that are currently installed in my Nitro.

Maybe I will find a genius car mechanic that also knows how to modify auto computers and can meld the Toyota/Lexus computer system that controls the Hybrid propulsion system and regenerative braking with my existing Nitro electronics.

Last edited by amigadave on 24-Nov-2014 at 05:46 AM.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@tonyw

Okay, just did some reading on both the Dodge SRT4, which I now think you are referring to the Dodge Neon SRT4 which was produced between 2003 and 2005 (or was it 2006), and more reading about the CAN bus standard.

I found a list here: http://www.auterraweb.com/aboutcan.html that attempts to show all of the models of vehicles made between 2003 and 2007 that contained the CAN bus.

All USA vehicles produced after 2006 were required to include CAN bus as part of the OBD II standard for vehicle communications. I skimmed quickly through these web pages, so don't take anything I write literally and find your own answers, but it appears that the Dodge Neon SRT4 is not one of the models that use the CAN bus protocol, so none of what I wrote in my first reply will help you.

What is your Son trying to do with his SRT4's dash gauges, by communicating with them?

That car won lots of awards and seems to be a very fun car to own. Not sure, but it appears that many are still being raced in several different types of races.

Last edited by amigadave on 24-Nov-2014 at 06:13 AM.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@amigadave

My son has a business designing and marketing after-market ECUs:
http://www.adaptronic.com.au/

His basic design is sufficiently flexible that he can adapt it to almost any new model that a customer brings along. His current project is this Dodge. It's an American model, LHD, FWD, with this weird means of talking to the dash instruments.

There is apparently another model of after-market ECU that keeps the original ECU just to talk to the dash instruments, but he wants to do better. A lot of his work these days is with race cars, using his adaptive firmware to get the most out of a modified engine.

You can communicate with him direct via that website. I'm sure he'd be pleased to hear from you.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

@tonyw

The only information I can find suggests that even the 1st generation Neons had OBD 2.

I found this page which mentions checksum 9D85. Not sure if that's what he needs.

http://www.austec.co.uk/performance-tuning.asp?Make=CHRYSLER&Model=NEON&Engine=2000%2016V

Is it not possible to control the instruments directly or does it need to be plug and play?

@Rob

Thanks for that info, I'll pass it on.

He wants his ECU to be able to drive the instruments directly. Apparaently no one else has been able to do it yet.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

Hi everyone, thank you!

This is for the SRT4, which isn't CAN.

Yes, it needs to be plug and play so we need to mimic the factory comms.

I was hoping to be able to crack the checksum so I could generate the packets myself to say whatever I wanted but so far I haven't been able to.

I have worked out it's definitely not CRC though.

So far I've been brute forcing it which should work well *enough* for now; it's just inelegant and if we need to more clever stuff it won't be enough...

Thanks all!
Andy

By the way, thanks to paternal influence and support I grew up programming C64s and Amigas :)

@AndyW

By the way, working now :)

On youtube!

Thanks!
Andy

Last edited by AndyW on 26-Nov-2014 at 05:49 AM.

@AndyW

Awesome Andy!

What was the trick, or must you keep it secret, now that you have cracked the code?

You sound like the right guy to contact if/when I ever decide to do any modifications to my Dodge Nitro RT 4x4.

Is it even feasible to mix and match parts from one manufacturer with those from another? I would only consider doing such a project after my current engine has reached the end of it's useful life and needs a rebuild.

As I wrote earlier, I would like to modify my Dodge Nitro with Lexus/Toyota parts to create a Turbo Diesel/Electric Hybrid vehicle, that might be capable of 30+ mpg, but retain it's 5,000lb towing capacity and sporty performance, which I currently enjoy with the 4.0L High Output gasoline V6 engine. I only get about 21 to 22 mpg on the highway and about 18 to 20 mpg driving around town where I live (would be even lower if I lived in a bigger city with more stop signs and stop lights) with the current engine & drive train combo.

The Lexus R400 gets close to 30+ mpg on gasoline, in both highway and city driving, due to the electric motor(s) combined with it's 250hp + V6 gasoline engine, so I am certain that I can do better with a Turbocharged Diesel combined with those same electric motor(s), but also need to choose the right size and output Turbo Diesel.

If I understand how the Lexus drive train works correctly, it is not like typical gasoline/electric hybrid vehicles, which use a large electric motor to propel the vehicle and use only a small gasoline (or diesel) motor to generate electricity to supplement and recharge the electric batteries.

The Dodge Nitro and Lexus R400 are approximately the same size and weight, which makes me believe that this crazy project might be feasible. I know it will cost thousands to accomplish, but buying a new vehicle with similar capabilities will cost over $35,000 by the time I am ready to attempt this project. Buying a new Lexus R400 Hybrid is probably about $55,000.

Last edited by amigadave on 26-Nov-2014 at 07:29 AM.
Last edited by amigadave on 26-Nov-2014 at 07:27 AM.
Last edited by amigadave on 26-Nov-2014 at 07:25 AM.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@amigadave

Hi Dave, no trick, just using a really big hammer.

I have a test jig which I can use to simulate different crank / cam angle waveforms (and RPMs), and can output a square wave or sine-ish wave for vehicle speeds, and that's controlled via USB.

So I programmed a micro to snoop the comms from the factory ECU and filter out the packets so I only have the one with the correct ID for the speedo / RPM (which I'd previously determined by wiggling those and seeing which packets had data that changed). That micro sends out the packet data over RS232 so it can be easily read by a PC.

The software then steps through the different RPM / speed combinations, reading out the packet values and waiting until the factory ECU has the correct combination of speed and RPM. For RPM it's not so bad, but for speed the signal seems to be a bit noisy, so it has to wait until it gets the exact 16-bit value that I've decided that I want (which is taken from feeding the speed into the unit and getting the most common value). Then when we have the correct 32-bits of data, it records the checksum.

It took about 5 hours or so to run through all 8192 combinations.

That way I just needed 3 tables:

1 x 64-entry, 16-bit table for the RPM values (every 128 RPM)
1 x 128-entry, 16-bit table for the vehicle speeds (every 2 km/h, factory speedo goes to 160 mph = 256 km/h)
1 x 64 x 128 2D array with the 8-bit checksums in it.

All of those can fit easily in my micro which has about 48 kb of flash that's easy to access.

I know, it's really inelegant; I'm not a fan and later on I'd like to do it properly so I can output whatever values I want, not just the closest 128 RPM and 2 km/h, but for what I need it for, it works.


Regarding the engine conversion... hmmm. I think Dad aka Tony will have some more well researched things to say than I do, but my initial main concern is that the drivetrain control system from the R400 is likely to be very highly integrated in itself and even with the braking system. I remember reading in the SAE mag (Auto Engineer I think it's called) that they had a lot of trouble with the R400 doing regen, because all their previous work was with the FWD Prius, so when you do heavy regen on that, it brakes the front wheels, which is where you want most of the brake bias anyway. With a RWD chassis you kind of have to start with rear brake bias when you're using regen and when the rear bias becomes too much for the current front / rear balance it reduces the regen and starts increasing the front brakes.

Also the control of the ICE would be heavily integrated with the electric drive, charging control and battery monitoring, I would have thought, so I would think that either means:

1) Making it think that it's driving a V6 gasoline engine when it's actually driving a Diesel engine. So that means using a normal Diesel ECU for the V6 turbo diesel engine, and somehow keeping the ECU from the R400 happy. Needless to say the rest of the R400 system including the electronic braking system to muck around with the bias as I mentioned before would need to be swapped over and made work on the new chassis, which would have wholly different braking hardware I'm sure (at least keeping it with the one manufacturer is a bit easier)

2) Rolling your own control system, and just using the motor / gearbox / whatever magic is on the R400. This is probably the way I would do it, and the way that people tend to do pure-electric conversions from what I can tell. Even then they sometimes keep the factory ECU to integrate with the rest of the car (one of my ex-colleagues has just done this with a Smart car, factory stability control etc all still works)

Either way, the control side of things will be a LOT of work!

@AndyW

Great to see you got it working. Very informative collection of videos too.

@amigadave

Quote:

Is it even feasible to mix and match parts from one manufacturer with those from another?

Anything is feasible if you're prepared to throw money at it.

Fitting the power train from a Lexus Hybrid into you're dodge sounds like it will require huge amounts of fabrication.

The engine in the R400 is transverse whereas it is inline on the Nitro.

Fitting the R400 gearbox into the engine bay may require the chassis being modified as well as custom mounts for it and the engine. The gear shift is located further forward and higher in R400 so that poses it own problems too.

You will also need custom brakets for things like Aircon and Power Steering pumps if they are driven by the engine, some vehicles use electric power steering.

Where will the rear motor and batteries fit? Will you lose that flat rear load area in the process.

Then you've got the rest of the running gear to sort out which is likely going to need a lot of custom parts including drive shafts, prop shaft and a hub assembly that works with the brake regen and also your existing ABS setup.

The R400 looks like it has a lot better aerodynamics too, how does that affect fuel economy.

To be honest I don't think it will be worth it for the time, cost and technical headaches involved.

Investigating a more modern conventional petrol engine may be worthwhile. I know that Ford do a 2.0 I4 Ecoboost unit that will produce similar levels of torque to your current unit and the 2.3 unit going in some Mustangs produces even more. There'll no doubt be quite a few crashed donor mustangs willing to give up their engine and electronics in the next few years.
You'd need a custom bell housing to mate the gearbox to the engine but over than than it's just mainly a case of brackets for fitting the engine and ancillaries.

Chrysler may even do a more efficient engine that will bolt straight up to the gearbox.

I think an LPG conversion would be the most cost effective solution and the vehicle would be off the road for a lot shorter period too. I guess that depends on the availability of LPG where you are.

Lastly it might be worth finding out if a simple ECU recalibration would give some improved economy.

@Rob

And we know who makes self-adapting, self-optimising ECUs...

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

@tonyw

I think in this instance a reflash of the factory ECU might be a better way to go than fitting an aftermarket ECU and getting it to integrate with the factory electronics. Also we don't optimise for economy or torque because no-one installs sensors to measure that!

But the only way you'll get more efficiency out of it through tuning is by making a different tradeoff from the ones made by Dodge. Eg if they run it at stoich to keep NOx emissions low, and you're prepared to make your car more NOxious in the name of fuel economy, then you could probably reflash it with leaner target lambdas on cruise and save a bit there, but it won't go from 22 mpg to 30 mpg by doing this.

The other way to do it is to get a lighter vehicle that has 30 mpg from factory as a runabout, and use the Nitro when you need to tow. Not sure how much taxes / registration etc cost to see if that's an economically viable option, and also I'm not sure how much of the plan is to overcome the fuel economy problem (I doubt the costs of the conversion would overcome the difference in fuel economy over the life of the vehicle for example) and how much is to pick a rewarding project to do...

I think the Euro market had a Diesel Nitro, so another option would be to swap the car for that if you can get one in the States (or swap out the driveline / electronics / fuel tank / etc... sounds like a lot of work though) - I know some people do LPG over Diesel conversions (mix LPG in with the air going in) - supposedly it helps with power but I've never had one on a dyno to actually see; and I don't know if it would help with economy on cruise (if you'd end up using less Diesel to produce the same torque)...

@AndyW

The Nitro was available in Europe, Canada and a few other countries outside the USA with a CRD Diesel, which probably would be the most logical choice to swap to and then find an engine tuner/builder, who could turbocharge it, to increase performance and possibly also improve fuel economy.

My idea to use some Lexus/Toyota parts from the R400h/R450h, or Highlander Hybrid to make a Hybrid Dodge Nitro possible did not include trying to use the transverse mounted 3.5 gasoline V6 driving the front wheels (the R400h/R450h All-Wheel Drive uses electric motors to provide propulsion to the rear wheels). All of my thoughts about such a project are purely fantasy and dreams at this point in time, but I think they could become reality, if I found the right mechanic(s) and/or engine tuner(s)/builder(s), to work with me, and can find someone like AndyW who knows how to reprogram the ECU's and adapt things to work together, that were never intended to be combined together.

The Dodge Nitro is a rear wheel drive vehicle and the 4x4 model uses an transfer case with a simple electronic "On/Off" switch, which provide either rear wheel drive, or 4 wheel drive in a locked differential condition, which can easily cause damage when drivers forget to turn the 4 wheel drive system off and go from icy or snow covered roads to pavement that is dry, or even wet pavement that still provides enough traction to damage the drive train during turns in the locked differential mode. Not a great design, for drivers who are not careful, or unaware of how much damage can occur. One of my thoughts was to remove the transfer case and drive shafts that propel the front wheels of the Nitro and to install the electric motors that are used to propel the rear wheels, from the Lexus R400h/R450h, or Toyota Highlander Hybrid. This would be one of the simplest designs, but might need some kind of torque sensors to balance the amount of power coming from the electric motors on the front wheels, to the power being provided by the rear wheel drive of the Nitro's existing transmission and engine. Once that was successfully implemented, I could also consider either adding a dual fuel system, using LPG and Gasoline with the stock RT Nitro High Output engine, or swapping out that engine for the CRD Diesel and adding a turbocharger.

There are dozens of possibilities and all of them are sort of crazy, but I will continue to think about them and do my research, including trying to find the right people who could work on such a project with me, who have the right skill set to solve some of the challenges. I have already solved the question of where the electric batteries would be located.

Thanks for everyone's feedback and opinions, and congrats to AndyW for his brute force success. Hopefully you will figure out a more elegant solution later.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

I also have a 1970 Volkswagen Karmann Ghia coupe that is a total restore project. It needs a new or rebuilt engine, interior and floor panels (or complete rolling chassis), as the floor has rusted out and the seats are ready to fall to the ground at any time. The exterior also needs body work in a couple of spots and then a nice custom paint job would finish it off perfectly.

I will probably keep it mostly stock with the exception of larger pistons and a longer throw crankshaft, with dual carburetors, or a single 2 barrel carburetor, but I have toyed with the idea of leaving the Nitro the way it is, and instead making the Karmann Ghia into an all electric car for local driving trips to the store and back, or to doctor appointments. That would be a much easier project than what I am thinking of for the Nitro. Ideally, I would like to eventually do both projects. Far more interesting than spending $40,000 to $65,000 on a brand new vehicle, and more satisfying when you get the results you really want and get to save a little money in the process.

Even if it costs me $20,000 to complete the Nitro project, I would end up with what I want and save at least half the cost of buying even a used, but in good condition, low mileage Hybrid Lexus/Toyota 4x4 SUV.

Last edited by amigadave on 28-Nov-2014 at 03:21 AM.
Last edited by amigadave on 27-Nov-2014 at 03:51 AM.

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@amigadave

Quote:

This would be one of the simplest designs, but might need some kind of torque sensors to balance the amount of power coming from the electric motors on the front wheels, to the power being provided by the rear wheel drive of the Nitro's existing transmission and engine.

If you're using a DC motor, the "torque sensor" can be simply a current sense, since the torque is directly proportional to the load current. For AC motors, it's not quite as cleanly proportional, but a good start.

_________________
cheers
tony

Hyperion Support Forum: http://forum.hyperion-entertainment.biz/index.php

What an absolutely cool thread, love it, very informative and interesting to read, thanks!

Rude!

_________________
2017 Camaro 2SS

Sorry for hijacking this thread TonyW, but since it appears that AndyW hasn't gotten any useful answers from members here on solving his questions, I hope you don't mind.

Does anyone here have any experience in doing a DIY electric vehicle conversion, and can give me advice on doing a conversion on the Karmann Ghia?

_________________
Amiga! The computer that inspired so many, to accomplish so much, but has ended up in the hands of . . . . . . . . . .

@amigadave

is the ghia a type 14 or a type 34?

Anyhoo, electric conversion kits for volkswagens and porsches are plentiful. Here's a company with DC and AC kits, including adapter plates for both Type 1 and 3 VWs:
http://www.evwest.com/catalog/product_info.php?products_id=167
http://www.evwest.com/catalog/product_info.php?products_id=242

So one of those should fit your Ghia. Note that the difference in price is because one has some extra cables, switches and whatnot in the package. They also have a "complete kit" for the type 1, which presumably includes EVERY little thing you would need for a type 1, but if the battery rack would fit a type 14 I'm not sure.

I really can't see any obvious flaws in buying those kits, but haven't actually tried one. I've got a gear box and center house/rear axle for an MF35 sitting outside here waiting for a project, while working on the adapters (housing to motor flange and axle to clutch) for that I'm thinking of using the center housing and rear axle for a PTO powered trailerfor my MF135. They should have identical ratio PTO to wheel (32/38 from PTO to pinion plus 6/37 pinion to ring gear) when set to "ground speed PTO".

_________________
This weeks pet peeve:
Using "voltage" instead of "potential", which leads to inventing new words like "amperage" instead of "current" (I, measured in A) or possible "charge" (amperehours, Ah or Coulomb, C). Sometimes I don't even know what people mean.