• NAXJA is having its 18th annual March Membership Drive!!!
    Everyone who joins or renews during March will be entered into a drawing!
    More Information - Join/Renew
  • Welcome to the new NAXJA Forum! If your password does not work, please use "Forgot your password?" link on the log-in page. Please feel free to reach out to [email protected] if we can provide any assistance.

Paging Renix gearheads!

5-90

NAXJA Forum User
Location
Hammerspace
Here's a fun one for all of you (could also apply to anyone with more computer software experience than myself - I'm a hardware guy so far...)

I'm looking for someone who has already captured and recorded the serial data stream from the Renix ECU, or for someone who can help me do it. I've got an oddball idea for a project, and it requires having a copy of the serial data stream and the conditions that generated it for me to work with.

Essentially, I want to duplicate the functionality of the MT2500 from Snap-On that allows it to display realtime driveability data to the user - and possibly make this something that can be permanently mounted into the vehicle. Any help on this one?

5-90

EDIT - Additional - My initial plan for this is as a standalone device, but a version that ports the serial data stream to a PalmOS device for decoding may also be a possibility. Either way, I need the serial data stream to take apart...

5-90
 
I've been thinking about doing the same thing. My idea was to use a laptop to capture the data. The biggest hurdle is figuring out the meaning of the data stream. There are a number of parameters on the data bus so one chore is to figure out what bits and/or labels are associated with each parameter. The other chore is to figure out what the scaling is for each of the parameters. It might be worth taking a first look at the bus with an oscilloscope. Another scheme might be to hook up an ECU on the bench and then set up to simulate one sensor at a time while monitoring the bus.
 
That is why I am also collecting a "bench set" of Renix electronics - but I also have occasional access to a Snap-On MT2500...

I'm thinking of using a BASIC Stamp to interpret the data stream and drive an LCD display, which will keep the package small and potentially permanently mountable. I'm actually thinking of something that could be mounted in place of the click module, which would yield constant feedback from the engine when needed.

I honestly don't think just reading with an o-scope will be useful, the data stream itself needs to be captured for offline analysis. That's part of the reason for the "bench set" of electronics (donations gladly accepted!) - this will allow me to simulate conditions and should help to reveal the meaning of various parts of the data flow. However, samplings of the data stream in situ will also be required - both to determine the effects of various conditions on the interpretation of the data flow and to sort out how it is all put together.

Therefore, I need to get my hands on A) captured data for offline analysis OR the means to capture data using (preferably) a PalmOS device to be loaded to desktop PC; and/or B) a full set of Renix engine control electronics - to include both primary variations for manual or automatic gearbox. I have a partial set of sensors, a spare distributor or two, a throttle body, and a TCU, but I need ECM, at least one full set of sensors, and some other stuff I've not figured out yet (including a working ignition module - there are some interesting things in there as well.)

Whether or not this turns into a saleable project depends on how much I can make work at this point - and how much help I can get amassing the electronics. I'll keep track of donors for the first run of working projects for Beta testing and updates, but this isa and shall remain a "blue sky" project until I can get a bench set and a captured data stream together...

5-90
 
5-90, I have collected an old 486 notebook with original versions of Windows and Win95, to set up a portable ASCII reader using Terminal or other VT100 emulator. This is what we used to use with other PLC control interfaces to read the data stream, although I have yet to try it on the RENIX.

If you can get anywhere, let me know?

BTW, I have a copy of the Superchips tuner Windstar ECU modification demo, if you want to compare what they did for ideas?
 
The nice thing about VT100 is it runs on pretty much anything. I've found a PalmOS version of Hyperterm, but I don't know how to use it (yet.)

Do you have any data that you might have collected already? Perhaps we should co-ordinate backchannel - I tried posting at JeepSpeed, and no-one's answered yet...

5-90
 
Post some more info, #####'s ,I'd looking into that for you as a really soon to be Electrical Eng., I'd love to play with the number's
 
What year is your XJ? I'm working on a project for 87-90XJ (mostly 4.0.)

The problem we have here is that there is no cost-effective data reader for the Renix ECU, since they are not OBD-compliant.

This is more of a problem for software engineering (one of my current minors) - most of the hardware should be doable with COTS goods. I am always willing to accept input if you have anything you think is worth adding! I have a couple other ideas I might want to talk to you about, tho...

While I've got some reverse engineering in mind (perhaps adapting the HO CPS to talk to the Renix ECU?) the main reason I want to collect Renix electronics is to take apart the data stream and find out what means what so it can be torn to bits...

5-90
 
Here's my dumb question of the day: how did Snap-on get the data? Do you suppose they reverse-engineered, or did they just ask the manufacturer for the pertinent data? I wonder if it would be possible to represent oneself as an aspiring manufacturer of test equipment (not really dishonest, just awfully optimistic :) ) and buy the data from somewhere? I would imagine that Renault or Bendix would be a better bet than Chrysler, for some reason. It's a long shot, but I also wonder if there is some legal requirement covering the disclosure of this stuff. Given the fact that the government has a role in many automotive areas including emission control, even to the point of now mandating OBD, there might even be a requirement that the information be publicly available.
 
The code and stuff should be able to be had from Atica since they bought the Renix venture in 1986 - it helps that they specialize in auto diagnostic electronics...!

I have a few of the RenixII ECUs sitting around - mostly from other projects and while you can easily get them to "idle" on the bench - generally I tire before I clock em to full running -- the damed O2 sensor (twisted-a$s resistance type) seems to be a major hassle - ass soon as I get to closed loop the ECU reacts quickly and tosses stuff into fail-safe....

I've been meaning to run some road tests with one of the portable scopes and all that - the diagnostic port does NOT appear to free-stream data... (IIRC, it's been almost 2-years) Sopme 35-pin ECUs can be had from Volvo 440s, 460s, & 480 (and Renaults) with 1.7L engines - the data should be similar in composition --

Renix ran a line of PLC software for a while (I saw it only in a few odd places running on DOS or UX), not in an embedded config, so it's not much help... Usually, that data was passed in a TLV message with header and trailer messaging - Can't say if the ECU does the same --

Charles White (a tech writer guy, kinda like someone else we all know...) built a breif Renix section into one of his many Haynes publications "Automotive Engine Management & Fuel Injection Manual" <--- there should be a "PIN" series for Renix stuff - this series had a BUNCH of good data on values - not necessarily how they were passed in the data stream, but an indication that "the mixture off-set as generated from O2 sensor data would hit 254 and latch there" in some condition, so you could watch the data and set the contition then look for a 254 to pop -up -- like that...

I played at it for a while a few years ago (as mentioned), but figured that the bang for the buck payoff was in sensor falsing and like that -- PLUS with all of the engine control systems that are out there and the totally basic needs of the RenixII 4.0s it'd prolly be easier to grab one of the newer controllers and start from that direction (at least that's what I tell myself)
 
After all that...

... it'd be very easy to watch the O2 sensor data, the pulse duration on an injector, data from the TPS, MAP, CTS, IATS with very simple hardware (buffered) - a small PIC could easily sequence through those items and display real-time running parameters based on mapped conversion data...

I've done each of those (never built a consolodated report tool) - and I imagine that feedback from the coil wouldn't take much (for output voltage) -- timing may be a bit difficult, but also probably possible...

if that's all you're interrested in...
 
Ooh, Jeeps and computer stuff, gotta love it. Any idea what the format of the datastream is? I mean, is it by some strange coincidence RS232 serial (maybe adjusted for TTL levels) or something a lot more odd?
 
Is this just for pure analysis or will it eventually turn into a performance or diagnostic product? I can't believe the difficulty in finding info on the RenixII computer, it was used in a ton of vehichles. Good Luck.
 
Both - it's something I started thinking about once I found out the Snap-On MT2500 is the cheapest scanner that will "speak Jeep" - at $1400.

I don't mind building a small dedicated unit that I could make standalone or installable, and could probably produce for a couple hundred or so either way (the average OBD reader is about $150-200, OBD-II $200-400.)

I found an outfit called Autodiagnos (European) that I locate with a search for Renix, but they said they can't do anything. Next stop - Actia! They grew out of a Renix purchase in the 80's...

5-90
 
Hey satan -
I was just rereading this thread (and wanted to get it up top again anyhow...) and saw your post - I don't suppose you still have the info on how you did all the "little" monitors? That's something else I'd like to play with, and it may be a precursor to what I'm trying to do...

5-90
 
This is really a great idea 5-90. One hell of an undertaking tho. I also own a Mitsubishi 3000GT and frequent a forums for them. There is a guy that did this for the hybrid ECU version of these cars ('94/'95 were hybrid OBDI/OBDII). It took him about 8 months to get all the software written and all the signals that the ECU sees figured out. Lots of work but deffinatly worth it in the end to be able to know what the ECU is seeing while you are driving. Makes tune ups and preventative maintenance ALOT easier.
 
I'm a EE/CS/ME and do engineering for a living. If you can get the info, I will design the schematic and layout your pcb and can get proto's for free. I also have all the development tools for PIC including the C compiler as well as Zylinx and the Motorola HC16 line. I've always wanted to do this but have been so busy building other things that I haven't had time to hook up an analyzer to the data stream to look at it.

Are the levels RS-232 or RS-485? If so we could spy it with a simple communications program and start decyphering it. I have the Renix diagnostics manual but never have taken the time to see if it has anything in it about the data stream. I do however remember seeing a Txd and Rxd set of lines on the schematics. I also have spare Renix computers.

Here is a link to a little device I just finished for a customer. With the mad cow scare, you can see how busy I've been.

http://www.optibrand.com
 
Back
Top