Chrysler 41te Trans Control

[Bernard Fife]

Thad,

You are *way* ahead in electronic knowledge compared to where I was when I started with MegaSquirt (way back in 2001)!

I believe you need both the 5V and 12V for the tach signal. The 5V powers the 654 (or 555 on the V1), and the 12V is used as the signal source (to give it enough voltage to be sure to trigger). That is, the 5V is used by the 555/654 and also powers a transistor. This transistor is used to switch the 12V (nominal - sometime a 9V battery, etc.) supply from a battery or wall-wart to create the tach signal.

I am not familiar with the JimStim, but I suspect it would be fine to pull the 5V supply off of it (the tach circuit current requirement is low). Your wall-wart should work for both stims as well, I think (check the voltage with it all hooked up, if it doesn't pull the voltage too low, you are good to go testing!).

Lance.

[tjabo]

Thanks for the vote of confidence Lance!

And by the way, the board arrived today, so I guess I'll get working on that while I wait for the JimStim kit to arrive.

Thanks!

Thad

[Bernard Fife]

Thad,

Great, I'm glad it got there!

Lance.

[tjabo]

Good morning Lance, et al,

I've been delayed a few days by a variety of "life happens" type things, but I hope to be back on track soon!

In the mean time, RacerStev has been forging ahead, and has discovered a couple of things. First, and most important to this thread, is that he has obtained a frequency number for the solenoids of 1.96kHz. Hopefully that is in the expected range that can be managed. Also, I'm not sure how we use that number to create the peak and hold effect that is deemed to be important by the factory, but hopefully you will have an idea on this Lance.

Secondly, and not so directly important to this thread although I think it indicates the future interest in this trans/controller, Steve just took apart the stock one with which he was cranking out repeated low 10 second passes, and had this to say:

I took the trans apart that was in the race car today. VERY happy to report no damage at all, the shafts look fine,
no undue wear on the clutch packs. Looks just like a good stock trans!

This is a 2002 trans so it has all the good stuff inside. It's looking like a great race trans right off the shelf!

Steve

[Bernard Fife]

Thad,

Yes, 1.96kHz should be no problem (especially since we are using the processor's PWM channels rather than 'bit-banging' the pin). I am implementing the refresh cycles in the code right now - it is fairly easy and will be fully user configurable (ex. user can command the controller that if the PWM is active on a channel, set PWM to 100% for 8 milliseconds every 50 milliseconds - where all the numbers are user-settable).

Lance.

[racerstev]

This should be interesting!

Steve

[tjabo]

Just for a quick status update, I finished the JimStim last weekend and tested it quite a bit (nice piece of equipment!). I also got the heat sink made and mounted on the GPIO board, and began on the build of it. I expect to have a bit of time over the weekend, so I'm hoping to have it built toward the end of the weekend. I've got to remember to order a case for it today, I've been using a half of my MS case for the fitup stuff.

Thad

[Bernard Fife]

Thad,

Thanks for the update! I always used 1/2 a MS case too - even in the car - because it makes troubleshooting in the car easier. Definitely not a permanent solution though!

On the code front, I hope to get some work done and post new code, with most if not all of the new features included (things like the refresh cycles on the PWM, a second independent speedo output (for ABS), the input shaft speedo sensor code, and improved VSS 'reset to zero' code). That's assuming I things go smoothly, though!

Lance.

[tjabo]

That sounds awesome Lance!

I don't think this crew will be hurrying you up as far as getting those coding objectives achieved. . . I've been trying to help out on the beta testing of the MS2Extra code for stepper motors this week, and that plus other things probably means I'm a couple of weeks from being done with the GPIO board. Steve just got his turbo for the van, so he's probably a little bit from having it driveable (but he's freakyfast at getting his cars together, so you never know).

On the plus side, another Chrysler fwd ATX savant with a 9 second Neon, Carl Buchbinder, has expressed interest in testing the MegaShift. He and Steve both have their Peloquin diffs for their 41te's in hand, so they're close to assembling their transmissions and getting them rolling. My goal is to have this done and tested by the time one of them has a car ready to accept it!

Hopefully I'll have some positive results to report toward the end of the weekend.

Thad

[tjabo]

I made it to build step "E" without having to make any hard decisions on 41te specific items, but it looks like that's the end of easy street. . . I've got to do some more reading to see what this all means, but the build manual says:

E. GPI1, GPI2, GPI5 (2/4WD, swB, downshift):

These circuits are arranged in a left-right line just above the center of the board. They are digital input circuits, and ground to activate (no pull-up).

Note, if you are not using a GM style shift lever input (3 digital inputs on swA, swB and swC) you can use swB and swC for datalogging analog voltage signals such as temperature or EGT. To do that, you need to build the circuits differently. See http://www.msgpio.com/manuals/gpi.htm for more details on different ways to build the GPI2 circuit. You can also use the 2/4WD input for logging digital inputs (if you don't have a 4WD), if you build the circuit appropriately and select the appropriate option in MegaTune.

1. Install and solder a 1.0K Ohm, 1/8 Watt processor pin resistors {brown-black-red, 1.0KEBK-ND} in R27, R29, and R35.
2. Install and solder 330 Ohm, 1/8 Watt resistors in R21 and R25. These are the pull-up resistors for the downshift button circuit on GPI5 and the 2WD/4WD switch on GPI1.
3. Install and solder 5.6 Volt Zener Diodes in:
* R28 - banded end goes towards Ampseal connector,
* R30 - banded end goes away from Ampseal connector,
* R28 - banded end goes towards Ampseal connector.
4. If (and only if) you are using the digital inputs for the manual gear lever determination such as with a 4L60E (as opposed to a variable voltage, Ford-style potentiometer) then you must add pull-up voltage to GPI2 (and EGT3 and EGT4 in later steps). Do this by installing a 330 Ohm resistor in R22.
5. Install and solder 0.1µF capacitors {399-4329-ND} in C14 and C22.
6. Install and solder a 0.001µF capacitor {399-4144-ND} in C16.

I need help figuring out what we are going to use these circuits for, and thus how I should build them. . .

Steve, got any thoughts?

Thanks,

Thad