Breaking the code ... 4L80 or other late model trans -DIY TCM/TCU

I've been tinkering for a while on my transmission control module (TCM) for my car. I do realize there are after-market units out there and other DIY sites and products. But, I like to learn even the nuts and bolts of how things work or are put together. I do buy things to get done and out of they way (the American way of life it seems); however, there is some wisdom gained when doing activities like this with out buying the solution. There is a growing trend in the “Maker” movement for people re-learning trades and doing hands on activities like welding, modeling etc. This is a good thing in my opinion.


Regarding the TCM, my first iterations of building my TCM involved a MicroChip[SUP]tm [/SUP]'PIC”. MicroChip is an actual company. A PIC is a Peripheral (hardware) Interface Controller (PIC) which has been used since the early days and are some of the most prevalent chips out there powering your watches, alarm clocks to more sophisticated things. PIC chips are to some engineers like comparing a PC fat with software and increasing sluggishly over time to something purposed, dedicated, and reliable such as a calculator (driven by a PIC). Sometimes being at the trailing edge of technology is smarter than being at the leading edge of technology. I don't have to worry about a PIC catching a virus or being controlled by some one else. So, while I am fascinated by the solutions from old technology pre-dating silicon technology and processors, I like to do graceful implementations of old and newer while learning.


The purpose of this thread is to share information and knowledge, not necessarily the pros and cons about electronic versus mechanical things. I have had some time to actually refine my skills and implement a “bench” version of my TCM. This is the more advance version using the ATMEL Atmega Chip set through the use of Arduino. This is a shift for me since I was originally coding in assembly language. I am very satisfied with its performance thus far.


Here is the description of the current capabilities .

1. Input Tach pulse
2. Input MPH pulse
3. Input Upshift gear selection button
4. Input Downshift gear selection button
5. Input Wide Open Throttle
6. Output on LCD Tach in RPM
7. Output on LCD speed in MPH
8. Output gear selection
   OR
9. if WOT is detected begin shift at designated RPM until final gear is reached or WOT is no longer true.

Not currently in use.

1. Input for detecting idle
2. Input for detecting brake

Right now I am toying with the idea of implementing the “full” automatic capability and to do this I was thinking some something like the following logic. Your thoughts? :beers2:


If WOT false
If Manual shift false
Then upshift @ 1500 RPM


to further advance this I would implement logic for partial WOT basically passing gear down shift.
This is what I am thinking:


If partialWOT True
If RPM less than 2500
Then down shift by 1 gear
Else
If partialWOT false
Then return to normal Upshift.


Now I also realize I could use MPH for this logic but that would get tricky if there is tire spin at a launch. For example, my dad's original 70 Stage 1, would upshift from 1-2 @ ~45MPH (w/ 3.42 posi) 2-3 was ~80 mph. The opposite was true for WOT down shift at less than those speeds. fwiw if I used only MPH that is effectively what some traction control systems do, up shift, apply brakes and limit throttle input. Would need to resolve this – maybe because it depends if I lift a little at the starting line as the tire spins I am out of WOT logic. I am trying to avoid using a vacuum sensor or MAP but may need to.


Fortunately or unfortunately depending on your perspective, my car is in the air as I resolved the rear brake maintenance item and then began installing a customer serpentine belt system. So the car is not road worthy yet to try my next TCM iteration, I also told myself that the car will not go back on the road till this serp belt system is in place...I am tired of the pile of parts I have laying around for it. So while I get cold in the garage I come back to the TCM and work out the logic.


If you guys want a video of this current revision let me know I'll post it. :TU:

For those familiar with software coding in C++ or the olden days of assembly language you can still relate, here is my code.

 

Pretty interesting stuff. I'm no computer guy by any stretch. What caught my attention was the serpentine belt setup. Who's kit are you using or what components? I have some plates I bought and some of the stuff to put the mid nineties GM small block stuff on mine. I haven't gotten to it yet but am interested to know how yours works out. Any pix of the install?

Keep up the good work on the trans stuff. I'm happy with the 2004r in my regal and will probably go that route in My GS but this 4l80e tech is still very interesting.

 

With out a TPS or map sensor how are you determining vehicle load and controlling line pressure so you don't burn up the clutches?

 

Thanks.

The serp system is the very same as the the one you mentioned, the old 90's style GM system; However, because the Riviera has less hood clearance AND the waterpump is shorter I have had to customize the water pump pulley. The Aluminum plates that are used I need to recreate both (left and right brackets) the Alternator/Powersteering bracket and the AC bracket - I need to lower the whole assembly. I bought bought the "kit" here on V8buick.

Go to http://www.v8buick.com/showthread.php?55804-Serpine-belt-systems-for-455/page3 See post #61 Picture 5. I am not using a v-belt to drive the fan.

I trimmed down a GN water pump pulley, but my measurements were still slightly off for my liking.

 

I don't moderate line pressure in this version or the version in the car now. Each shift is firm @ full pressure. If I want a "softer" shift I ease up on the gas pedal and shift. I should modulate line pressure at some point, maybe do a little more research on this to PWM the pressure solenoid. :TU:

 

Does it bang when shifting from park to rev. or to drive? seems like that would be not good to have full pressure shift from park.

 

Not too hard, if on the right surface it will "cherp" or "bark" the tires slightly -Park to drive or Park to first. The engine high idle is usual where it does this most. From my reading you must be at or above 1200 RPM to shift from any other gear from 1st which assures appropriate line pressure. The whole performance of the shifts is very similar to my Dad's original '70 Stage 1 which is exactly what I was trying to mimic when I put the car together. Firm, crisp shifts. So far it has not caused any issues since I got it on the road several years ago.

 

Hi Chad,

I think this is a super cool project, keep us posted and yes make a video!

I still have the setup you built for me, for manual shifting a 4L80e with push buttons on the steering wheel. I will get it into use one of these days... Heck I thought about plugging it into the 4L60E in my truck to see if it would operate that.

 

Where are you picking up the RPM and the MPH signals from? Is the RPM directly from the neg side of the coil? Not sure about the MPH at all.

How hard would a TPS be to install on the throttle?

Looks like an interesting project.

 

the spedometer has the speed sensor, I think there is another that picks up a pulse from on of the drums as it spins around in the trany.

 

Sean, It will work but the gear sequence will be reversed 1st is 4th, 3rd is 2nd etc. I am not 100% on the pinout at the trans though the connectors are the same between the 4TXX and 4LXX transmissions.

Anson is correct. The 4L80e comes with input and output speed sensors (output shaft sensor is the speedo sensor in the 2wd configuration), temp sensor, gear solenoids, pressure control solenoid, and pressure switch manifold (for detecting shift lever location PRNDL321)

The TPS (which is a potentiometer) is not hard to implement it would be harder to install in the car, I was thinking a contact switch for idle, part WOT, and WOT. But mounting may be more difficult than the TPS. Im in the air in this point.

At the moment RPM will be sourced from the coil when I get to install in the car, the input shaft speed I am not too concerned with. Trans temp is used to PWM the torque convert clutch (TCC) but a firm continuous connection in light weight (not a dump truck) low temp application is OK. There are debates on why the early 4L80e had a ON/Off TCC solenoid versus the later version which is Pulse Width Modulation (PWM) - this is also discussed because the added cooling holes (94 and newer) in the input shaft compromise the input shaft strength. I "technically" have a '96 4L85e in my car. This was a transmission sold to me by the lead GM Turbo Hydromatic engineer several years ago in AZ. So long as I am not using this in a dump truck with some crazy load pulling a hill, I am with the crowd that says lock the TCC's PWM solenoid ON/OFF and not worry about actually PWM it. If I was worried I would swap the solenoid to the older part number if mine ever fails.

 

A TPS and a MAP aren't really that hard to add to a car. There are 2 sets on my 79 Eldo, so I can run
the original or a more modern EFI. Bruce Roe

 

Ok, Revision 10 is out. In this case I added or modified the following.

• Added the transmission temperature input. - it was actually in version 3.
• Added the pressure control solenoid (PCS) output which is PWM.
• Changed the WOT input to be analog so that a Throttle position sensor will need to be used.
• Added logic in the WOT and normal up shift routines. So, If WOT is detected, full line pressure is applied for each gear shift. Otherwise, a 60% duty cycle is applied to the PCS for each shift. In scenarios where Temp exceeds 200 degrees F, Full line pressure is then used for every gear all the time until a cooler temperature is detected.
• A few other misc things like pin re-arrangements that don't affect functionality.

Sean, I'll try to get a video posted up some time soon but videos with a clear explanation is taking me some time. Stay tuned...

For you math-letes, I realize the frequency for the PWM output of the chip set Im using is 490Hz and is not matched to the PCS 614Hz, but that I believe "could" only effect the responsiveness of the solenoid. I only did a cursory search to see what the power difference would or could be...no luck. So, I am not entirely sure this will matter much except dithering may occur. In which case I could do a WOT "pedal stomp" to clear the dithering effect if it does occur.


For those that know nothing about code, this should make you cross-eyed j/k, but nonetheless here it is:

 

Video Pending upload ...looking at this post time stamp plus 2 hours is when the video will complete upload to youtube.

[video]https://youtu.be/QrLO6A98ISg[/video]

<iframe width="640" height="360" src="https://www.youtube.com/embed/QrLO6A98ISg" frameborder="0" allowfullscreen></iframe>

 

Very cool.

 

This reminds me of when I was in Electrical Engineering school... And, Or, switches... Normally open, normally closed switches... I had to design a AC to DC converter from the PCB to the actual converter was all hand built... Fun stuff, and you take it another 1000 steps further with your peojects... Much respect, I will pay you for the product once its done.. Over my head.

 

Priced out the TPS and MAP from a turbo regal on Rock auto. Seems like these would be cheap and easy to install. I am thinking the MAP is still not required unless I wanted to implement better logic for down shifting when pulling a grade up hill. The TPS is clearly going to be needed for what I am trying to achieve.

I do have a question for you guys. Anyone measure vacuum for the transmission while under normal acceleration? What is the average expected in/HG under normal acceleration?

Thanks,
-Chad

 

That's going to depend on converter stall and cam. Mine pulls great vac until it locks up the converter, drops 400-500 RPM and the vac falls, kicking out the lockup, which raises the vac and locks up the converter ad naseum.

 

I'll just put logic in to leave the TCC on when speed is equal to or greater than 45, RPM is above 1200 and throttle change is in a measured range not greater that some value (meaning disengage TCC in heavy acceleration to WOT). Some of this will apply to "passing" gear selection. Need to give this some more thought when I don't have a migraine. Thanks Greg. Happy New year.:beer

 

Sounds like you will have a sweet program when you get it all sorted out. Being mine is a turbo 200-4R, I'm eyeing your project as something I can use to tune my lockup. Right now is just the vac switch that I finally tuned to the point it locks in 4th as soon as it shifts to avoid the in/out/in/out dance it did before. I think it would have worked a lot better with a mostly stock motor but a 2500 stall and TA-212 is just a bit more of a vac spread than it is happy with.