700R4 TCC Wiring

[pjmoreland]

This is the vacuum sensor I'm using:

NXP MPXV6115VC6U

https://www.digikey.com/en/products/...115VC6U/951851

[HO455]

Originally Posted by pjmoreland

I agree 45 mph would be a good cutoff. I am not sure if I'm going to be able to make the vacuum sensor work to my satisfaction. My impression is that you don't have to press the gas pedal down very far to make the vacuum level go almost completely away. I'd like to have the TCC stay locked maybe all the way to half throttle. Maybe a pedal-mounted switch from a 71-72 truck with a TH400 would be a better way to go. Another approach might be to tap into one of the pressure ports on the side of the 700R4 transmission to read the pressure associated with the TV cable position. I have thought it might work well to base the decision to lock or unlock the TCC on whether the gas pedal has been pressed down slowly vs. quickly.

On my 91 S-10 (700R4 behind a 4.3) the converter stays locked up when the throttle is pushed down slowly to maintain speed when going up a grade but unlocks immediately if I push down the throttle quickly. It seems that how guick the throttle moves is part of the equation.
Also the converter doesn't lock up below 45 mph on the S-10.

Thanks for the link to the sensor!

[Kenfuzed]

Is the High/Low vacuum switch solution still a good option, or did you find this to be unacceptable? I am in the process of putting a new engine and rewiring much of this. My truck is an 84 but since it was a California emissions model it had an ECM and most of the vacuum devices were already removed by the PO. I'm trying to set this up to use a vacuum trigger.

I also noticed your solution does not use the 4th gear oil pressure switch that so many earlier setups used.

[pjmoreland]

It works fairly well, but it has the one main flaw that it locks the TCC at too low of a speed and lugs the engine. The work around is to shift into 3rd when driving around town. I did use the 4th gear pressure switch, but it is mislabeled on my schematics as the 3-4 switch.

[Kenfuzed]

Okay, I didn't see a pressure switch in this last schematic. My 700 is an early version so not sure all of this will work without upgrading the switch at the connector.

[pjmoreland]

Originally Posted by Kenfuzed

Okay, I didn't see a pressure switch in this last schematic. My 700 is an early version so not sure all of this will work without upgrading the switch at the connector.

The 4th gear switch is only mentioned in the label for the B pin of the transmission connector. The assumption is that one of the switch terminals is connected to ground inside the transmission and the switch is normally open. That might not be the case in a stock '84 700R4, although it was for my '93 model. The following thread includes pages and pages of all the different ways these transmissions were wired internally.

https://67-72chevytrucks.com/vboard/...15#post8974715

[Kenfuzed]

Thanks for sharing

[HO455]

You mentioned using 3rd gear around town as an option. I my opinion it is something of a must do to insure the transmission has a long life.
In my experience 700r4's have a tendency to constantly shift in and out of OD when your driving in the 35 to 40 MPH range depending on throttle position. This continuous shifting causes heat to build up in transmission.
I don't shift mine into OD under 45 MPH and then only if I will stay above 45 for several miles.

[pjmoreland]

I hooked the Arduino controller up to my truck and spent some time today working on the code. After trying several different approaches, I was able to read the speed sensor reliably. I have to count the amount of time it takes for 8 pulses (one revolution) to be received. Trying to measure the time for one individual pulse was not consistent. I was also able to read the 4th gear switch, although the switch seems to be loose inside or something because it sends occasional brief signals when it is off, and especially when I drive over some bumps. I cleaned up that signal by reading the switch 10 times in a row and assuming it is off if any of the 10 readings say it's off. The signal is solid when it is on. I setup a simple speed-based control that locks the TCC in 4th and above 50mph, and then unlocks under 45mph. The logic works well and is consistent. I haven't worked on the vacuum sensor yet other than to verify I can read the signal.

I had a major setback on this project today though, unfortunately. Since the TCC was locking above 50mph without a vacuum switch to unlock it, I was able to give it more gas while locked than I have been able to before. That's when I discovered the TCC is slipping. I had been suspecting it was doing that previously, but I had assumed it was just the vacuum switch unlocking it. I dropped the transmission pan this afternoon and replaced the TCC solenoid, but it didn't make any difference. The old solenoid bench tested just fine. I guess this is when I find out if TCI honors their one year warranty on torque converters. I'm annoyed about having to drop my transmission.

[HO455]

UHHHGGGGGGGGGG!
Man I'm sorry to hear that.

On a different note congratulations on the other progress!

[pjmoreland]

Originally Posted by HO455

UHHHGGGGGGGGGG!
Man I'm sorry to hear that.

On a different note congratulations on the other progress!

Yeah, I'm not too happy about it, but that's life with these old trucks!

Here's my setup and the output from the software.

[pjmoreland]

Here's how I have it wired right now. If all goes well, I'll design a PCB, and I'll 3D print a box for it.

[pjmoreland]

I did some more testing of my Arduino code today and noticed that I'm getting a fair number of inaccurate speed readings. I found an Arduino library named FreqMeasure that is designed to work with low frequency pulse signals like I've got. Unfortunately, my Arduino Nano Every board has too new of a microcontroller to work with this library. Fortunately, I also have a regular Arduino Nano board, which is supported by the library, so I gave that a try. The FreqMeasure library works much better than the routine I had written. I had to move the speed sensor to digital input 8 to make FreqMeasure work.

[pjmoreland]

Dropped my transmission last night and shipped my torque converter off to TCI to have the slipping TCC checked out. The o-ring on the end of the input shaft looked good, and I don't see any junk inside the hole in the end of the shaft. The ball in there moves around freely.

[pjmoreland]

I started working on the Arduino PCB design today. I'm using a program named DipTrace. I got all of the components selected, modeled and placed in a schematic. The next step is to do the actual PCB layout (board outline, component positions, routing traces). Here's my schematic.

[HO455]

Very nice!
I may have to switch mph sensors. I have been trying to use the Gearvendor 2 wire signal generator but been having trouble with repeatability. Your PMW 3 wire method may solve my problem.

[pjmoreland]

What kind of signal does your Gearvendor sensor produce? Is it an AC sine wave? Have you tried using the Arduino FreqCount library with the sensor connected to input D5?

[HO455]

I have been going off the assumption that it is the same as this one. I wasn't able to get any information from Gearvendor to be sure.
I had loaded a frequency counter and had tried getting signals but it was all over the place so I came to the conclusion I needed a better test rig drive the generator. But the weather has been nice and too many outside side projects needed attention so I haven't gotten started on that yet. However it looks as if the rains have arrived so I should be able to get back to it soon.


https://www.summitracing.com/parts/d...SABEgLWCvD_BwE

[pjmoreland]

According to the documentation for that sensor, the output is a sine wave and the voltage (amplitude) increases as the speed increases. That might make it difficult to count with the Arduino. That being said, people have been successful at reading tach signals with an Arduino, and that voltage is all over the place. The nice thing about the sensor I used is that it outputs a square wave On-Off-On-Off signal.

https://www.dakotadigital.com/pdf/SEN-01-1.pdf

https://www.dakotadigital.com/pdf/SEN-01-4280.pdf

[pjmoreland]

I completed the Arduino TCC control PCB layout this evening. I still need to find some sockets to make the Arduino removable.

[HO455]

Very impressive sir! It looks like a production unit.

[Accelo]

It likely works better than the GM design. Of course, the electronics "back then" would have been expensive. I was around then the chief complaints, about GM transmissions, was the lock up timing. The pathetic power levels and early lock up, to keep mileage up, were additional issues that GM had to deal with. You may have a commercial product when you are finished. However, I do know many a transmission transplant that just ignores the lock-up converter. They just run them "old school." I suspect the marketing would be the difficult part.
Kudos to you for staying on this project.

[pjmoreland]

Originally Posted by Accelo

It likely works better than the GM design. Of course, they electronics "back then" would have been expensive. I was around when the chief complaints, one GM transmissions, was the lock up timing. The pathetic power levels and early lock up, to keep mileage up, were additional issues that GM had to deal with. You may have a commercial product when you are finished. However, I do know of many a transmission transplant that just ignores the lock-up converter and they just run them "old school." I suspect the marketing would be the difficult part.
Kudos to you for staying on this project.

I have never driven a vehicle that came stock with a 700R4. I wish I had that experience. It would be interesting to analyze the behavior, especially on later fuel-injected systems.

The speed based B&M TCC controller is, in my opinion, the best one on the market right now, and it costs $250. It doesn't have a vacuum sensor though, which is unfortunate. Based on the low quantity cost of the parts I'm buying for this Arduino-based design, I'm probably going to be really close to that $250 mark. At this point I'm feeling more inclined to share my final design with anyone who wants to build one for themselves, assuming all goes well. I chose through-hole-style electronic components that are easy to solder by hand with that in mind, despite the fact that it made the PCB layout more difficult to design.

I designed a box for the PCB over the weekend. It's pretty small at 2.5" x 2". I'm planning on mounting it to the bottom edge of the dash. A friend of mine is going to 3D print it for me. I ordered the PCB today, along with the components.

[Accelo]

It would be interesting to analyze the behavior, especially on later fuel-injected systems.
With the 700R4 transmission debuting in 82 and Fuel Injection 87 and full electronic (4l60E's) in 1993. It continued to get better as time went on. By the time it full electronic in 93 it was sorted out pretty well. The 2.8L Blazers were some of the worst. I purchased an 87 Blazer new. I was aware they were problematic so I purchased on with a 5 speed manual.
At that time I hadn't realized the 87 was completely upgraded. Or I would have gone for the automatic. You are basally reverse engineering what GM did in 93 on the converter. What would be cool is if you could get the GM code.

[pjmoreland]

Originally Posted by Accelo

It would be interesting to analyze the behavior, especially on later fuel-injected systems.
With the 700R4 transmission debuting in 82 and Fuel Injection 87 and full electronic (4l60E's) in 1993. It continued to get better as time went on. By the time it full electronic in 93 it was sorted out pretty well. The 2.8L Blazers were some of the worst. I purchased an 87 Blazer new. I was aware they were problematic so I purchased on with a 5 speed manual.
At that time I hadn't realized the 87 was completely upgraded. Or I would have gone for the automatic. You are basally reverse engineering what GM did in 93 on the converter. What would be cool is if you could get the GM code.

That would be interesting to see the GM code. I imagine it factored in the throttle position sensor data, which I don't have, unfortunately.