MT643 Downshift 3rd lockup to 2nd under load

[cadillackid]

I know I have one of those damn things in my garage.. I found out a little too late that a bad modulator would destroy an AT545.. so I replaced the modulator and then the tranny bit it a few months later so I kept the almost-new modulator and ditched the tranny

[Simplicity]

Originally Posted by cadillackid

I know I have one of those damn things in my garage.. I found out a little too late that a bad modulator would destroy an AT545.. so I replaced the modulator and then the tranny bit it a few months later so I kept the almost-new modulator and ditched the tranny

Do you want to sell it? I'm not sure what shipping would be, but if we could make it work in both our favors, that would be great.

[cadillackid]

I would be glad to just send it to you. Below is what I have if it matches yours. Re use your old O ring or use a standard one. I’m sure mine went bye bye with the 545.

PM me and you can make a shipping label online for UPS, email to me , I’ll pack it in a box and it will gladly go back to being gainfully employed vs taking up space in my drawer of allison parts.

[Simplicity]

This is my tranny fluid. When it was draining it looked more light brown. Obviously it's darker in the cup. I allowed some to coat the side of the cup to give a clearer idea of it's transparency when thin.

My friend who has a much more sensitive nose and experience with drivetrain fluids said it may have a very faint scent of burn to it. I put my face right into the bucket and I can't smell much of anything.

I rubbed the fluid between my fingers and there was no obvious grit. I held my finger into the light and do not see particles or shimmering.

I'm choosing Valvoline full syn ATF and am changing both filters. More info revealed after dropping the pan.

[Simplicity]

UPDATE:

I've changed the internal and spin on tranny filters, cleaned the governor screen, new gasket and filled the tranny with 16qt of Valvoline full synthetic ATF.

Here's a link to another post showing some pics of that fluid change.
https://www.skoolie.net/forums/f33/t...ind-35074.html

I've got my replacement modulator, tested it and it works.

I've changed my relay and tested it before installing it, yet I'm not getting any voltage out of the modulator connection.

Here's how we tested:
Key on, engine off.
Accelerator to the floor.
Multimeter on 12v with correct polarity.
Nothing??

Is this a viable test? I don't want to connect the new modulator to an electrical system that may have an issue and damage the modulator.

Any suggestions / input is appreciated.

[cadillackid]

On my navistar the original config is that when certain speed, pedal position and engine loading conditions the modulator would be activated to either trigger a downshift or increase line pressure. In reality this only worked sort of..

WW Williams in Columbus Ohio developed an add-on micro switch that was mounted to the physical pedal. When the pedal went over like 75% it triggered the trans relay on the firewall which activated the modulator.

A diode was fitted so that the ECM could also still call the modulator if it wanted but wouldn’t be back fed by the pedal switch.

The pedal switch is still on my red bus as it was outfitted with this “fix”. Of course it’s not in use but if you find in road testing that you still don’t modulate let me know and I’ll remove that assembly from the bus and send it to you if you want it.

My allison 1000 gets it’s modulation from the data link now

[Booyah45828]

I believe the ecm will activate the modulator without the engine running, so that should be a good test.

Is the relay clicking when you do this? If not, is the fuse blown to the relay? The ecm gives a ground to the relay, so if you have power to the relay, see if the ecm is giving a ground to it.

I don't think you can ruin the modulator in this scenario, it's just an electromagnet and is pretty robust.

[Simplicity]

According to the Electronic Control System Diagnostic manual, this is how it works (note, I've attached the related pages from the manual, so you can read those and check them against my interpretation of how it's supposed to work).

Summary:
ECM monitors Engine Data Line
Two Modes:
1) Closed Throttle Mode (moderate engine load of less than ~65% engine load), relay is energized;
2) Open Throttle Mode (heavy engine load of greater than ~80%), relay is de-energized

The transmission uses Engine Data Line (EDL) from ECM connector 379, terminal 32 for the appropriate shift schedule.

When ECM terminal 32 goes high, 12v, Open Throttle Mode is used.

Here's where it gets confusing. They call the relay connector female pins one set of numbers and the actual relay male pins another number...cute huh? So, I'm going to just use the relay male pin numbers to keep things hopefully more simple.

RELAY PINS
30 Common
86 coil
85 coil
87a normally closed with 30
87 normally open with 30

30 and 86 are normally energized with ignition power.

OPEN:
When ECM terminal 32 goes high (12v) the "relay does not energize".

Now the ignition power goes through circuit 928 (in the relay schematic), which is the normally closed 30 / 87a and energizes the modulator.

CLOSED:
When ECM 32 is LOW (0 volts) the relay is grounded which opens the normally closed 30 / 87a and de-energizes the modulator.

Flash Code 244 sets if relay connectors circuits (not the relay, the relay connector) finds an open or short. When I was connected to my ECM (while changing some parameters and before starting addressing the modulator issue) there were no codes showing, so I think those relay connector circuits are okay. Yet, I will reconnect to the ECM and double check.

The actual relay and modulator are not associated with the 244 code, so there are separate tests.

Again, the modulator gives a resistance of 5.x and clunks when energized with a 12v battery.

The relay is brand new and while tested before, I will retest for conductivity between the coil pins (85/86) and normally closed 30/87a.

There are also tests for the relay connector side of the circuits. I know I have 12v to 30 with KOEO. I need to recheck the other three tests.

Thanks for any assistance!

[mokibrabrant]

Originally Posted by Simplicity

According to the Electronic Control System Diagnostic manual, this is how it works (note, I've attached the related pages from the manual, so you can read those and check them against my interpretation of how it's supposed to work).


Summary:
ECM monitors Engine Data Line
Two Modes:
1) Closed Throttle Mode (moderate engine load of less than ~65% engine load), relay is energized;
2) Open Throttle Mode (heavy engine load of greater than ~80%), relay is de-energized

The transmission uses Engine Data Line (EDL) from ECM connector 379, terminal 32 for the appropriate shift schedule.

When ECM terminal 32 goes high, 12v, Open Throttle Mode is used.

Here's where it gets confusing. They call the relay connector female pins on set of numbers and the actual relay male pins another number...cute huh? So, I'm going to just use the relay male pin numbers to keep things hopefully more simple.

RELAY PINS
30 Common
86 coil
85 coil
87a normally closed with 30
87 normally open with 30

30 and 86 are normally energized with ignition power.

OPEN:
When ECM terminal 32 goes high (12v) the "relay does not energize".

Now the ignition power goes through circuit 928 (in the relay schematic), which is the normally closed 30 / 87a and energizes the modulator.

CLOSED:
When ECM 32 is LOW (0 volts) the relay is grounded which opens the normally closed 30 / 87a and de-energizes the modulator.

Flash Code 244 sets if relay connectors circuits (not the relay, the relay connector) finds an open or short. When I was connected to my ECM (while changing some parameters and before starting addressing the modulator issue) there were no codes showing, so I think those relay connector circuits are okay. Yet, I will reconnect to the ECM and double check.

The actual relay and modulator are not associated with the 244 code, so there are separate tests.

Again, the modulator gives a resistance of 5.x and clunks when energized with a 12v battery.

The relay is brand new and while tested before, I will retest for conductivity between the coil pins (85/86) and normally closed 30/87a.

There are also tests for the relay connector side of the circuits. I know I have 12v to 30 with KOEO. I need to recheck the other three tests.

Thanks for any assistance!

I have never seen or heard a more cogent reason or explanation for an all mechanical drivetrain. Oh. Well not true there was that one post I read where a rig was towed to three different Allison authorized repair facilities, all of which failed to locate the problem. Resulting in the owner some 2 months later removing the loom and inspecting each wire for deterioration and cracks. A crack in the wire insulation it was to the ECM was found to be the culprit. Something about "crosstalk" from the alternator to the ECM was disabling the bus. At that point I needed no further reason to ever risk going to an electronically controlled machine. I had experienced the wonders of "limp mode" and that's enough. Advocates love them until they don't. I have achieved don't. Once bitten twice shy.....good luck

[musigenesis]

Originally Posted by s2mikon

They also state 100 as the minimum.

Uh, does that mean you're not supposed to drive the bus until the transmission heats up to 100?

[Booyah45828]

Originally Posted by Simplicity

According to the Electronic Control System Diagnostic manual, this is how it works (note, I've attached the related pages from the manual, so you can read those and check them against my interpretation of how it's supposed to work).

Summary:
ECM monitors Engine Data Line
Two Modes:
1) Closed Throttle Mode (moderate engine load of less than ~65% engine load), relay is energized;
2) Open Throttle Mode (heavy engine load of greater than ~80%), relay is de-energized

The transmission uses Engine Data Line (EDL) from ECM connector 379, terminal 32 for the appropriate shift schedule.

When ECM terminal 32 goes high, 12v, Open Throttle Mode is used.

Here's where it gets confusing. They call the relay connector female pins one set of numbers and the actual relay male pins another number...cute huh? So, I'm going to just use the relay male pin numbers to keep things hopefully more simple.

RELAY PINS
30 Common
86 coil
85 coil
87a normally closed with 30
87 normally open with 30

30 and 86 are normally energized with ignition power.

OPEN:
When ECM terminal 32 goes high (12v) the "relay does not energize".

Now the ignition power goes through circuit 928 (in the relay schematic), which is the normally closed 30 / 87a and energizes the modulator.

CLOSED:
When ECM 32 is LOW (0 volts) the relay is grounded which opens the normally closed 30 / 87a and de-energizes the modulator.

Flash Code 244 sets if relay connectors circuits (not the relay, the relay connector) finds an open or short. When I was connected to my ECM (while changing some parameters and before starting addressing the modulator issue) there were no codes showing, so I think those relay connector circuits are okay. Yet, I will reconnect to the ECM and double check.

The actual relay and modulator are not associated with the 244 code, so there are separate tests.

Again, the modulator gives a resistance of 5.x and clunks when energized with a 12v battery.

The relay is brand new and while tested before, I will retest for conductivity between the coil pins (85/86) and normally closed 30/87a.

There are also tests for the relay connector side of the circuits. I know I have 12v to 30 with KOEO. I need to recheck the other three tests.

Thanks for any assistance!

That's backwards to what it should be in my experience. The modulator coil won't receive power until the ecm grounds the relay coil. The relay receives power at key on, but the ecm won't ground it until around 3/4 throttle. A lot are saying 65%, which might be standard, but I'm thinking that the % was an adjustable value, because I'm thinking we changed it on some of them.

What I can't remember is if the engine has to be running for the ecm to ground the relay coil. I don't think it does, But it's been a decade since I've messed with one.

Regardless, full throttle should provide power to the solenoid coil and kick the plunger out. If it doesn't, figure out why.

A member on the schoolbusfleet forum said that some body manufacturers have wired the relays wrong, never wired them at all, or didn't install the relays to begin with. I've never seen that, but I wouldn't walk into this assuming this was ever right from the factory. I have seen pins placed in the wrong spots on fuse panels before.

[cadillackid]

It uses engine load factor from the ECM not TPS.. the engine load Is a value calculated by the ECM based on the throttle position which on a Navistar the TPS sets a desired RPM.if the RPM is less than the desired then the load factor goes up. Until I believe load factor is like 18-20% at desired RPM. The fueling progressively gets heavier as your actual RPM goes below the desired. Thus creating the load factor .. somewhere around load factor 75-80% the ECM calls for a downshift (or higher line pressure). This activating the modulator.

The ECM won’t always call for the modulator of the RPMs are at or above what it perceives will be too high after the downshift.

The action of the modulator is also dependent on those seemingly useless settings in the ECM for gear ratio, tire revs per mile and final drive trans ratio.

When they key is on and engine off the load factor is always 0 even at wide throttle

[Booyah45828]

Originally Posted by cadillackid

It uses engine load factor from the ECM not TPS.. the engine load Is a value calculated by the ECM based on the throttle position which on a Navistar the TPS sets a desired RPM.if the RPM is less than the desired then the load factor goes up. Until I believe load factor is like 18-20% at desired RPM. The fueling progressively gets heavier as your actual RPM goes below the desired. Thus creating the load factor .. somewhere around load factor 75-80% the ECM calls for a downshift (or higher line pressure). This activating the modulator.

The ECM won’t always call for the modulator of the RPMs are at or above what it perceives will be too high after the downshift.

The action of the modulator is also dependent on those seemingly useless settings in the ECM for gear ratio, tire revs per mile and final drive trans ratio.

When they key is on and engine off the load factor is always 0 even at wide throttle

Hmmmmm, I thought it used tps value. If it uses load value, you'll have to have it running then, and even so, I don't think you'll get any significant load value idling.

I'm really starting to wonder how it works.

I remember testing them in the shop, and I don't think they were running when we did it. I could have swore we removed the thing, and had someone sit in the seat with KOEO, floored the accelerator, and you could watch the solenoid function. If it didn't kick out, you then hit the + terminal with a test light and repeated the test. If the light didn't illuminate, you checked the relay and fuse. If it did illuminate, you replaced the modulator solenoid, and then verified operation.

Maybe the buses had one of those aftermarket switches on the pedal activating the relay, vs relying upon the computer to do it.....

[cadillackid]

WW williams sold a ton of those switch kits because the ECM was erratic at best or had a lot of hunting. .



when I fixed my modulator I first had the switch on the pedal whiuch worked well.. then I tried disconnecting it and connecting the ECM just to make it factory.. and ha! it worked 20X worse so I hooked back up the pedal switch.. then my 545 blew up and now things are **MUCH** better

[johnbloem1974]

Originally Posted by cadillackid

then my 545 blew up and now things are **MUCH** better

Best thing that could have happened for me too! Although I am still jealous of your 2000 setup in RedByrd... but hey, I'll take the 643 any day!

John

[cadillackid]

the DEV bus got a 643 when its 545 popped also.. and its Much better as well, I really was not anti-545 until I blew up 2 of them and upgraded both to the Greener side of the street now im like forget-the-545 (except in my classic Superior.. i'll keep the AT540 in that one just because its iconic)

[Simplicity]

After isolating as much of the electrical as I could, I determined the following:

New to me electric modulator passes bench test. Thanks Christopher.
Modulator relay is good.
Modulator relay socket is good.
Pin 56 at ECM connector tones to the relay socket.
Relay socket to modulator connector tones.

I did not test pin 56 on the ECM itself, as I didn't want to push my luck and mess up the ECM. It seems pretty obvious that the ECM is not sending to the relay. Why? Who knows.

As I finally determined I needed to use the accelerator switch work around and am waiting for that to come, I really wanted to test the modulator.

So, I connected a + and - wire to the battery, added a toggle switch and a 15a blade fuse, got my friend to come over to activate the switch while I drove.

35mph, pedal to metal, toggle switch on, immediate downshift....rpms ran right up to about 2500rpm! Tested once more, same results.

Then, we went to a solid 6% grade, about 1.5 mile climb, 50mph speed limit. Going 50mph, started to loose speed, pedal to metal, toggle switch on, downshift once, pulled a bit but still to steep a grade, downshifted again, pulled the grade at 2350rpm.

Woo hoo!

Thanks to all, especially Christopher, for all the help.

[cadillackid]

try hooking it to the factory ECM setup.. it very well might work for you.. or you tried it asnd determined the ECM pin is never grounding the relay coil? or is the ECM calling but the relay not acting?

[JT20]

Originally Posted by Simplicity

After isolating as much of the electrical as I could, I determined the following:

New to me electric modulator passes bench test. Thanks Christopher.
Modulator relay is good.
Modulator relay socket is good.
Pin 56 at ECM connector tones to the relay socket.
Relay socket to modulator connector tones.

I did not test pin 56 on the ECM itself, as I didn't want to push my luck and mess up the ECM. It seems pretty obvious that the ECM is not sending to the relay. Why? Who knows.

As I finally determined I needed to use the accelerator switch work around and am waiting for that to come, I really wanted to test the modulator.

So, I connected a + and - wire to the battery, added a toggle switch and a 15a blade fuse, got my friend to come over to activate the switch while I drove.

35mph, pedal to metal, toggle switch on, immediate downshift....rpms ran right up to about 2500rpm! Tested once more, same results.

Then, we went to a solid 6% grade, about 1.5 mile climb, 50mph speed limit. Going 50mph, started to loose speed, pedal to metal, toggle switch on, downshift once, pulled a bit but still to steep a grade, downshifted again, pulled the grade at 2350rpm.

Woo hoo!

Thanks to all, especially Christopher, for all the help.

Sweet I’m glad you got it working!
So you have it working like it should on a toggle switch?
I’m trying to understand how and what went wrong.

What do you have to do to make it work without manually switching a switch?

Sounds like you are making progress figuring it out.

[Simplicity]

Originally Posted by JT20

Sweet I’m glad you got it working!
So you have it working like it should on a toggle switch?
I’m trying to understand how and what went wrong.

What do you have to do to make it work without manually switching a switch?

Sounds like you are making progress figuring it out.

Read this:

https://www.skoolie.net/forums/f33/a...ory-35465.html