T87 Swaps

[kidturbo]

Discovered a slightly BIG freaking issue with sticking this T87 into the 2005-2010 truck chassis.

While we can make the TCM speak happily with the E35A/B ECM by tagging on a few missing packets of required data, that can be assembled from existing CANbus data, the TCM really doesn't like any of the other nodes on our CANbus network. GM added a extra byte or two of data to almost every packet sent by our BCM and EBCM after model year 2010. And our T87/A TCM's are coded to accept only certain packet lengths per each ID. The same issue I ran into with the E38 gasser swap. And was able to fix with a simple CAX file mod to the ECM OS, adding 1 Byte onto 1 ID. In this case, we need to modify about 5 ID's in 2 differnt devices..

On the favorable outlook side, for stand alone ECM/TCM swaps and probably all LBZ trucks, I think we can make this little swap work just fine without much trouble.. Will either need to disable the U DTC's in the TCM, or feed it enough garbage to satisfy that the missing nodes are alive. Either way, I don't see a problem making the ECM/TCM run without any BCM/EBCM data. But I now see the reason why nobody with fancy hardware has been pushing an LMM T87A option. The BCM and EBCM of that year are just not very compatible with the newer version of GMLAN.

That doesn't mean one couldn't update the BCM and EBCM modules to fix this issue also. Several options to try, starting with just disabling the DTC's in the TCM and see how it responds. But any tables related to cruise control aren't gonna be happy.. It also expects to see 4 wheel speed sensors, rather the the 3 wheel speeds in a LMM. There are also some rolling counters and brake pedal pressure paramaters missing from our LMM chassis data that is going to take a bit of reading, and hacking thru the TCM bin files see if they hold any importance. For now, it's 50/50 on the LMM's..

 

[kidturbo]

Also making progress on the 80 pin connector. Outer shell is fitted. Took a few attempts to get the fit I'm told. But I have a young and eager Engineer student on it. With a bag of assorted pin terminals, some good software, and a couple 3D printers. He's already surpassed my skillset, so...

 

[pl_silverado]

2009-2010 LMM’s did have 4 wheel speed sensors. The TCM I sent you was off of an 09. 4 channel EBCM

 

[pl_silverado]

Likely going to have some sensor issues here also, 2wd Allison applications would likely be fine but the transfer case housings are different between the two generations.

 

[Cougar281]

Welll..... Maybe, maybe not... Part number 19302667 lists applications from 1999 all the way up to 2020 (there are actually some applications as far back as 1990 ), and it seems to apply to the Diesel up to 2016. For 2017 and up, Diesel is no longer in the 'applies to' list . I don't see a fitment list for part 24276403, but 24276402 seems to apply from 2015 through 2019, and almost exclusively diesel. So maybe if we are using a 2016 T87 cal, the input needs to be changed but not the output? I should add that I believe 2020 is where the 10L1000 Allison-not-Allison was introduced, which I'm sure used the T93, not the T87A.

 

[pl_silverado]

Yea I beleive the same output is still used on the gassers. The diesel is the only one that’s changed, but loooking at a 01-10 2wd Allison the output would fit right in the tail shaft extension housing.

 

[kidturbo]

Gonna leave this one up to the knowledgeable transmission builders on here.. From what I have read, and seen with a volt meter, both of these new speeds sensors are a 9V Hall Effect. The pins on TCM output 9V, and I'm not sure what the return signal actually reads yet. If someone has a good scope and a 16 or newer truck, would be quite handy to have a pattern capture.. I would expect a 9v square wave, but know it will accept a 5v square wave from my Arduino emulator.

Plan is to just convert this wave pattern on same board that will handle the CANbus conversion work. Bur for now, I'd like to make up a couple circuits on a breadboard for theory testing.

 

[kidturbo]

Likely going to have some sensor issues here also, 2wd Allison applications would likely be fine but the transfer case housings are different between the two generations.

Did a little digging this evening on what our T87 expects to see for Input and Output speed signals. First off, I also verified the # 19302667 is used on gassers to at least 2020, by just looking up the speed sensor for a 2019 Tahoe. Everything comes back as 19302667... So that's interesting, and makes me want to swap back to the gasser OS and test the signals. See no reason the hardware couldn't handle both type of sensors, possibly on different pins.

Next I did some digging through the Allison T87A bin, and found the stated Min/Max signal values for the input and output speed sensors. Both lines expect to see between .250v - 4.70v at all time. Outside those limits, sets a sensor fault code. The supply to each line is 9v according to what I read on the pins, and found in documentations. So we can pretty much assume it's looking for for a 0-5v square wave on both lines, for the Allison OS at least.

I hope to test all this further using the emulator over the next week. Still think the VR to Hall converter approach is gonna be the easiest. But for now, think we should nail down the who's who of output shaft values.

 

[Baddiode]

Did a little digging this evening on what our T87 expects to see for Input and Output speed signals. First off, I also verified the # 19302667 is used on gassers to at least 2020, by just looking up the speed sensor for a 2019 Tahoe. Everything comes back as 19302667... So that's interesting, and makes me want to swap back to the gasser OS and test the signals. See no reason the hardware couldn't handle both type of sensors, possibly on different pins.

Next I did some digging through the Allison T87A bin, and found the stated Min/Max signal values for the input and output speed sensors. Both lines expect to see between .250v - 4.70v at all time. Outside those limits, sets a sensor fault code. The supply to each line is 9v according to what I read on the pins, and found in documentations. So we can pretty much assume it's looking for for a 0-5v square wave on both lines, for the Allison OS at least.

I hope to test all this further using the emulator over the next week. Still think the VR to Hall converter approach is gonna be the easiest. But for now, think we should nail down the who's who of output shaft values.

I have schematics for the t87 pinout as well as all other transmission control schematics for both 2019 tahoe and 2019 duramax l5p. They move the input speed wire from 12 to 15 between the 2 as well as a few other differences that could cause some of the speed sense issue. Also I read your mention of the tap shift and tow haul on the 07.5-10 lmm trucks, those are to the bcm then commanded from the bcm to the tcm. The 09-10 trucks use the same a50 tcm as the 11-15 lml trucks and have 4 wheel sensors as long as the truck has traction control and stabilitrac.

 

[kidturbo]

That's some great info. Thanks for sharing

I was gonna start looking for a later LMM and early LML truck to test drive. If we can get the ABS happy, the BCM stuff shouldn't be a huge deal either. So far, I've been going off 08 LMM recordings, 17 l5P, and a 16 LML ecm stand alone.

The way GM did these OS changes, nothing is very backwards compatible past certain model releases. I'll be looking at those other pins differences. Some gassers had 3 sensors I know. Would be sweet if could just swap em in the code.



Sent from my SM-S901U using Tapatalk

 

[Baddiode]

That's some great info. Thanks for sharing

I was gonna start looking for a later LMM and early LML truck to test drive. If we can get the ABS happy, the BCM stuff shouldn't be a huge deal either. So far, I've been going off 08 LMM recordings, 17 l5P, and a 16 LML ecm stand alone.

The way GM did these OS changes, nothing is very backwards compatible past certain model releases. I'll be looking at those other pins differences. Some gassers had 3 sensors I know. Would be sweet if could just swap em in the code.



Sent from my SM-S901U using Tapatalk

I've noticed doing swaps that the backwards compatibilities aren't there and finding compatible OSs and calibrations can be a challenge. I know the bcm calibrations and os are all compatable from 2007.5-14 lmm and lml doesnt seem to matter they all speak to the ecm and a40-a50-t14 tcms. As for the abs/EBCM the 09-10 trucks with trac and stabilitrac have a different module from the 11-14 trucks but do have 4 wheel control and internal pressure sensor that relays data for the TCM and TBCM/trailer brake module, so the data should be there. My truck is a 2010 GMC 2500HD with trac and stabilitrac and currently i have the a40 tcm as my a50 died like the rest. If there is any data you want to see, I can see what I can capture for you.

 

[kidturbo]

If you have any ways to capture raw canbus data, csv format is good. Would like to compare the 2010 ebcm and bcm data to others.

Just now happened to find what looks to be a 2010 bcm in a box of parts. From what, I have no idea. But the more we know about each Hardware and OS, the easier it is to map out who speaks to what..

It's all the same language fromn 2004 onward. But sorta like Queens English vs American English. Depends on where you're standing, if everyone else sounds weird or not.



Sent from my SM-S901U using Tapatalk

 

[Baddiode]

If you have any ways to capture raw canbus data, csv format is good. Would like to compare the 2010 ebcm and bcm data to others.

Just now happened to find what looks to be a 2010 bcm in a box of parts. From what, I have no idea. But the more we know about each Hardware and OS, the easier it is to map out who speaks to what..

It's all the same language fromn 2004 onward. But sorta like Queens English vs American English. Depends on where you're standing, if everyone else sounds weird or not.



Sent from my SM-S901U using Tapatalk

I actually have that same BCM sitting on my bench right now that has a blown out diode for the 12v output used for the serial enable. Had a short in the engine harness and guy decided to use a higher amp fuse in place of the 3 amp that kept blowing lol. The one I have is programmed for a 2008 but that same p/n can be used 07-14 with trucks that dont have door modules and that calibration is used 07-13 so that should give you the data you want as long as it is still functional. I havent recorded raw canbus in a long time. Have to figure out what software i was using with my MDI. Any recommendations?

 

[Cougar281]

Did a little digging this evening on what our T87 expects to see for Input and Output speed signals. First off, I also verified the # 19302667 is used on gassers to at least 2020, by just looking up the speed sensor for a 2019 Tahoe. Everything comes back as 19302667... So that's interesting, and makes me want to swap back to the gasser OS and test the signals. See no reason the hardware couldn't handle both type of sensors, possibly on different pins.

Next I did some digging through the Allison T87A bin, and found the stated Min/Max signal values for the input and output speed sensors. Both lines expect to see between .250v - 4.70v at all time. Outside those limits, sets a sensor fault code. The supply to each line is 9v according to what I read on the pins, and found in documentations. So we can pretty much assume it's looking for for a 0-5v square wave on both lines, for the Allison OS at least.

I hope to test all this further using the emulator over the next week. Still think the VR to Hall converter approach is gonna be the easiest. But for now, think we should nail down the who's who of output shaft values.

I really don't know what to tell you on the OSS... I grabbed the VIN of a 2016 Diesel Silverado off fleabay, punched into gmpartsstore.com, which validated the VIN and allowed me to search, and according to GM, that sensor is the proper outsput speed sender for a Duramax/Allison equipped 2016 Silverado... As it has been for a LOOOONG time. GM Parts lookups say that the same sensor that applies to the 2016 (but not the 2017 and later) T87 equipped Duramax/Allison applies to my 04... If we were looking at a 2015, then there is that breakpoint, so if they were different, you could get either, but with 2016, there were no T14 equipped trucks, only T87, so GM saying this sensor applies to the 2016 T87 equipped Allisons, as wall as the T87 equipped gassers.... This one is a mystery.

 

[kidturbo]

Only one way to solve this. Think we need a spinning reluctor wheel work bench setup.

I grabbed a front wheel hub off an F350 my buddy changed out last week. Only to git it home and discover it doesn't spin freely.. Imagine that. Back to the drawing board..

Need another 40 tooth wheel like I have on the boat driveshafts. That worked flawlessly using that same old output shaft sensor. And on the bench, we can test different sensors to the same setup.. Ok great, now who's gonna build one?? LOL

Believe we will ultimately find that the rear sensors are the same up until 2017, when they switched to the T87A release. The code on the T87 and T87A is just slightly different. And the 2015-16 likely works fine with the VR style sensors, as does the gassers.. Just a setting in the OS.

Here is why.

The a2l data we have been basing these hardware changes off of for the 2017 T87A, specifically states both sensors send back a 0-5v square wave pattern on specific pins. What we don't have such exact specs on the gasser, or the 2015 T87 OS. If so, we'd know what pins accept what wave type and voltage levels. So it's entirely possible this early Allison OS runs a square wave input, and a sine wave output shaft speed just fine. Meaning just 1 sensor to change..

According to this 87A bin file header, we are on "Update for HWIO v2.13" since 2013. I'm counting 10 plus Hardware IO releases from start. So to have changed the pin from a Sine to Square wave, would just be a few linses of code in an official release.. I suggest we test some sensors on the bench or in a truck. There is likely a few release vesions of this OS we just haven't found yet.

 

[kidturbo]

Ok, hooked up my latest Arduino board with the same 5v Crank/Cam emulator software used for running ECM's. Modified it to have a single 40 pulse wheel, and connected it to the T87 Input and Output speed sensor lines. With the Allison OS, it tracks RPM perfectly. And I quickly noticed the Input reluctor is not a 40 tooth. And best I found posted is 64 Teeth on the PTO style drums.. So if anyone knows the exact tooth count on that drum reluctor, please share for science.

Anyways, I was able to verify the T87 works with 0-5v square wave pulses on the Pins listed in our 16 Allison diagrams. What it does with a sine wave wired in there, I am planning to try and emulate that next...

Another couple points to note, the PID for Transmission Input Speed is actually Engine RPM as received from CANbus. Turbine Speed is our actual Input speed sensor value in EFIlive. Transmission Output Speed is our actual 40 tooth reluctor in rear housing or transfer case.

 

[1FastBrick]

@kidturbo what was your plan with the wheel hub?

 

[kidturbo]

@kidturbo what was your plan with the wheel hub?

This is what I have to work with..

What ya got?

Sent from my SM-S901U using Tapatalk

 

[kidturbo]

@kidturbo what was your plan with the wheel hub?

Sorry, I missed your actual question.

Use the wheel hub and pickup to generate a live sine wave signal. But unless I mount a wheel on it, ya not gonna spin very fast by hand or drill..

Sent from my SM-S901U using Tapatalk

 

[1FastBrick]

I have some GM hubs is why I asked. But I could also have a small reluctor made if that's what you need.