So just to make sure I have this correct…. The ecm tos is accepting the same square Hall effect signal that the t87 uses as the output shaft speed sensor input in 5v form? So if this is the case then we can change the turbine sensor to the new one and leave the original vss alone and use a vr to hall converter to feed the t87 and the e35b tos input. I am working on an adapter to use the new bolted vss sensor in place of the old threaded sensor but that is a 9v, do you think we could tap that and simply use a resistor in tos line to keep voltage down?Ok kids I just ran out of Caffeine, but solved it. And luck have it, with results in our LMM/LBZ favor...
I reset the speed limit in the LMM ECM to 80km/h. Which should reflect as fuel cut at a much lower output shaft rpm. Tested it, and now cuts the fuel around 2000rpm. Which verifies the ECM speed limit tables are actually using the VSS raw pulse data supplied by TCM over GMLAN.
View attachment 120874
Next I wanted to test the VSS High signal wire to the J2 45 pin on the ECM. So I connected it to our Arduino emulator that supplying the 5v pulse to our TCM inputs. I left the TCM off, and the ECM picked up the 40 pulse correctly as a 5v square wave.. GOOD SIGN..
So with our engine spinning 1250rpm, and fuel injectors hammering away in the background, I spun the output shaft speed up to what read 200km/h. And guess what, no fuel cut via VSS single wire only.. It's not interested in the pin 45 signal.. Also a good sign.
View attachment 120875
Last I connected powered up the TCM, and the emulator is now feeding 40 pulse per rev signal to the TCM turbine speed line, output speed line, and the TOS / VSS High pin on our LMM ECM. Some freaking way, everyone is happy receiving a emulated 5v pulse, AND the CANbus and TOS speeds all match up in the data.. Super news...
So I cranked the speedo POTs up, and soon as output shaft RPM hit 2000, the ECM VSS singal crossed 80kmh, and the ECM cut the fuel. The lack of pecking in my ears confirmed the data below.
View attachment 120876
Everybody happy with the same 40 pulse per rev 5v Square wave. So we can easily Tee off our Hall converted T87 output sensor signal line, and fill that TOS hole in the old connector @Cougar281..
It's gonna be an octopus looking adapter cable.So what do the last 20-or-so posts mean to us average people?
That the T87 will directly swap back to an LBZ and LMM?
Or do you have to make an "adapter" cable that will cross-connect the VSS signal to multiple pins in order to make it work that far back?
@Cougar281So something I noticed on the pinout chart I made to figure out what pin on the A40/A50/T14 needs to go to what pin on the T87 that may be a problem or may not - I think what is the line pressure control solenoid on pin 2 of the T87, which the LMLs got but the LBZ/LMM didn't have, I listed as going to pin 74 on the A40/A50, which is the 'mainline pressure solenoid control', which I'm 98% certain is the 'G' solenoid, which we don't actually need. In the calibration for the '15 LML, there are D5179-D5182, which are for 'Mod Main Pressure (G-Sol)'... Since we don't actually need the G Solenoid, and don't have the mainline pressure solenoid, I wonder if connecting it that way and setting the values to that which will prevent it from ever being allowed to be enabled?
@Cougar281 I don’t think that will be much of an issue. The 07-10 Allison still has the PCS1 mainline solenoid. But they changed its design for 2011. Before it was more of an on and off and in 2011 they changed the valving and used the other style solenoid so it could use pwm to control pressure rather than to feed the spring regulated circuit. Hence why you can take an A40 from a 2008 and stick it in a 2013 and it will work just fine but it will just keep line pressure at normal/max vs the A50 allows pressure to drop when not needed to gain fuel mileage and allow the trans to run cooler by not building as much hydraulic heat from the higher pressure. So all we have to do for a 07-10 truck should be to just max the pressure which should max the pwm and it should work as intended.So something I noticed on the pinout chart I made to figure out what pin on the A40/A50/T14 needs to go to what pin on the T87 that may be a problem or may not - I think what is the line pressure control solenoid on pin 2 of the T87, which the LMLs got but the LBZ/LMM didn't have, I listed as going to pin 74 on the A40/A50, which is the 'mainline pressure solenoid control', which I'm 98% certain is the 'G' solenoid, which we don't actually need. In the calibration for the '15 LML, there are D5179-D5182, which are for 'Mod Main Pressure (G-Sol)'... Since we don't actually need the G Solenoid, and don't have the mainline pressure solenoid, I wonder if connecting it that way and setting the values to that which will prevent it from ever being allowed to be enabled?
I'm located in southeast Ohio, so I'm pretty close to ya. @pl_silverado said he was gonna send me a new style sensor, but I can grab one at the local parts store also if need.@kidturbo where are at anyways? I can send you that new vss and pigtail if you want to do some testing with it. I tested it on my t87 and directly wired to the tcm it reads speed as expected. But the battery in my 440 scope crapped out so testing is a little limited at the moment til the new battery pack gets here.
That is true but seeing where they are heading with the a40/50/t14 modules becoming obsolete and already in short supply and heavily over priced used and even worse new going for as much as $3000 even though they are $700 through gm if you can ever get one. I ordered a replacement a50/t14 back in June and hasnt shown up yet. Plus I’ve already seen a hand full of replacements fail after 18 months. if we can get these t87 modules to work properly then there will be no reason to go back and may not have the option to ever go back anyways as we are finding no replacement modules for earlier duramax trucks for pcm, gpcm, abs, etc. I did see the 2 channel vr to hall converters but then we also have to get a 5v source to power them. Though there are plenty of 5v references unused on the ecm that we could use to power the converter. But if I get this vss adapter right then just changing the 2 sensors “one of which simply unbolts and pulls out and new one goes in, the other threads out and will need an adapter” and plugging in the wiring adapter for each would eliminate the possibly of having a converter go bad 1000 miles from home and nowhere to get one same day. But the be converter is definitely more on track with K.I.S.S. for a plug in adapter harness not requiring getting under the truck.I'm located in southeast Ohio, so I'm pretty close to ya. @pl_silverado said he was gonna send me a new style sensor, but I can grab one at the local parts store also if need.
IMO, the easy fix is just stick one of these little dual line VR to Hall converters in the harness and leave all the factory speed sensors alone. Then, if any reason ya wish to swap back to an A40/50, it's simply unplug the T87 adapter cable, connect the old style TCM, and done.
Sent from my SM-S901U using Tapatalk
Somehow I missed this post yesterday. That's kind of what I was hoping.@Cougar281 I don’t think that will be much of an issue. The 07-10 Allison still has the PCS1 mainline solenoid. But they changed its design for 2011. Before it was more of an on and off and in 2011 they changed the valving and used the other style solenoid so it could use pwm to control pressure rather than to feed the spring regulated circuit. Hence why you can take an A40 from a 2008 and stick it in a 2013 and it will work just fine but it will just keep line pressure at normal/max vs the A50 allows pressure to drop when not needed to gain fuel mileage and allow the trans to run cooler by not building as much hydraulic heat from the higher pressure. So all we have to do for a 07-10 truck should be to just max the pressure which should max the pwm and it should work as intended.
It is looking for a specific resistance but it will still read even if resistance is high or low. We can disable resistance checking easy enough. Using a converter the only thing we would want active would be circuit performance. That would just be in case the converter ever started messing up.Isn't the TCM expecting a certain resistance across the sensor? If so, how will a converter and/or pulling power from the 9v reference affect that? Wouldn't that annoy the TCM?
According to the signal data from the bin posted a couple pages back, and what I've noted so far in bench testing, the 87 TCM looks for a 0-5v range rather than a resistance value on the ADC pins. For the 9V Ref lines, these VR converter circuits don't actually draw any power to operate. They just switch a supplied 5v On/Off.Isn't the TCM expecting a certain resistance across the sensor? If so, how will a converter and/or pulling power from the 9v reference affect that? Wouldn't that annoy the TCM?
You are right. I just looked through testing and diag and the t87 protocol is testing against voltage vs testing resistance. Honestly didn’t realize they changed. Not very often you deal with a short to power or short to ground situation, usually it’s a sensor performance or broken wiring causing intermittent signal.According to the signal data from the bin posted a couple pages back, and what I've noted so far in bench testing, the 87 TCM looks for a 0-5v range rather than a resistance value on the ADC pins. For the 9V Ref lines, these VR converter circuits don't actually draw any power to operate. They just switch a supplied 5v On/Off.
Without going back and rereading a lot, are all the canbus messages to and from the tcm needing converted or just certain ones? I starting to think I just need to go from a drive that way and let you plug into my truck and do some testing. Need to run to Logan to Bazzel and get some race fuel one of these days anyways.Two ways this could be wired up. Since we need a CANbus chip and processor manipulating some data packets, that same board can convert 12v to 5v an 3.3v, which we share with our 5v VR converter. Second way, we design or use an off the shelf PCB circuit that works with our existing 12v or 9v supply, and keeps the VR converter separate from the CANbus. Advantage here is, if you lose the CANbus board for some reason, the TCM will likely still shift into gear so long as it gets a valid speed signals. Probably be in limp due to the U0100, but would likely still roll ya down the road. We would just want to incorporate a circuitry on that TOS line, so if it shorts out due to another node, it doesn't knock down our speed sensors. Or we just replicate it using the VSS as our tigger..
If anyone has an A40/50 truck and a good scope, please check that TOS line voltage and wave pattern. When testing ECM pins VSS Hi Lo on the bench, I noticed the VSS Lo goes to Batt Voltage when ECM boots. Then it drops off to a few mv within a couple seconds. That same line is typically unused in out automatic trucks. And while I have custom copy of what's supose to be the non-existing LMM standard transmission OS, I've never been able to get it working with just the VSS line as our only speed input.
The CANbus messages "Required" by the 87 TCM to keep it running happy vary by OS. What's required by the 16 gasser version, doesn't line up exactly with the 16 Allison version. But mostly it's messages missing in these earlier models, which was added by GM to the later platforms. Those are fairly easy to fix. However we've found a few messages that GM added bytes to the same ID. So what was 2 bytes long on a LMM truck, might be 3 or 4 bytes in length on the L5P models. And some of those are from the BCM and EBCM. So while I've been able to sort and verify all traffic related to ECM, have no set plan for the BCM and EBCM stuff yet. If we have to run a gateway between the T87 and the rest of network it becomes more complicated. Hence the reason the LML swap is fairly easy, and nobody has tackled the LBZ/LMM yet.Without going back and rereading a lot, are all the canbus messages to and from the tcm needing converted or just certain ones? I starting to think I just need to go from a drive that way and let you plug into my truck and do some testing. Need to run to Logan to Bazzel and get some race fuel one of these days anyways.
Amazon. $13.77. Now, it's not gm, but cheap enough for experimentation.Here ya go @Baddiode. This early model VSS is what I was thinking "might" be of some use for anyone wishing to convert to Hall sensors. If could source this threaded steel housing part, finding a new style hall sensor that fits the hole might be easier than printing a custom adapter.. Made in Taiwan it says..
GM 15547452