E67 to T87/T87A TCM interface needed
Re: E67 to T87/T87A TCM interface needed
Simply because its connected to the CAN line. It wouldn't be a bad idea to add a relay which can be switched on and off.
This would allow either external software to control it for whatever reason, or even put into custom code on the engine or transmission computer side.
This would allow either external software to control it for whatever reason, or even put into custom code on the engine or transmission computer side.
Your Local Aussie Reverse Engineer
Contact for Software/Hardware development and Reverse Engineering
Site:https://www.envyouscustoms.com
Mob:+61406 140 726

Contact for Software/Hardware development and Reverse Engineering
Site:https://www.envyouscustoms.com
Mob:+61406 140 726

Re: E67 to T87/T87A TCM interface needed
Yeah anything attached to GMLAN needs to adhere to a specific set of proper protocol rules, or you'll fail flashing, and possibly brick whatever. Has been an big issue over the years with aftermarket head units.
Over the years I've logged a bunch of manufacture and aftermarket CANbus tools. And for GM at least, whenever ya enter into read or write mode, the first thing an MDI, HPT, EFI or any other turner typically does, is broadcasts a "shut up until told to speak again" message. If flashing in a car, response is typically denoted by every node going into some sort of sleep mode. Then once read/write is complete, the tool sends a Reset / All Clear message to notify all the nodes it's done taking priority status on the bus. However, unless someone added those specific message commands into their CANbus code on a new device, IE an emulator or a full gateway, ya just created another cheap aftermarket head unit bug that's gonna bite someone who least expects it..
I've noticed on the bench that my new HPT i3 is most picky about spotting unknown devices on the bus. It complains about my CANbus tools running in Normal mode and not Silent, as where EFIlive doesn't have a problem with sniffers running while reading or loading. On my Marine gateways we use a DPID start up script to grab a couple extra parameters not in the normal broadcast data, and mode 22 it into a continuous PID stream, followed by a 0x101 Test Present every few second. HPT is good with it for normal scanning, but EFI V3 won't even detect an ECM now if it sees another 0x101 ID on the network...
So yes an easy way to shut off anything additional to the bus is always tops on my list. And a big reason for perusing this setup without requiring a full Gateway involved. While I spend my days building CANbus gateways for everything imaginable, I'm still not confident enough to say "sure program that box thru my gateway.." It's just a dumb risk, and someone out there is sure to find that bug you never considered...

Over the years I've logged a bunch of manufacture and aftermarket CANbus tools. And for GM at least, whenever ya enter into read or write mode, the first thing an MDI, HPT, EFI or any other turner typically does, is broadcasts a "shut up until told to speak again" message. If flashing in a car, response is typically denoted by every node going into some sort of sleep mode. Then once read/write is complete, the tool sends a Reset / All Clear message to notify all the nodes it's done taking priority status on the bus. However, unless someone added those specific message commands into their CANbus code on a new device, IE an emulator or a full gateway, ya just created another cheap aftermarket head unit bug that's gonna bite someone who least expects it..
I've noticed on the bench that my new HPT i3 is most picky about spotting unknown devices on the bus. It complains about my CANbus tools running in Normal mode and not Silent, as where EFIlive doesn't have a problem with sniffers running while reading or loading. On my Marine gateways we use a DPID start up script to grab a couple extra parameters not in the normal broadcast data, and mode 22 it into a continuous PID stream, followed by a 0x101 Test Present every few second. HPT is good with it for normal scanning, but EFI V3 won't even detect an ECM now if it sees another 0x101 ID on the network...
So yes an easy way to shut off anything additional to the bus is always tops on my list. And a big reason for perusing this setup without requiring a full Gateway involved. While I spend my days building CANbus gateways for everything imaginable, I'm still not confident enough to say "sure program that box thru my gateway.." It's just a dumb risk, and someone out there is sure to find that bug you never considered...

- Tre-Cool
- Posts: 414
- Joined: Tue Oct 16, 2012 12:17 pm
- cars: VY SS UTE, VX Drag Car
- Location: Perth
- Contact:
Re: E67 to T87/T87A TCM interface needed
Hopefully that efi cax file i sent to you with what i suspect are all the tapshift parameters might help with the standalone tap shifting options. The car i will try the 10 speed out in first uses direct wire method, so will require changing.
I managed to find a different calibration but same OS to what i used for finding the tapshift options & doing a compare in efilive they also have different CAN broadcast settings between the 2 cals.
1 was a 2018 tahoe, the original file i have no idea, was 1 of the files uploaded to the forum here.
I managed to find a different calibration but same OS to what i used for finding the tapshift options & doing a compare in efilive they also have different CAN broadcast settings between the 2 cals.
1 was a 2018 tahoe, the original file i have no idea, was 1 of the files uploaded to the forum here.
Re: E67 to T87/T87A TCM interface needed
If only speaking of GM, just have to look for a few commands:kidturbo wrote: ↑Mon Jun 17, 2024 4:48 pm So yes an easy way to shut off anything additional to the bus is always tops on my list. And a big reason for perusing this setup without requiring a full Gateway involved. While I spend my days building CANbus gateways for everything imaginable, I'm still not confident enough to say "sure program that box thru my gateway.." It's just a dumb risk, and someone out there is sure to find that bug you never considered...
![]()
1) 101 FE 01 28
2) 101 FE 02 A5 XX ( xx may be a few numbers).
3) 101 FE 01 3E
Typically I setup my own stuff so they go quiet when seeing 1 and 2.
And will put a timer that exits silent mode if it does not see message 3 (Tester present) after 10seconds (which is way larger then needed).
Your Local Aussie Reverse Engineer
Contact for Software/Hardware development and Reverse Engineering
Site:https://www.envyouscustoms.com
Mob:+61406 140 726

Contact for Software/Hardware development and Reverse Engineering
Site:https://www.envyouscustoms.com
Mob:+61406 140 726

Re: E67 to T87/T87A TCM interface needed
Wrapped up the Camaro SS E38 stand alone T87A bench testing last week. This particular 10Lxx OS we have been testing with runs clean on the bench, only requiring a total of 14 PCM ID's, but no BCM or EBCM node data. This looks to be a good base starting point, with no ECM or TCM bin file CANbus modifications required to keep everyone happy.
The required PCM ID set seems to match up pretty closely with a 2016 E92 OS I have recorded. So now we know what years this particular OS is good back to with NO emulator or gateways involved. Believe this 10Lxx OS might match up perfectly with the earlier 8L90 T87 OS version I'm guessing. To many bits on the brain at the moment, but will verify when I get a time.
Planning to take this E38 with an emulator on a test run to the LS3 engine setup again next week. Last thing is to verify is writing out the single wire Range Switch Inhibitor code. Not a problem if you are running an E92 setup, or just plug one wire and power into the harness.. But the T87A won't even begin to go into gear with that ECM link missing. So hopefully calling this version a wrap, and now just need a guinea pig to take one for a drive.
Next up, will be the 10L-Anything..
Holley, FuelTech, Bosch, Haltech, MoTeC, Emtron, Ford, GM, Chrysler, and probably toss a J1939 version in for good measure. Since I already have the gateway architecture and every major ECU protocol coded in, it's just a quick revision and to have a truly 100% stand alone, just give it a pulse and it will shift version. If I could only find someone to make up a very nice stand alone harness now... Humm..
And while cleaning off my desk for first time in over a decade, I found this old CANbus worksheet from 2015ish when I was first reverse engineering that 2012 Corvette [Holden] protocol. Sniffed this out with an original CANbusTripple and WireShark.. No idea how accurate any of this is now, way to many moons have passed. But figured you would appreciate this one Tre-Cool.
Thinking maybe I'll frame it for nostalgic purposes.
The required PCM ID set seems to match up pretty closely with a 2016 E92 OS I have recorded. So now we know what years this particular OS is good back to with NO emulator or gateways involved. Believe this 10Lxx OS might match up perfectly with the earlier 8L90 T87 OS version I'm guessing. To many bits on the brain at the moment, but will verify when I get a time.
Planning to take this E38 with an emulator on a test run to the LS3 engine setup again next week. Last thing is to verify is writing out the single wire Range Switch Inhibitor code. Not a problem if you are running an E92 setup, or just plug one wire and power into the harness.. But the T87A won't even begin to go into gear with that ECM link missing. So hopefully calling this version a wrap, and now just need a guinea pig to take one for a drive.
Next up, will be the 10L-Anything..
Holley, FuelTech, Bosch, Haltech, MoTeC, Emtron, Ford, GM, Chrysler, and probably toss a J1939 version in for good measure. Since I already have the gateway architecture and every major ECU protocol coded in, it's just a quick revision and to have a truly 100% stand alone, just give it a pulse and it will shift version. If I could only find someone to make up a very nice stand alone harness now... Humm..

And while cleaning off my desk for first time in over a decade, I found this old CANbus worksheet from 2015ish when I was first reverse engineering that 2012 Corvette [Holden] protocol. Sniffed this out with an original CANbusTripple and WireShark.. No idea how accurate any of this is now, way to many moons have passed. But figured you would appreciate this one Tre-Cool.
Thinking maybe I'll frame it for nostalgic purposes.

Re: E67 to T87/T87A TCM interface needed
Pulled up some old high speed CAN captures from a 2016 LT4 Corvette, and ran that ID and byte list against our modified 10L T87A OS required list. Found they match up perfectly. Next test, see how it replays against the T87A live.
So if this current T87A data set match up correctly, and everything plays nicely, then this bit switch actually swaps everything back to 2015 platform requirements. Meaning anything factory equipped with an 8spd, can easily swap up to a 10spd with just the TCM change. Covering any E92, but what about our E67's?
Recall reading guys doing early E92 - 10L swaps, having to change the ECM or OS out also? If so, this new T87A OS should cover that issue. It's basically our early T87 6-8spd GMLAN set on the newer T87A 10L90 10spd hardware. So who wants to test this one in real life? I'm really dying to know if it actually was that simple.
Also found this elusive 6L80 TCM connector if anyone wants some. A buddy had me track these down. Which wasn't easy.. Much like the Allison T14 TCM header, these valve body connectors were never released for a general purpose outside of Ford, GM, Opel. But at least they do exist in the world, with a somewhat in-line format.
Below, this T14 TCM side was never produced in a terminal connector, and I had build it from scratch. So far about a grand into making 1 stinking connector. For the love of hacking. LOL. But it's finally hitting OEM specs on tolerance and materials. The A50 TCM swaps are way hotter commodity right now than 8L/10L suff. Simply because all of those factory A50 units are dying off of internal board issues. So tackling that 2007-2014 Allison T87 swap first is what really helped hash out this 10L data stuff so quickly. But has taken 4 months and 8th rendition into this simple 80 pin connector. And I thought this would be the Easy Part..
So if this current T87A data set match up correctly, and everything plays nicely, then this bit switch actually swaps everything back to 2015 platform requirements. Meaning anything factory equipped with an 8spd, can easily swap up to a 10spd with just the TCM change. Covering any E92, but what about our E67's?
Recall reading guys doing early E92 - 10L swaps, having to change the ECM or OS out also? If so, this new T87A OS should cover that issue. It's basically our early T87 6-8spd GMLAN set on the newer T87A 10L90 10spd hardware. So who wants to test this one in real life? I'm really dying to know if it actually was that simple.
Also found this elusive 6L80 TCM connector if anyone wants some. A buddy had me track these down. Which wasn't easy.. Much like the Allison T14 TCM header, these valve body connectors were never released for a general purpose outside of Ford, GM, Opel. But at least they do exist in the world, with a somewhat in-line format.
Below, this T14 TCM side was never produced in a terminal connector, and I had build it from scratch. So far about a grand into making 1 stinking connector. For the love of hacking. LOL. But it's finally hitting OEM specs on tolerance and materials. The A50 TCM swaps are way hotter commodity right now than 8L/10L suff. Simply because all of those factory A50 units are dying off of internal board issues. So tackling that 2007-2014 Allison T87 swap first is what really helped hash out this 10L data stuff so quickly. But has taken 4 months and 8th rendition into this simple 80 pin connector. And I thought this would be the Easy Part..

- Tre-Cool
- Posts: 414
- Joined: Tue Oct 16, 2012 12:17 pm
- cars: VY SS UTE, VX Drag Car
- Location: Perth
- Contact:
Re: E67 to T87/T87A TCM interface needed
My drag cars have Haltechs in them so that would be something i'd love to test, even with a 6l80e. There is a local mob that made a can gateway for the t43/haltech stuff, but i don't think they ever really sold them in great numbers. I've heard feedback that it didn't work well.. it was certainly something that was "in-testing" for a while.
I've got the new output shaft for my 10L, just haven't done anything with it yet until I've got a harness/connector i can wire in.
I've even got a plug n play harness to adapt from e38 to an elite ecu that i've never used.
I've got the new output shaft for my 10L, just haven't done anything with it yet until I've got a harness/connector i can wire in.
I've even got a plug n play harness to adapt from e38 to an elite ecu that i've never used.
Re: E67 to T87/T87A TCM interface needed
Haltech it is..
See it has a little TrqMgmt data streaming that we can likely make use of. Just recently added it to our N2k gauge set, so still fresh in the mind at least.
A drag car is the perfect setup for testing thru my existing gateway hardware. As I don't have any extra GPIO's pinned out on the IP67 models. Just data in - data out, on two M12 connectors. While great in harsh environments, the design could use a few "Optional" pins for our P/N, tap, and mode settings requirements the real world might expect. Guessing you be Ok without Cruise ??
See it has a little TrqMgmt data streaming that we can likely make use of. Just recently added it to our N2k gauge set, so still fresh in the mind at least.
A drag car is the perfect setup for testing thru my existing gateway hardware. As I don't have any extra GPIO's pinned out on the IP67 models. Just data in - data out, on two M12 connectors. While great in harsh environments, the design could use a few "Optional" pins for our P/N, tap, and mode settings requirements the real world might expect. Guessing you be Ok without Cruise ??
Re: E67 to T87/T87A TCM interface needed
And Link too please??? I've got a conversion coming up for a rb25 Nissan to 6L80.kidturbo wrote: ↑Tue Jul 09, 2024 2:47 am Haltech it is..
See it has a little TrqMgmt data streaming that we can likely make use of. Just recently added it to our N2k gauge set, so still fresh in the mind at least.
A drag car is the perfect setup for testing thru my existing gateway hardware. As I don't have any extra GPIO's pinned out on the IP67 models. Just data in - data out, on two M12 connectors. While great in harsh environments, the design could use a few "Optional" pins for our P/N, tap, and mode settings requirements the real world might expect. Guessing you be Ok without Cruise ??
- Tre-Cool
- Posts: 414
- Joined: Tue Oct 16, 2012 12:17 pm
- cars: VY SS UTE, VX Drag Car
- Location: Perth
- Contact:
Re: E67 to T87/T87A TCM interface needed
Well the elite/nexus can do the cruise through the software if your running DWB, which i am in 1 of the cars. I have a customer with a manualised t700 with an elite too. when that box dies again i think it's highly likely i can squeeze a 6L in the tunnel.kidturbo wrote: ↑Tue Jul 09, 2024 2:47 am Haltech it is..
See it has a little TrqMgmt data streaming that we can likely make use of. Just recently added it to our N2k gauge set, so still fresh in the mind at least.
A drag car is the perfect setup for testing thru my existing gateway hardware. As I don't have any extra GPIO's pinned out on the IP67 models. Just data in - data out, on two M12 connectors. While great in harsh environments, the design could use a few "Optional" pins for our P/N, tap, and mode settings requirements the real world might expect. Guessing you be Ok without Cruise ??
Ive just taken a look in the haltech nsp software to see how it could take the torque reduction request, i might need to reach out haltech to see what or how it can be done as they only have a couple of supported can protocols. i do remember that in the latest update they added a whole bunch of more customizations to can speeds recently.