Will's Bug Out Rig: Duramax Van

[The Dark Side of Will]

As noted previously, in the vans, the V6 & small block powertrains are centered, while the big block and diesel powertrains are offset 2" to the right.
The diesel powertrains are centered in the pickup trucks.
I snagged a pickup truck engine mount and a van engine mount tower/bracket. The combo is shorter than the left van mount + van mount tower/bracket. However, a RIGHT van mount would be even shorter than that, as confirmed by photos on RockAuto and eBay. The van right mount can be used on the left as demonstrated by the fact that there's only one pickup truck mount. (Yes, GM's assembly is differentiated left from right, but that's because of the heat shield. All the aftermarket manufacturers list one mount and say to reuse the heat shield)

So I could use right van mount on both sides and raise the bottom surface of the van tower/bracket even further than if I used the pickup truck mount. Of course aftermarket suppliers have pickup truck mount from about $20, while GM is the only source for van mounts, and Rock has them for $77ish.

The 8100 and 5.7 (& 6.5 diesel) in the 2002 vans use the same engine mounts, so there must be different left/right mount towers/brackets for those engines.

The Duramax is bigger and bulkier than an LS, and has a bunch of junk associated with the oil filter and cooler low on the left. It probably needs to be shifted to the right to clear the diff at all.

My engineering effort required to design the frame mods to lift the engine mount pad up to the point that I can use a shorter mount, thus freeing up space under the mount for the diff, will *ONLY* be applicable to the Duramax vans. With the same engineered parts, more fab will be required to utilize the mount towers/brackets for an LS engine. Interdasting.

[The Dark Side of Will]

I've been accumulating parts for the van. I now have:

=>'11+ 3.73 front diff
=>'11+ CV axle (8 bolt/36 spline)
=>'87-'10 3.73 front diff
=>'99-'10 CV axle (6 bolt/33 spline)
=>1500 AWD van suspension crossmember
=>1500 AWD van torsion bars
=>1500 AWD van torsion bar crossmember
=>Duramax van engine mount bracket that bolts to frame (same both sides)
=>Duramax truck engine mount

On the way:
=>Right hand Duramax van engine mount

Because the big block and diesel powertrains are shifted 2" to the right in the vans, the right engine mount assemblies are shorter and the left engine mount assemblies are taller. In the big block & 6.5 vans, the mounts are the same, but the brackets are different. In the Duramax vans, the brackets are the same, but the mounts are different. The Duramax van mounts resemble the Duramax pickup truck mounts. The truck mounts are the same height L to R, but the van right mount is shorter and the van left mount is taller.

I previously snagged a pickup truck mount and van bracket to check out. Going to a shorter mount allows me to raise the frame bracket to create space for the diff. I later realized that the right van mount was shorter than the pickup truck mount. Going to an even shorter mount allows me to raise the frame bracket even more, creating even more space for the diff.

I'm not sure exactly how much space I need, so I'm going for as much as I can get right out of the gate.

[The Dark Side of Will]

Over the last few weeks, the van had developed a tendency to not crank after it sat for a few days, which it does every week. I'd turn the key, but the starter solenoid would chatter. After a couple of false starts in diagnosis, I discovered that the terminal bolts on the 2nd battery had worked loose. That would not be difficult to discover, except that the 2nd battery is hung on the outboard side of the left frame rail. You have to lie down on the ground next to the van to see the terminals. I snugged the terminals and it's been fine since.

The PO found the batteries dead and replaced them right before he shipped the van to me. Of course he used cheap batteries. I'm not sure if they are the batteries it's supposed to have or just whatever size the parts store had. They are two 600 CCA units. It's been close to 2 years, so I was thinking the batteries could be starting to go south, even though they've had easy lives. Apparently I was wrong and they're fine BUT a single 600 CCA battery just does not have the grunt to crank a Duramax.

Also, the spec torque for a side-terminal battery is 11 ftlbs.

[The Dark Side of Will]

When you own three diesel vehicles that are all due for oil changes...

[The Dark Side of Will]

Trying to get data comms info for the Allison swap:
https://www.duramaxdiesels.com/forum/th ... ons.88679/

The Allison was a 5 speed through 2005.
Starting in 2006, the Allison was a 6 speed. There were no changes to the transmission rotating components... just the valve body. Allison discovered an additional pattern for clutch engagement that gave them a 6th gear with a higher overdrive than 5th.

A 5 speed Allison can have 6 speeds just by swapping the valvebody and controller. HOWEVER, the factory 6 speed Allisons have rotating component upgrades vs the 5 speed transmissions, so a factory 6 speed is NOT just a factory 5 speed with a different valve body. In particular the 2017-2019 Allisons were built for use with the L5P Duramax handling 910 ftlbs of TORK! with a warranty. In 2020 GM introduced the 10 speed, which is a completely different transmission.

Due to cascading system changes as GM updated the trucks, then updated the vans in a limited way a year or two later, finding the right combo of ECM & TCM comms and features to make an Allison swap in the van work has not been figured out yet. The people who've done it either have integration problems or rewired the van to become a truck

[The Dark Side of Will]

A while back, the HVAC "mode selector" stopped working. That's the system that controls what vents the air comes out of. The vans are alte schule in that the mode selection is operated by vacuum servos. Since a diesel engine doesn't make vacuum, GM added a vacuum pump. Failure of the hoses connecting this vacuum pump to the HVAC system is common.

Since GM also did not want to add steps to the main line, they incorporated the vacuum pump into another assembly. In this case, the vacuum pump and accumulator are attached to the bottom of the battery tray. So the battery and tray have to come out in order to replace the hoses. I found that the old hoses had disintegrated, so I replaced the hoses and everything works again.

The van was advertised showing an engine driven air compressor in the back (the seller did specify that the compressor was NOT included in the sale). The starting current for that engine came off the vehicle batteries. The van has had that pair of cables ganged on replacement battery studs for the entire time I've owned it. Since I had the battery out, I disconnected them. They still go into the interior of the van, so I'll have to spend some time crawling around underneath it cutting zip ties to extricate them.

The negative stud would not go back into the battery straight and the positive stud would not take 11 ftlbs, so it looks like I need to replace these batteries sooner rather than later.

[pmbrunelle]

Will has a hypothesis that if the Allison transmission speaks SAE J1939, and that if the van's Duramax engine also speaks J1939, then both components could function together.

Not a bad hypothesis...

I would clarify that J1939 is pretty broad, so while I'd agree that it is necessary for both modules to speak J1939, I'd say that J1939 alone is not sufficient.

********************************************************************************

The approach I would take is to record the CAN bus traffic on the van, and on an Allison vehicle (a running and driving one).

If the CAN bus traffic is on the van is mostly indistinguishable from that of the Allison vehicle, then probably an Allison could be installed on the van.

********************************************************************************

For observing in real time, or for recording CAN bus traffic, we use the Ixxat USB to CAN interface at my work:
https://www.ixxat.com/products/products ... 0281.12001

We normally buy the cheapest variant:
Ixxat PN 1.01.0281.11001
https://www.digikey.com/en/products/det ... 01/9554731

In addition to reading CAN bus messages, the Ixxat can also be used to transmit messages on the CAN bus. I don't think you'll need this capability for your investigation.

I mention the Ixxat because it is what I know; other products may do something similar.

********************************************************************************

Here is what a CAN bus recording looks like (using the computer program that's included with the Ixxat):

400_mini_overview.png (52.35 KiB) Viewed 1674 times

The wires of the bus convey information using two distinct voltage levels; binary. In the human-oriented representation of the computer program, hexadecimal digits are shown. One hexadecimal digit represents four binary digits.

The recording is a series of messages, each one having a unique timestamp. Only one message can be sent on the bus at a time; a sender must wait for the bus to become silent before transmitting.

Every CAN message begins with the identification (ID) of the sender. For instance, the ABS may have one CAN ID, while the steering angle sensor may have another CAN ID.

Some devices may transmit using two or more CAN IDs (even though they only have ony physical CAN transceiver). For instance, you may present yourself as Will in a casual setting. In a military setting, you may present yourself as Rank Surname.

So, the engine may transmit RPM on one CAN ID, and shifting-related messages using another CAN ID.

Besides the obvious function of identification, CAN IDs also serve a role in bus arbitration. While senders must wait for the bus to be silent before transmitting, sometimes two senders may begin transmitting at the same time. In the event of a collision, the sender with the lower-valued CAN ID is permitted to continue transmitting, while the sender with the higher-valued CAN ID must silence itself (it may retry again later) and cede the bus.

The integrator who is setting up a CAN bus on a vehicle must ensure that no two devices transmit using the same CAN IDs. Also, knowing that CAN IDs are used to determine priority in the case of a collision, the integrator should make sure that high-importance messages are transmitted using low-value CAN IDs to make sure that they get through, even when bus load is high.

After the CAN ID is the Data Length Code, DLC. It is the length of the data packet which follows.

The data packet is the actual content of the message!

********************************************************************************

It is somewhat possible to interpret raw CAN datalogs. For instance, if you blip the accelerator while stopped, and you see some bits in a data packet change, you know it might be something like RPM, and definately not vehicle speed.

Maybe looking through the SAE J1939 standards you can attempt to decode the meaning of the CAN messages yourself.

Ideally, you can find a computer program (I don't know of any off the top of my head) that will take the raw recording and interpret the messages (per SAE J1939) directly into a human-readable format similar to the below image:

J1939 Shift.jpg (215.21 KiB) Viewed 1674 times

This is a repost of an image I posted on Old Europe (a screenshot of the intro document to SAE J1939).

********************************************************************************

The CAN bus messages (such as engine RPM) which are transmitted at a regular frequency should look the same between the van and the Allison vehicle.

Also, the conversations when the engine and transmission are interacting (during key moments) should look the same. I am talking about things like upshift, downshift, Park->Drive, Drive->Park, things like that.

I think it is important to study the key moments that occur while driving. Some integration problems may not be apparent if the vehicle is stopped and idling, because the conditions needed to make the problems occur haven't arisen yet.

As an example, on an OBD2 car, you can unscrew the fuel tank cap, and it won't throw an evap code right away; not until you drive the car some.

[The Dark Side of Will]

Yeah, I guess to do it right, I'll need to snag a bus sniffer. It's probably pretty easy to come up with a series of standardized short test sequences that would probably be illuminating.

I was kind of hoping I could just get the Allison TCM, plug it into the van bus without removing the 4L85E TCM and see if two TCMs showed up on a bus scan.

I'm also wondering if the vans didn't go to J1939 and stayed with Class 2 data... but since Class 2 is single ended and J1939 is differential, I should be able to tell the difference by studying wiring diagrams.

[pmbrunelle]

If the Allison and van TCMs speak the same language, then they would both transmit messages using the same CAN IDs... a shitshow on the bus would ensue.

If you do want to hook up a sniffer of sorts, be aware that J1939 specifies a maximum bridge tap length of 1 m from the main bus.

Connection-of-bridge-tap-between-central-office-and-modem.png (15.26 KiB) Viewed 1653 times

Probably since Class 2 is slower, you wouldn't have to worry about bridge tap length.

I'm not sure what buses you could access via the OBD2 port.

You might have Class 2 for the cluster and hotel stuff, and J1939 for the powertrain. It's common enough nowadays to have multiple buses in one vehicle. Yeah, you'll be able to distinguish the two based on the number of wires. But if there's also "GMLAN" which also works on a pair of wires, then the identification of the bus by schematic diagram may remain ambiguous.

J1939 actually requires a twisted pair, and I believe shielded as well, but not all OEMs will follow that requirement to the letter. Sometimes, "good enough" is sufficient, and the rest of J1939 can be implemented even though the physical layer doesn't fully meet recommendations.

[The Dark Side of Will]

The DuramaxDiesels.com link above has some info about comms protocols.

2001-2005 Pickup trucks used Class 2 between chassis modules and ECM, Class 2 to reflash all modules in the vehicle and J1939 within the powertrain (ECM, FICM, GPCM, TCM)
2006+ (later transitions not important for this discussion) Pickup trucks still use Class 2 between chassis modules and ECM, but GMLAN within the powertrain and to reflash powertrain modules. J1939 and GMLAN use the same physical layer, but GMLAN has a 500kbit data rate vs J1939 250kbit.

I need to look up the diagrams to be sure it's not Class 2, but I'm @$$uming that the 2006 van uses J1939, since that's the older way of doing things that fits GM's paradigm of only using older cheaper tech in the vans. I guess another way to tell would be if the van powertrain module reflashes happen over Class 2 or GMLAN.

The 6.5 diesel GMT400 pickup trucks have a TCM for the 4L80E, that I @$$ume talks Class 2 because it was the '90's and GM didn't have anything else. If GM recycled that module for the van 4L85E TCM, and there's no J1939 or GMLAN on the vehicle, that would suck. I guess that's the first possibility to rule out, but I can do that with just wiring diagrams. Being old school, I have the complete paper manual for the '06 GMT610 (>$300 used on Amazon... ) so I can actually look up the diagrams.

[The Dark Side of Will]

The other avenue that I'd thought about was using a TCM from a medium duty truck (Kodiak or TopKick), which may have been more likely to stick with J1939.

[The Dark Side of Will]

Also, there's Don's CANBUS interface thread: viewtopic.php?t=21768

Edit:
Medium duty truck stuff:
https://www.duramaxforum.com/threads/c4 ... t.1021635/
Not much here

https://www.duramaxdiesels.com/forum/th ... col.67025/
This guy says there are no 2006 Duramax ECMs that still talk J1939; another poster mentions marine oriented bus gateway
There's apparently a lot about bus stuff here... cool.

And a quickie intro to J1939... probably at a lower level than we already are:
https://www.csselectronics.com/pages/j1 ... o-tutorial

[The Dark Side of Will]

https://www.dieselplace.com/threads/250 ... st-9901277

If I had gone the IFS route, I would have definitely started with a truck lift that included a differential drop structure (then just modified that to my liking).

The van frame is really too flat to do a "zero lift" 4x4 conversion (with a duramax). You just don't have the necessary realestate between the diff housing and the oil pan and the drivers side motor mount solution is very tough to work through.

If your intent is to use the entire truck drivetrain(including allison) you really need to consider getting a newer truck donor(07.5 or later LMM would be good) so that will allow you to use the newer van dash wiring and cluster(2010 and later) and solve a bunch of other issues related to BCM communication with the ECM/TCM.

if you are going to stick with the 4l85e then it's much easier..

Just a snippet for the notepad.

[The Dark Side of Will]

https://www.duramaxdiesels.com/forum/th ... col.67025/
This guy says there are no 2006 Duramax ECMs that still talk J1939; another poster mentions marine oriented bus gateway
There's apparently a lot about bus stuff here... cool.

I've been dorking around in this thread, and while it's very informative, it doesn't solve the VanAllison problem.

It does link to some interesting CANBus capture tools, which may be useful.

From this post: https://www.duramaxdiesels.com/forum/th ... ost-940323

GM’s “Low Speed GMLAN” is a Single Wire CAN operating at 33.33kps.
It conforms to SAE standard J2411 & GM standard GMW3089.
Its Higher Layer protocol is OSEK.

From the now defunct mp3car forum via the Web Archive:
https://web.archive.org/web/20160114122 ... rface.html
Of course that's also YEARS old at this point.
OSEK: https://en.wikipedia.org/wiki/OSEK

So if the 2006 van powertrain comms work the same as the 2006 pickup truck powertrain comms, then the ECM, GPCM & TCM talk via high speed GMLAN, can be reflashed over the DLC via HS GMLAN, but communicate to the IPC and other modules via the older Class 2 protocol. The vans use a T42 TCM for the 4L85E, while the trucks use an A40 ('06-'08?) or A50 ('09+?) TCM.

The Vortec 8100 was available from 2001 - 2007 with the Allison in pickup trucks. The 8100 was also available in the vans, Suburbans and Avalanches, but with the 4L85E transmission. Now I wonder if the 8100/4L85E combo is set up like the DMax Vans with separate ECM & TCM, or if they're set up like the LS vehicles with a combined PCM that runs both units.

If all the 2006 Duramax ECMs speak the same HS GMLAN language to the TCM, then either the 2006 T42 or 2006 A40 TCMs *should* be plug/play.

An '05 TCM will not work with an '06 ECM because the high speed protocols are different. Also, the '05 TCM would be reflashed over Class 2, while the '06 ECM would be reflashed over HS GMLAN. An '07.5 (GMT 900) ECM calibration in a 2006 van would not talk to the rest of the vehicle because the vehicle uses Class 2 and the ECM low speed channel is Low Speed GMLAN (LS GMLAN).
However, if everything is from 2006, or from 2007 GMT 800's, then the Allison should be plug/play in the van, as far as protocols are concerned.

I can't help but think that the people who've tried the VanAllison swap and been unsuccessful may have screwed up by not being careful which year of ECM and TCM files they used. IE, for a 2006 or early 2007 van, there's a version of 2007 files that may work, and there's a version of 2007 files that definitely will not work in a 2006 van.

[The Dark Side of Will]

https://www.duramaxdiesels.com/forum/th ... ost-940418

On 2001-2007 trucks, tap-shift (on 06-07) and cruise control are discrete signals wired right to the TCM and ECM. Tow/haul goes over the Class 2 bus as a message from the BCM. Ive already figured out the tow/haul class 2 messages, so now its easy to keep tow/haul functional on 2001-2007 swaps that want to keep tow/haul without having to keep the BCM wired up.

I already kind of knew this, since on the '06 vans, removing the non-cruise multi-function lever and replacing it with the cruise control multi-function lever gives the vehicle cruise control with no other changes necessary, but it's good to see it in print. My van has tow/haul mode, but of course not TapShift.

[The Dark Side of Will]

The Vortec 8100 was available from 2001 - 2007 with the Allison in pickup trucks. The 8100 was also available in the vans, Suburbans and Avalanches, but with the 4L85E transmission. Now I wonder if the 8100/4L85E combo is set up like the DMax Vans with separate ECM & TCM, or if they're set up like the LS vehicles with a combined PCM that runs both units.

If all the 2006 Duramax ECMs speak the same HS GMLAN language to the TCM, then either the 2006 T42 or 2006 A40 TCMs *should* be plug/play.

Ok, the T42 was used on most of GM's product line from '06-14. It's a generic 4 speed automatic trans controller with 4L60, 4L80 and 4T80 options... probably 4T60's as well. I assume the 8100 / 4L85E vehicles would have this, just like my DMax / 4L85E can does.
https://www.eficonnection.com/home/prod ... ss-adapter

Also for the A40:
https://www.eficonnection.com/home/prod ... ng-harness

Oooh, there's a T43 that's used with 6 speed transmissions:
https://www.eficonnection.com/home/prod ... ng-harness

And *maybe* a way of installing TapShift on any T42/T43 vehicle?
https://www.eficonnection.com/home/prod ... nal-module

[The Dark Side of Will]

Thread on an interesting looking Marine CAN gateway gizmo:
https://www.duramaxdiesels.com/forum/th ... way.68389/

[The Dark Side of Will]

The Vortec 8100 and 4L80E were used in the vans in 2001 & 2002. That was old enough that they probably use a PCM to control both engine and transmission.
The Vortec 8100 and 4L80(E?) were used in the Avalanche 2500 and Suburban 2500 through 2006. That year probably had an ECM & TCM, since it came after the switch to 58x.
The Vortec 8100 and the Allison were used in the GMT 800 trucks through 2007.

So in 2006, there were the following combinations:
DMax + Allison
DMax + 4L80
8100 + Allison
8100 + 4L80

The powertrains in '06 should all communicate via GMLAN at 500k. This contrasts with the '01-'05 Duramax/Allison powertrains that had a J1939 bus at 250k. However, the T42 four speed auto TCM and the A40 Allison TCM might use different "vocabulary", meaning a different set of bus messages. If that's the case. then the ECM's from the 8100/4L80 and DMax/4L80 combos speak with that vocabulary. That might explain why the Allison TCM doesn't work well with the van calibration. I'm not sure what "doesn't work well" means, when coming from the one person who's tried it.

Sill thinking... there's something here that *should* work. Explanations always sound simple... once you find them.

[ericjon262]

I've been quietly following this for my DMAX 2500/Suburban mash up.

[The Dark Side of Will]

I doubt you'll have much trouble... the Duramax/Allison as a standalone powertrain swap is pretty well sorted. I'm only having questions (I don't know yet that there are actually issues or problems) because I need to integrate the Allison it into an existing system that did not include the Allison to begin with.