Sprinter 4WD Conversion Idea, GMT-800 IFS.

[luthj]

I don't think any suburban came with 1410 ujoints and I think only the 2500hd and up trucks came with 1410 on the rear shaft only. Worth a double check on that.

You are probably right. I wast doing a few googles trying to roughly estimate driveline torque. I do need to work out what GM used on the front, but thats a ways down the road at this point.

Just curious, what Factor of Safety are you using for the subframe and other fab parts that will see dynamic loads?

Honestly I am not sure yet. The issue with dynamic loads on a vehicle are their unpredictability. Especially with regards to shock loading. Thankfully the tires and wheels serve as buffers, bending or deforming to absorb energy. Same goes for suspension bushings.

Right now I am going to simulate worst case loading. Such a single wheel catching a rock/curb during a lateral slide, hitting a curb at speed, etc. From there I am looking for a FOS of at least 1.5 for these. For cyclic loading, like shock towers, etc. I would use FOS of 2 as the minimum value. The endurance limit for steel is usually about half the ultimate tensile strength, so staying above 2 takes fatigue failure mostly out of the picture.

Other worst case scenarios would be a bogged recovery using on lower A-arm, dropping the subframe on a rock, and jacking from one of the cross-overs.

I am more accustomed to working with aluminum, which has no fatigue/endurance limit, and has much lower toughness. Steel is so much more forgiving, as it can absorb so much more energy in deformation, and still carry a load.

 

[luthj]

Taking another approach to calculating diff transmitted torque. With dry road friction coefficients of around 0.7 with street tires, and a front axle weight around 4000lb (backing up a hill for example). I would be looking at 2,800lbf on the front wheels. Tire diameter is 31" (radius 15.5). Yield a diff torque of 2,800lbf x 1.29ft = 3,616lb-ft of torque. With an FOS of 1.5 x 3,616, and adjusting for mount offset, give me a diff mount loading magnitude of ~5,800lbs.
5,800lb seems like a rational number. At least compared to other diff mounts I have seen? Its a place to start at least.

 

[Len.Barron]

IIRC they run something in the 1310 series for front shafts.

 

[luthj]

IIRC they run something in the 1310 series for front shafts.

Okay, thanks. That matches what I think is on my Jeep NP242. 1310s are rated around 1,600 lb-ft for yield, and 800 continuous. Based on that I think reasonable design parameters for diff loading are shock for 5,968lb-ft and continuous of 2,984lb-ft. That 5,900lb-ft value is very close to the worst case wheel loading scenario I used above. So both the wheels max traction and UJ max load are roughly in the same ballpark.

 

[luthj]

Put a few more pieces on my redneck jig.
IMG_20190123_144721 by J Luth, on Flickr

Steering angle is somewhere around 70 degrees total. It isn't symmetric though, so with the wheels connected its probably around 60 degrees.

IMG_20190123_144620 by J Luth, on Flickr

The Gm stops are not linear with suspension travel, so the farther from the natural ride height you get, the smaller the allowed angle is.

The digital angle finder has been indispensable. Camber gain is pretty flat, less than 0.5 deg per inch.

IMG_20190123_145112 by J Luth, on Flickr

GM was nice enough to put a few surfaces that are perpendicular to the kingpin axis. I am aiming for at least 2.5 deg, and no more than 3.5 deg of caster. This is roughly the range the sprinter uses.

IMG_20190123_145133 by J Luth, on Flickr

The jig has already proved its value, as I needed to adjust my upper arm pivot nearly 1/3 of an inch.

 

[locrwln]

Anyone have suggestions on typical loads for diff mounts? Transmitted torque numbers would work as well. The NP242 can handle about 1500ft-lbs of input torque. Multiple that times low range 2.7, then times 3.73 gears yields a whopping 15,000 lb-ft of output torque... Which is obviously doesn't pass the sanity test. 5,500lb-ft is may be more than my current rear mount can handle. I need to account for the bolt clamping force still.

View attachment 495123

I think this Diff has a yoke for 1410 UJ. Which is rated at 1500ft-lb for short bursts. So 1500x3.73 is 5,600 ft-lbs at the CVs. Reaction forces would be about 40% higher due to level arms, so about 7,800lbs at the mounts worst case? That is pretty significant. I may need to rethink forward mount ideas.

The front u-joint at the differential is a saginaw 3R joint (which I think is in the 1300 series family size wise). The Duramax 2500HD trucks use a 1410 on the rear shaft. I can't tell you for sure what will handle what, but keep in mind that some of the D-max's are putting out ~650-750lb/ft of torque (stock and even more on modified trucks) through a 2.72 t-case on 1/4" bracket lifted trucks and I don't think I have ever seen a breakage due to torque input on any of the lift brackets.

Jack

 

[luthj]

I was trying to keep the welding to a minimum on the subframe. After working through the loads on the diff mount, I am expecting to partially or fully box half the part... I ended up going to 1/4" stock on the front bit, the rest is 3/16 plate.

The front u-joint at the differential is a saginaw 3R joint (which I think is in the 1300 series family size wise). The Duramax 2500HD trucks use a 1410 on the rear shaft. I can't tell you for sure what will handle what, but keep in mind that some of the D-max's are putting out ~650-750lb/ft of torque (stock and even more on modified trucks) through a 2.72 t-case on 1/4" bracket lifted trucks and I don't think I have ever seen a breakage due to torque input on any of the lift brackets.

Jack

Thanks!

That sounds about what I would expect. With torque its all about how long the lever arm is, with the diff mounts only 7" from the centerline, it does amplify it quite a bit.

 

[luthj]

Still trucking on the Subframe. Lots of reinforcement, and a couple rounds of load simulations. It is over 90lbs now... Rear drivers side diff mount is pretty much finalized. I roughed in the rack mounts. I need to work out a method that will allow me to adjust its angle/position without needing to cut the bracket up (during fabrication) as I anticipate the final positioning will change when everything is bolt up. I am trying to incorporate slots and some play in the various attachments. This should allow some fudge-factor in both assembly, and my design.



After some head scratching I think that I can re-use the factory passenger side mount. I am going to flip it upside down, and left to right. All 4 studs will be pressed out and reversed.

The drivers side forward mount is gonna have to come off. I will make a bolt-on bracket, and hopefully pickup some of the front cross-over. It will need to be a bolt on attachment, so the diff can drop out for maintenance.

My current plan is to use a normal front driveshaft (2 UJ), and get the diff and Tcase angles matched. Though maybe I should just plan on using a CV driveshaft and having the diff tilted up more? Say 10 degrees? I guess I need to mock up the driveshaft to see what the angles will look like. Its tough to estimate as I don't have the Tcase bolted up. Which reminds me, I need to get a trans housing and adapter casing...

 

[Len.Barron]

The eccentric/cam bolt that they use on the solid AAM 9.25" lower trailing arm (for caster adjustment) would be a cool fix for adjustability.

 

[luthj]

The factory bracket would either require a 6" riser (and be pretty close to the bell housing) or interfere with the rack mount. So I will just fabricate a bracket. The slots allow for a combined adjustment of plus/minus 4 degrees. I will use some weld-in urethane bushings at both ends. If you look close you might see slots/tabs starting to appear. Its time consuming and fiddly, but will pay dividends when i go to assemble everything.



I am trying to get the wheels nearly centered in the wheel wells. So I need to get some measurements of where the subframe locating bolts are with respect to the wheel center. Not really possible to do with the factory subframe in place. I am a bit concerned that the 4 bolts are going to interfere with the major lateral plates of the design. If so I am going to have to be creative I think.

On a related front I am in contact with a EE/CE about a fix for the low range TCM issue. After gathering some info from an Australia programmer, I think we have a good approach. We are going to use an arduino with 2 CANbus segements. One for the TCM, the other for the rest of the vehicle. The arduino will pass all frames between the two. When in low range it will modify the wheel speed frames (one for rear, and one for front). Multiplying the wheel speeds by the low range ratio. That should keep the TCM happy.

 

[mgmetalworks]

You may not have this issue because of the year of the vehicle you're working with (and it's CAN bus simplicity compared to newer vehicles) but I found the Arduino to be too unstable to handle too much "traffic" on the bus. My application was a bit more complicated (actually, a lot more complicated) but just a heads up that Arduinos have limitations.

If you can find one, a CanBusTriple will do exactly what you're wanting to do and it's basically set up for you with a great interface. https://canb.us/

 

[b dkw1]

So 1500x3.73 is 5,600 ft-lbs at the CVs.

Unless you have a spool in that diff you need to divide that by 2. Even that will be generous. Doing FEA on suspension mounts will just give you some interesting pictures. Your loadings will come from so many directions that it is damn near impossible to get an accurate result.

 

[luthj]

You may not have this issue because of the year of the vehicle you're working with (and it's CAN bus simplicity compared to newer vehicles) but I found the Arduino to be too unstable to handle too much "traffic" on the bus. My application was a bit more complicated (actually, a lot more complicated) but just a heads up that Arduinos have limitations.

If you can find one, a CanBusTriple will do exactly what you're wanting to do and it's basically set up for you with a great interface. https://canb.us/

Thanks for the feedback. I was following your trials with the Cummins swap. The sprinters CANbus is 500kbit, I am hopeful that a modest arduino can handle the traffic. There really isn't a whole lot going over the T1N sprinters canbus anyways. If necessary I can always identify and drop inbound traffic from the A/C head unit and immobilizer, as the TCM doesn't monitor those.

Unless you have a spool in that diff you need to divide that by 2. Even that will be generous. Doing FEA on suspension mounts will just give you some interesting pictures. Your loadings will come from so many directions that it is damn near impossible to get an accurate result.

The dynamic loading on the arm/bushings is hard to anticipate. That being said I can gauge which direction the mounts are weakest in, and use FEA to assess some worst case loading. Its not a complete assessment by any means, but its much better than my guesses. Because the upper arm mounts are on risers/levers, they producing significant torque when pulled away from the frame. I may end up welding a tab to the sprinters frame rail. I can attach that to the upper arm mounts with a bolt. This would allow direct load transfer to the frame, and still keep the subframe removable.

 

[luthj]

Rounds of refinement are rapidly finalizing the design. Clearances are coming into focus, and I am double checking all my measurements. I am aiming for .4-.5" of clearance around the diff and output flanges. Still adding tabs... slot/tab groups don't mirror across different bodies, which adds a fair bit of work.



PS diff bracket is finalized.



Final tasks are:
Rail mounting bolt locations
Steering rack mounts
Bump stop calibration (for ride height)
Engine mount towers (location, angle)
Drivers side forward diff mount

The rail bolts will have to wait until I tear the van down. I have a used rack in storage I can measure. I need a 2500 bump stop to put on my jig to adjust its contact at ride height. The engine mounts will also have to wait until I drop the MB subframe.


Looking forward. Here is the jeep 242 tcase I grabbed at the local yard.

IMG_20180903_095638 by J Luth, on Flickr

IMG_20180903_110413 by J Luth, on Flickr



Here is what the MB rack looks like.

IMG_20180907_182606 by J Luth, on Flickr


Other major components I need.
Wrangler NAG1 to NP/NV adapter case and shaft
Jeep NAG1 main case
Custom front/rear driveshafts. The rear may be interesting, as MBs input flange is not common.
Tcase shift handle and cable (should be pretty simple)

In addition to the subframe, I need to design/fabricate the following.
Drop/mount brackets for GM torsion carrier
Drop spacers for MB trans crossmember
Adapter for Jeep trans isolator/mount to MB cross member
Tone rings for the outer CVs

Unknowns
Tcase to fuel tank clearance

 

[mgmetalworks]

Custom front/rear driveshafts. The rear may be interesting, as MBs input flange is not common.

Flange adapter?