DIY Tesla Drive Unit Repair

[jatguy]

Yeah, that's a datum I'm looking for as well. 0.2mm interference (~0.008") seems like an awful lot. In our design exercise, I used .0025" (~.06mm) as the interference fit. But this particular metals engineering is something I don't do; I've merely watched engineers I've worked with spec this stuff.

I've done a couple of measurements with the Mitutoyo bore gauge, and it seems that the surface is actually conical (at least on my part), and not super precise. The cone could be a result of deformations after the seal replacements in the past (it was replaced at least 2 times).
The inner diameter on the bottom of the groove is anywhere between 54.86 and 54.9 mm. And at the top of the groove 54.97-55.0 mm.

I also don't have any expertise in interference fit. But there are many online metal interference fit tables, I'm just not experienced enough to know how to interpret them or know any assumptions / caveats

It is quite simple and mainly depends on the material and the temperature difference. I usually calculate the thermal expansion of the material.
For 55mm OD aluminum cap the temperature difference of 100 degrees gives you the expansion of approx. 0.1-0.12 mm.
So if you want to squeeze the 55.12mm cap to the 55.0mm hole one will need to heat up the hole (or cool down the cap/shaft) by 100 degrees.

 

[mr_hyde]

What are your thoughts on a stainless cap pressed into the 55.0mm to 54.9mm bore? Would 55.1mm be too tight at the bottom and cause the cup to buckle?

 

[jatguy]

What are your thoughts on a stainless cap pressed into the 55.0mm to 54.9mm bore? Would 55.1mm be too tight at the bottom and cause the cup to buckle?

I think it really depends on the cap design.
If this will be a hollow cap made out of bent thin sheet metal (similar to the Chenming design but w/o hole) this would be possible.
In case of a solid machined steel cap it will shrink/expand by approx. 0.05mm (depending on the alloy) for 100 degrees temperature difference.
So probably that the 0.1mm tolerance is too big.

 

[sirpewe]

So, small update:
After some thinking, measuring and testing I've ended up with the idea of a design, similar to QC.
Mine is just a bit more simple and could be done on a 2D (CNC) milling machine to make it cheaper and easier.

Also, following the cap idea discussed before, I did a drawing with the dimensions that I think fit best. The main issue is the OD of the cap, which I believe should be 55.2mm + 0.01 or 0.02 for the hot press fit.

I will be testing both options in the nearest future.

View attachment 1027203

View attachment 1027197View attachment 1027199

Have you managed to test some?
If you have a CAD file for it then I can get one manufactured and do some testing.

 

[hanno.parksepp]

Have you managed to test some?
If you have a CAD file for it then I can get one manufactured and do some testing.

I'd be interested also in getting manufactured a more cost friendly version, especially for those not in the US. I probably made a mistake by ordering from QC Charge over a week ago. No shipment, no replies - so basically same attitude as already mentioned by someone before, so I guess better resort to some better alternatives.

 

[chrisro]

I can supply for you an alternative. We are in Germany.

 

[jatguy]

Have you managed to test some?
If you have a CAD file for it then I can get one manufactured and do some testing.

Yes, it is still in the development. I will share the CAD file once it will be reliable enough.
Today I've got isolation error on my Model S, so I need to do a rebuild asap...

 

[cleverscreenam]

I'd be interested also in getting manufactured a more cost friendly version, especially for those not in the US. I probably made a mistake by ordering from QC Charge over a week ago. No shipment, no replies - so basically same attitude as already mentioned by someone before, so I guess better resort to some better alternatives.

QC Charge sent me an email receipt immediately and my coolant-delete arrived in just 4 days... (This was 2 months ago)

 

[sirpewe]

I can supply for you an alternative. We are in Germany.

Do you have any more info abut this? I'm interested. I'm located in Sweden.

 

[ajbessinger]

I'd be interested also in getting manufactured a more cost friendly version, especially for those not in the US. I probably made a mistake by ordering from QC Charge over a week ago. No shipment, no replies - so basically same attitude as already mentioned by someone before, so I guess better resort to some better alternatives.

I would recommend calling the shop directly to talk to Tony (message/email are not always the best way to get ahold of him...). I know that supply was running low last I spoke with him, so they might be out. Another batch of the manifolds is being built, and should be finished in about a week and a half.

 

[hanno.parksepp]

I can supply for you an alternative. We are in Germany.

Excellent! I dropped you a line via your rosamotors website.

 

[geordi]

I would recommend calling the shop directly to talk to Tony (message/email are not always the best way to get ahold of him...). I know that supply was running low last I spoke with him, so they might be out. Another batch of the manifolds is being built, and should be finished in about a week and a half.

Is this something that requires the drive unit removed to install? I'm guessing the answer is yes... How much (ballpark) would that run? I've got one right now (2013 S P85 RWD) that makes a faint welding type sound when using or generating power, but silent when just coasting in neutral. I have another (2014 S base RWD) that makes no noise at all.

 

[asavage]

Is this something that requires the drive unit removed to install?

In the Model S, the LDU coolant manifold is right up against the subframe and cannot be replaced without removing the LDU from the subframe.

Additionally, if your LDU has any symptoms (noise, fault codes, driving issues, or coolant on the speed sensor) it has to come apart anyway, and I mean completely apart. Stopping the coolant leak doesn't magically fix existing damage.

 

[jatguy]

The main issue is the OD of the cap, which I believe should be 55.2mm + 0.01 or 0.02 for the hot press fit.

View attachment 1027197

After some hours on the lathe and a couple of unsuccessful caps, I've finally made a proper one and press fitted it for my future rebuild.
The problem was in the inconsistency of the ID. On my part the surface is more like oval, and the ID also different on the top and bottom of the groove.

Yeah, that's a datum I'm looking for as well. 0.2mm interference (~0.008") seems like an awful lot.

In the end I've used a 55.19 diameter of the cap, put some Permatex ultra black inside the groove (on the bottom) and Loctite 262 around the cap.

I will also fill the outer chamfer with some metal epoxy to be 100% sure.

Personally I think the quality and longevity of this "cap" solution depends on the person, who is doing a modification, and requires a lot of tools/materials/skills. For an average person, a CNC-machined replacement part (like from QCcharge) is the easiest solution, so I will finalize my design for the future.

 

[asavage]

In the Model S, the LDU coolant manifold is right up against the subframe and cannot be replaced without removing the LDU from the subframe.

As I'm doing this today myself, I will note that if you have a RAV4 EV with the LDU, and your LDU has NO SYMPTOMS or issues, the coolant manifold appears to be able to be R&R'd without removing the LDU from the vehicle, via the right fenderwell.

[The LDU is installed "backwards" in the RAV4 EV, from the way it's installed in the Model S.]

Pics shown with various plastic bits removed and lower "frame" members unbolted & removed, maybe a hour work with hand tools.

 

[brainhouston]

...
The problem was in the inconsistency of the ID. On my part the surface is more like oval, and the ID also different on the top and bottom of the groove.

...
In the end I've used a 55.19 diameter of the cap, ...
...

Nice work, i'm actually working on making one on my lathe too
Did u end up with smaller diameter towards the back of cap? or just 55.19 across the surface?
A bit confused by ur statement about ID difference top to bottom...

 

[jatguy]

Nice work, i'm actually working on making one on my lathe too
Did u end up with smaller diameter towards the back of cap? or just 55.19 across the surface?
A bit confused by ur statement about ID difference top to bottom...

I made 55.19 all across the surface. But, the ID of my coolant manifold indeed different top to bottom. I just thought the fit will be more tight on the bottom, which isn't bad

 

[ajbessinger]

As I'm doing this today myself, I will note that if you have a RAV4 EV with the LDU, and your LDU has NO SYMPTOMS or issues, the coolant manifold appears to be able to be R&R'd without removing the LDU from the vehicle, via the right fenderwell.

[The LDU is installed "backwards" in the RAV4 EV, from the way it's installed in the Model S.]

Pics shown with various plastic bits removed and lower "frame" members unbolted & removed, maybe a hour work with hand tools.


View attachment 1036514View attachment 1036515View attachment 1036516

Yep, on the RAV4 EV it is VERY easy to install. Mercedes B-Class is the next level up, as the "frame" rail is in the way, but by removing the side mount and lowering the motor you can get the manifold to fit without removing the drive unit. Both Model S/X of course require the drive unit to be removed.

 

[ajbessinger]

Is this something that requires the drive unit removed to install? I'm guessing the answer is yes... How much (ballpark) would that run? I've got one right now (2013 S P85 RWD) that makes a faint welding type sound when using or generating power, but silent when just coasting in neutral. I have another (2014 S base RWD) that makes no noise at all.

As @asavage said, the coolant delete manifold can't be installed on Model S/X without removing the drive unit, as the subframe side rail is in the way. At my shop, I normally quote about 6 hrs of labor (~$1000) for a coolant delete install. Even in a shop environment with lifts and all the tools you could want, it's nearly an all day job.
However in your case, being that that drive unit is making noise, it really needs to be taken out and rebuilt anyways... Especially if the noise is from coolant washing the grease out of the rotor bearings; that would likely mean that a fair amount of coolant has already made its way into the motor, which can continue to cause damage, even if you fix the source of the leak. In the event of pre-existing coolant intrusion, the only way to be sure that the motor will last long term is to take it apart, and clean out any coolant/corrosion that is already in the motor, and be sure to test stator isolation and all.

 

[asavage]

I made 55.19 all across the surface. But, the ID of my coolant manifold indeed different top to bottom. I just thought the fit will be more tight on the bottom, which isn't bad

I dropped off my 6061 roundbar and the manifold and a drawing and a rendering of the plug I wanted made, to my machinist on Tuesday; it's ready for pickup tomorrow morning (yea!).

He & I agreed that .003" (~.08mm) interference ought to be plenty on a part this size, and that's what I had him lathe it to.

In prep, I hacksawed off the aluminum tube, removed the circlip, then used an end mill to remove the leftover stub, down to the casting. Then I used a smaller end mill to cut a channel "north" (I bumped the seal counterbore, ouch, but NBD). So, my plug can use all the available space, and there's plenty of meat in the bottom of the plug and it should clear the rotor by a wide margin.