Tesla Large Drive Unit (LDU) Motor Teardown and maintenance

[howardc64]

DELETED

[howardc64]

DELETED

[2oldteslas]

I'm going to try this 2 lip PTFE seal for my 2012 P85. It took 8 days to come in the US mail from New Jersey to the Chicago area. It ships with an internal collar spacer fitted to keep the seals oriented and stretched until installation.

[2oldteslas]

Howardc64 RE: Spline grease.
A good sticky grease used for propellor shafts is made by Overall Supply Inc from New Jersey. It is called "super grease" and is very sticky. It worked well on tug boat prop shaft bearings and seals. A couple other synthetic greases are made by Hydrotex and Certified Laboratories. I used the super grease on my rotor splines.

[howardc64]

DELETED

[jrbe]

Has anyone looked at making a new coolant extension tube (blue) an adapter lip / flange (aqua green - black is an o-ring) that extends into an internal seal (red) in the motor shaft?
The coolant extension tube could be 2 pieces (at the dashes) to be a bit more cost effective.

Anyone have the shaft ID where the red is?

This would cut down on coolant flow somewhat. It would also give a smaller seal diameter / surface speed. Might give a bunch of seal options flipping it like this. The wear item would be the adapter lip / flange and should be easy to replace if needed.
Could also be a solution for motors with mangled sealing surfaces.

*edit, the seal lip would probably open from centrifugal force if one tried to use a seal pressed into the motor shaft as I drew. Probably would have to flip it to an external lip seal.*

[jrbe]

This is generic info but worth a read for people considering multi-lip seals. PTFE might be allowed to run dry but I wouldn't expect that to be the case at the speeds that these are spinning.

For a radial shaft seal to seal efficiently over a long period, the sealing lip must be lubricated. This reduces friction and wear to the sealing lip and shaft. Dry running of sealing lips made of standard materials should always be avoided. To prevent dry running, coat the counterface surface with a suitable lubricant prior to seal installation.

The lubricant must not only lubricate the sealing lip to reduce friction and wear, but also dissipate heat generated by the seal. To promote heat dissipation, a sufficient quantity of lubricant must be able to reach the sealing lip from start-up.

Some rolling bearings, such as angular contact ball bearings, tapered roller bearings and spherical roller thrust bearings, as well as gears, create a pumping action by virtue of their design. This means that the sealing lip can either be starved of lubricant, or subjected to excessive quantities of lubricant. In either case, steps must be taken during the design stage to make sure that the proper amount of lubricant reaches the sealing lip, as too much or too little can affect seal performance.

To prevent lubricant starvation, lubrication ducts can be provided. If the seal is subjected to excessive amounts of lubricant, a flinger can be installed between the bearing and seal.

In applications where the sealing lip is not exposed to a lubricant, for example when two seals are installed in tandem, grease or oil must be supplied separately to provide lip lubrication. In some cases, it may be sufficient to provide an initial grease fill between the two lips.

under lubrication.
Trick would be to find a PTFE compatible lubricant that doesn't gum up when expose to g48.

[howardc64]

DELETED

[howardc64]

DELETED

[howardc64]

DELETED

[2oldteslas]

Thanks, that grease looks thick! haha

Few questions

- Did you mix it with the original grease?
- Any longevity data? I guess need to pull again to see for sure.. PITA haha
- Any thoughts on electrical conductivity characteristics?

Thanks

-I cleaned off most of the old grease but not all so a small amount is mixed in with the super grease.
-no information on longevity or conductivity. Judging by experience it is a good heavy duty grease.

[2oldteslas]

These are the 3 coolant seals I bought. Single lip, double lip and triple lip. Time will tell which works best.

[howardc64]

DELETED

[howardc64]

DELETED

[howardc64]

DELETED

[howardc64]

DELETED

[muehlpower]

So you assume that the Tesla solution is to omit the rotor cooling, because it doesn't do anything anyway?

[jrbe]

Has anyone reached out to Parker Hannifin for a 19657H1L5PTFE seal?

I'm not sure why they couldn't list that as a metric seal.
Might be worth trying through motion industries or other reseller if they don't want to be bothered directly.

It's an ss style lip,

Screenshot_20230805_072510_Adobe Acrobat.jpg (9.88 KiB) Viewed 12007 times

My previous scribbles on this might not work out well as I first thought. I added an edit note to my post on that. With a seal in the shaft spinning, centrifugal force will probably open the seal lip. But they have a bunch of external lip seals if someone wants to experiment with flipping the sealing method inside out. With a sleeve pressed (and sealed) inside the hollow shaft it could give a decent surface for an external lip seal.

They also have FKM seals with a bonded PTFE section to have the flexibility of rubber and a sealing edge / wear / friction surface of PTFE. Click on the image and zoom in, white rectangle is PTFE,

But no off the shelf offerings for this that I saw in their design guide / catalog.

They also have seal designs with a molded in spring like section to allow for greater misalignment (some shown in the design style image above.) I don't think it's needed here though. Lots of interesting things in their 400 page design guide,
https://www.parker.com/content/dam/Park ... s/5350.pdf

[howardc64]

DELETED

[howardc64]

DELETED

[jrbe]

Yes, I reached out to Parker and Saint Gobain custom seal divisions. Neither wanted to engage. Saint Gobain said the seal was proprietary in nature so didn't want to provide services. I'm guessing all these big manufactures already contacted by Tesla/Tier 1 supplier. Maybe some are past/current LDU seal manufacturers.
...
viewtopic.php?p=49553#p49553
viewtopic.php?p=46858#p46858

Sorry I missed those posts.

I find it strange that potential OEM suppliers are not willing to sell parts. I could see them not using a Tesla logo or the Tesla part # on it. Maybe they know it's a bad design, they might just see it as a liability that they know will fail and shut it down asap. Maybe that's Tesla's refusal to sell seals too.

For clarity, 19657H1L5PTFE should be an off the shelf part. No need to go the custom route or even mention Tesla while trying to order some. It doesn't seem to be a thick sealing surface but that could be a good thing given the speed it can run at.

For the rotor cooling temperature reference, using an IR sensor should get the rotor temp without complications or assumptions. They should be pretty immune to electromagnetic environments. There are also temp recording paints that could be manually checked for testing / verification.

[howardc64]

DELETED

[howardc64]

DELETED

[SuperV8]

All I can tell you is its pretty clear Tesla is anti aftermarket therefore DIY. Most of the 5-10 seal makers I've contacted told of prior engagement with Tesla/Tier1 supplier on the coolant seal. One told me they stopped after finding out was in some kind of design competition to make one that doesn't have prominent leak problem. Tesla is a big gorilla in relation to these seal makers (large/small) in terms of business and PR leverage. By contrast, other car manufacturers don't really police parts manufacturers selling aftermarket. I've done much DIY on Volvo+VW/Audi repairs, OEM manufacturers grinding off car manufacturer logo and sell aftermarket is common place. So my guess is Tesla is pushing legal restrictions to their manufacturers.

Tesla's SC policy is also anti in-field end point learning, modding, enhancement. Tesla provide no solutions for their SC to service LDU+battery pack at SC (I think finally enabling contactor repairs etc) So its ship to repair center and back. Huge downside of this policy is the constriction of design improvement (many brains working at the end point rather than few (if any on last gen car) at HQ) and the core cost of such integrated and expensive item. Currently port warranty customer is paying for the expensive core. If you want an LDU+battery pack, need to turn in your old one and pay for the rebuilt one ($~7.5k for LDU, ~$15k+ for battery pack)

With off the shelf Parker parts, its basically going through distributor (if call seal manufacturer, they just point you to distributor) I've had no luck with distributors (Motion Industry + others) on DIY effort/1x volume. EU's much stronger right to repair laws seems to have a much more robust aftermarket DNA. Silicon Valley companies like Apple+Tesla is amongst the worst in aftermarket enablement. If you try to get the SKF ceramic bearing for the LDU in US, its basically impossible. But through EU, distributors are even willing to ship SKF bearings to US which is basically grey market. Talking with bearing distributors, they've also used Canada's more EU business relations as a grey market sourcing for US buyers for decades haha. US trade laws seems much more restrictive here.

Anyhow, tons of business mechanics, right to repair laws, and international trade laws are probably involved. Apple's anti-right to repair results in expensive battery+screen service cost (by them, even batteries are married to logic board these days and aftermarket methods and programmers become more and more challenging/less cost effective) but its still just a $1k asset. Tesla disabling right-to-repair for post warranty repair on a premium car asset is much more painful financially. Car value will drop like a rock post warranty. Probably same or worse than European premium cars (20%+ original price post warranty). Add in the periodically gov financial incentives on EVs that depresses used EV prices, post warranty price shoots below cheap subcompacts of the same vintage haha.

Usually depends on who owns the tooling along with master supply contract agrements.
If the OEM owns/paid for the tooling they usually dictate where the parts go. If the tier one supplier owns/paid for the tooling they have more flexibility on selling versions in different/aftermarket sales channels. Obviously usually without OEM branding - which is why you sometimes see aftermarket parts with OEM branding ground off.

Some OEM's like to pay for all their tooling - keeping the supply chain locked down, others are more open - and let the tier one supplier take the upfront costs, with the benefit of some aftermarket sales revenue.

[howardc64]

DELETED