Dana 35 Pinion Seal Replacement

[Juggernaut]

I recently pickup up another TJ with a weeping pinion seal on the dana 35, so I decided to through a $10 seal at it. Full disclosure, I know I'm taking a short cut here by replacing the seal without replacing the crush washer, but I'm trying to preserve a newly paved driveway and I don't have the time to remove the carrier in the immediate future.

I did try to go about this in most scientific hack way possible and before removing the pinion nut I did the following:
- center-punched the relationship between the pinion, nut, and yoke at the pinion's 6 o'clock position.
-used calipers to measure the distance between the top of the pinion and the top of the nut at 9 and 3 o'clock positions in an effort to confirm the nut will be back to the same depth/position. (.22, .21)
-counted the number of exposed threads above the nut (4 threads)
-counted the number of turns required to remove the pinion nut. (14.5 turns)

Here is where things get interesting.

I was curious about the torqued value of the nut so before disturbing the nut from its installed position, using a shitty torque wrench I incrementally increased the setting on the torque wrench until the nut ever so slightly loosened. The value reached was only 35ft-lbs. I took the nut off and reinstalled with a good wrench and it took a value of 35# to set the nut back to its noted relationship with the pinion and yoke per the above. At face value, this torque seems so minimal, but the axle was good, no noises, no play in the yoke, just a very slight weep of the seal. It doesn't exhibit any symptoms of the nut backing off at all in the 2k miles that I've driven it.

My other TJ has a 44 and is the only comparison I have at the moment. The nut looks similar in terms of thread count to the dana 35.

This got me thinking - Is there a scenario where after using a torque of say 160ish ft-lbs to initally compress the crush sleeve, the preload then results in a torque of roughly 35ft-lbs on the pinion nut once the sleeve collapses?

Any one have any related experience with this?

 

[Gilaguy23]

My memory is a tad foggy, but on my 99 TJ it took a lot more than 135 to crush the sleeve. Im 250lbs and used a 3' pipe on the end of a breaker bar basically hanging off it to incrementally get it to the right preload. It sure wasn't that hard to take the nut off by a long shot. I'll admit ive done as few pinion seal the way you describe and never had any problems or leaks. Some will argue its a hack job. My opinion is the 35 is a weak sauce turd anyhow. When/if it craters I'll deal with it properly. Only I didnt see in your post is did you use 271 on the nut??

 

[lumberjacklloyd]

I would be severely concerned if 35 ft lbs loosened my pinion nut. Furthermore, I would be concerned for clamping force on pinion stack retorquing to 35 ft lb.

 

[machoheadgames]

35 sounds way low to me. The initial crush of the crush sleeve takes several hundred pounds. I hammer the hell out of it with my Milwaukee high torque impact to get it to start crushing. Even after that, it still takes quite a bit to get it to crush further and further. Once preload is correct and you stop there, typically the high torque that it takes to continue crushing is enough to keep things tight long term, and in my experience that holds up because it pretty much takes an impact for me to get the nut back off.

All of that is why most people recommending the hack job method (changing seal without sleeve) recommend torquing to a simple 160 ft lbs or so, because it's enough to keep the nut from backing off but not too much to crush the sleeve further.

I would be concerned about only putting 35 ft lbs back into it. Although it is a lock nut so maybe it will hold anyways. But technically you should be able to crank on it a bit to at least over 100 ft lbs without crushing it further. My guess is someone has been in there before and didn't tighten it very tight when they reassembled, and you are just now finding that out by you now needing to do the same job.

 

[Juggernaut]

The plan is to apply loctite to the nut on final assembly but again I was put off by the low torque on the nut so I haven’t completed the job. I agree that the figure sounds way low but there was no issues with the axle that would indicate a problem with preload/bearings. I was apprehensive to increase the torque and potentially create a problem.

 

[lumberjacklloyd]

The plan is to apply loctite to the nut on final assembly but again I was put off by the low torque on the nut so I haven’t completed the job. I agree that the figure sounds way low but there was no issues with the axle that would indicate a problem with preload/bearings. I was apprehensive to increase the torque and potentially create a problem.

There is no way to beat around the bush here. Something is not right.

Does rotational resistance get increase drastically if you torque to 40-50 ft lbs?

 

[Juggernaut]

I haven’t increased the torque and tried, although if it does drastically increase the rotational force required, I’m cooked at that point because there’s no going back.

 

[Juggernaut]

I hate the thought of arbitrarily increasing the torque on the pinion nut, but I don’t feel this is correct and it sounds like you all agree. Thinking out loud, I could measure the current rotational force as assembled at the pinion nut and incrementally increase the torque setting until a change in rotational force is noted? I don’t know where that will ultimately land for a torque value on the nut, but is there a general consensus of a value (75-100#) that is tight enough provided it doesn’t increase the force required to rotate?

 

[machoheadgames]

I hate the thought of arbitrarily increasing the torque on the pinion nut, but I don’t feel this is correct and it sounds like you all agree. Thinking out loud, I could measure the current rotational force as assembled at the pinion nut and incrementally increase the torque setting until a change in rotational force is noted? I don’t know where that will ultimately land for a torque value on the nut, but is there a general consensus of a value (75-100#) that is tight enough provided it doesn’t increase the force required to rotate?

I really think somebody probably went in there before you and did not tighten it back up properly. In all reality, 35 ft lbs worked because at that point the bearings are seated against the crush sleeve and are running the same whether they are torqued by 35 ft lbs or 200 ft lbs, provided the crush sleeve does not crush further under the higher value. The only variable is at what point the nut will back off, which is why 35 ft lbs scares me, even if it was running fine that way. Especially if you used the old nut and not a new one, the locking function of the nut is further reduced if reused.

No crush sleeve I've ever seen would continue to crush without at least probably 220 or so ft lbs. The few that I've done have all been a major pain in the ass, so much so that I pretty much required an impact because underneath the Jeep I can't develop the torque to crush it (or have the room to work the long breaker bars), and with the axle out it wants to flop around under that much torque.

 

[Juggernaut]

Thanks for the sanity check everyone. I think I’m going to get a new nut, gradually torque it while checking rotational drag. Hopefully get it to 160-180# with loctite. I’ll keep and eye on it and redo the crush sleeve after the holidays.

 

[JMT]

Thanks for the sanity check everyone. I think I’m going to get a new nut, gradually torque it while checking rotational drag. Hopefully get it to 160-180# with loctite. I’ll keep and eye on it and redo the crush sleeve after the holidays.

I did the same thing with a Dana 30. Just returned it to the same preload. Was also surprised it didn’t take tons of torque, but it’s been fine for 20,000 miles. My goal was preserving the same rotational torque.

I presume the crush washer is crushed and at this point just a little more will restore the torque value. I also presume it will change the rotational torque

 

[Juggernaut]

Just to close out this thread, everyone’s instinct was spot on and the advice was solid.
I had the opportunity to speak with Tom Woods Shafts so I asked them as well.
Their advice and apparently a fairly common practice is torque to 160ft-lbs and red loctite the nut.

Thanks again everyone.