BMW diff removal

[ChrisC]

My point to "Billy.Whizz" is there are a lot of cars out there with the carbon plates which Caterham no longer use.  Only Caterham know for certain how many cars, but it could be all R models from 2016 to 2018(ish). This will look like there is a problem with the part being used, but Caterham had already addressed this, mainly because they are starting to fail now in numbers. 

[ScottR400D]

You may be right. According to RRT the carbon plates were fine for a period but then something changed. As I’ve posted before, he thought it was a change in adhesive (?) or something. 

They changed to sintered, because of this, and as far as I can understand no one fits carbon now. 

My diff was made in 2014, so it was before 2016 that there was an issue. The problem has been known about for some time so 2018 could be right for the change to sintered. 

Thing is I don’t think my current issue is anything to do with the LSD itself, though I may well be proved wrong.  

[John Vine]

Ref #100:

It was Phil at RRT told me the engines were canted...

He said the same to me when I was discussing a refurb of my Ford diff -- something like 1.5 deg IIRC. 

JV

[Tony Leigh]

Peter,

Just catching up on this post, missed a few days and can see there have been lots of comments. If it would help I could pop over with my 2019 420R (so presumably sintered plates in LSD) so you can listen to the noise it makes and take any measurements for comparison. It has Motul SAE 75w140 oil in it as supplied by CC for build, and my car has now done 900 miles and I am pretty sure it doesn't sound like yours, although I do wear ear protection when driving.

[ScottR400D]

Hi Tony. 

Yes, that would be great. I suspect yours will be more like mine used to be!

My daughter and grandchildren are here for a few days, I’ll message you when they’ve gone and we’ll see what we can arrange. 

Thanks for the offer! 

[ScottR400D]

JV, I’ve asked CC to confirm the engine cant and asked if the diff is similarly angled, as it should be, and if not why not. 

I will let everyone know what the answer is. 

[ScottR400D]

Here’s my answer from CC. 

“Answer: Thanks for your e-mail. I have no idea if there is a slight slant on the engine but I don't think so, having built a couple I can't see that is true as the propshaft is aligned with the differential etc. LSD's do need servicing and sometimes the plates changing, early cars had carbon plates which although quieter would wear out quicker than sintered ones. Hope this was useful. Regards”

So that’s helpful. 

[Nigel B]

I have a 2017 420R with 8000 miles on the clock. It does have a ‘chattering whine’ when I lift off the throttle and there is a bit of a whine on a constant, level throttle opening, that wasn’t there when I picked up the car last year, (at 5000 miles) so I think the diff is becoming noisier with the additional miles I’m covering. 

That said, at present it has, to my ear an ‘interesting, motorsport-ish’ sort of sound. But it may well become annoying with further miles, especially considering it has become noisier over the miles I’ve covered. It has a titan diff, but I have no idea if it has carbon or sintered plates. 

Caterham’s 420R demonstrator at Crawley was a fair bit noisier though, with a low speed 'clunking' that my car doesn't have. 

I hope that helps & I hope you get it sorted.

Nigel.

[ChrisC]

It might be worth doing a quick spanner check on the diff and the diff carrier, if you haven't already, because it's not uncommon for the bolts to work a little lose over time. 

[Nigel B]

Thanks for the suggestion, I'll add that to the jobs list!

[ScottR400D]

Did a quick check with an inclinometer. The diff slopes downwards 1.7 degrees more than the engine does. The engine appears to have a 2 degree forward cant relative to the top chassis rail. 

I’ve no idea if that’s enough of a difference to cause an issue but when I’ve time I will slacken the front of the diff and see if it will lift at all. 

[Shortshift]

So when you say that, you mean that the engine is 'nose down' by 2 degrees and the diff is also 'nose down' by 1.7 degrees?  If so, the delta is only 0.3 degress which is probably within your measurement accuracy tolerance...  Those look good figures to me.

By the way - was that measured by using the iphone spirit level/inclinometer/compass app or similar?

James

[ScottR400D]

Sorry James. No, the diff is pointing down by 3.7 the engine 1.9/2 so a difference of 1.7. 

Is that significant? I have no idea. 

I have a digital inclinometer. Bought several years ago for checking camber, castor etc on race cars. Put it on a chassis rail and zero it then lay it, in the same orientation, on parts of the engine that you’d expect to be parallel to the crank etc, and on to the machined blocks on the base of the diff, which are perpendicular to the flange. 

[ECR]

Am I missing something here? Why is non linearity a problem, surely that is what U/J's are there for. Incidentally, I was told that perfect alignment of the propshaft is to be avoided (something to do with wear patterns in the joints).

I wait to be informed ....

[ScottR400D]

Apparently:

Yes, perfect alignment is to be avoided because the UJ bearings take the load in the same place all the time and wear quicker. 

But if there’s an angle on the joint it actually rotates in a way where it goes fast and slow twice in each rev, setting up vibration. (I don’t fully understand but it’s what I’ve read).

If the other joint is at the same angle and the UJs are phased correctly the vibrations are cancelled out. If the angle is different the vibrations cause wear and can excite other parts of the transmission. 

So it’s fine, even desirable, for the prop to have an angle but the flanges in each end should be parallel, or in this case the diff flange should be at right angles to the engine crank/output shaft. The diff should tilt at the same angle as the engine. 

I readily stand to be corrected or to see a better explanation! 

[Shortshift]

Ref #113 - sorry, my bad - I misread #111.  I'd say that a delta of 1.7 degrees is still likely to be insignificant, though I've no verified grounds to say that!  Hopefully someone with more knowledge will come along and help explain what sensible limits are.

Ref #114 - I think, Roger, that the issue here is that a pair of simple UJ's (Hookes joints) are pretty poor performers in terms of maintaining constant angular accelerations when off-centre.  So with one end of a prop fixed at a different angle to the other end, there is going to be cyclic speed variation measurable at the output end of the shaft, compared to the input end, in any condition other that when the end flanges are parallel to each other (a CV joint - as its name implies, is a better performer in this regard).

With the shaft angled but the end flanges parallel, the UJ joint accelerations cancel each other out.  You're probably right about the wear tendency, though - but I think the key point here is having the end flanges parallel, but not necessarily an exact 'straight through' arrangement.

Could well be wrong on all of this, and I know I haven't explained it very well!

James

Edited to add: a one-liner to explain what I took half a page to cover, posted above as I was writing my guff:  "So it’s fine, even desirable, for the prop to have an angle but the flanges in each end should be parallel, or in this case the diff flange should be at right angles to the engine crank/output shaft. The diff should tilt at the same angle as the engine."

[ScottR400D]

To add, I’m not particularly optimistic about the alignment thing, I’m just trying to rule out everything outside of the diff. 

 

[ScottR400D]

Here a little video explains better. Far more extreme of course. 

[ECR]

Yes, that makes some sense. Thanks

[ChrisC]

Every days a school day :-) 

Based on that information there is a load more things to which could introduce driveline noise, like gearbox mounts and engine mounts. 

If It was my car, I would remove the diff in the carrier,  torque the diff in the carrier (off the car),  refit the diff in carrier, replacing the spacing washers.  replace the engine and gearbox mounts.  Basically a full reset.     

[Geoff Brown]

Historically from what can recall from K series on engine, gearbox & other components have always been slightly 'misaligned'. Wether the odd half a degree here or there makes a substantial difference is any ones guess especially where the gearbox to engine input (first motion) shaft is concerned.

As to the metal plate LSD noise - yes they are to a certain degree - my 2013 LSD is 'noisy' until it warms up after about 20 miles. The 'quietest' it has ever been is the present with Castrol Syntrax LSD 70w-140 installed.

[ScottR400D]

#120 Chris. Why do you suggest removing the diff in the carrier, torqueing the diff on the carrier and then refitting it, etc?

I'm not totally averse to doing the work would would like to hear the reasoning behind that suggestion. 

[7 wonders of the world]

I haven't seen the ironmongery close up but I'm wondering if there may be a degree of flex in it, therefore torquing it off the car will get it seated and tight, any clearance can then be accommodated with shims into the chassis mounts.

It is certainly possible to get a fixing to the correct torque without it being correctly seated.

[ScottR400D]

Yes, I can see that and there’s something in the back of my mind about why Caterham supply the diffs already in the frame but I can’t remember what it is. 

I’m beginning to doubt that I’ll get much movement in situ. The bracket to chassis fasteners are through metalastic bushes and the others go through reasonably tight holes. 

 

[ChrisC]

Yer what 7 wonders said.