LSD Bolts - Question

Non-engine, eg: aerodynamics, gearboxes, brakes, suspension
Post Reply
TR-Spider
Posts: 132
Joined: June 23rd, 2006, 8:37 am
Location: Rekingen / Switzerland
Contact:

LSD Bolts - Question

Post by TR-Spider »

I do have a small question:
the flywheel bolts on the Fiat 1600TC do have M 10*1.25 treads (according to my books).

Could somebody please tell me if these bolts are of 12.9 grade and how long they are (bolt length without head).

The reason I am asking is that I am just finshing my LSD-diff conversion...and can't find M 10*1.25 bolts with 25 (or 30) mm length with 12.9 grade to fix my crownwheel...not funny really.

The original bolts are 20mm long with 8mm flange thickness, new flange is 14 mm thick, so 6 mm tread lenght will be too short. 12.9 grade seems a must due to the high torqe setting of 98Nm.

Thanks,
Thomas
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

Hello Thomas,

yes the Fiat TC 105TC type fw bolts are certainly 12.9 grade and the underhead length is 23mm. 62lbf ft is the OE torque setting for them.

I hope this is helpful - and by the way - thank you very much for generously assisting my client Dave Storey recently with conversion of Argenta sump to 'TR' big-wing spec.

GC
TR-Spider
Posts: 132
Joined: June 23rd, 2006, 8:37 am
Location: Rekingen / Switzerland
Contact:

Post by TR-Spider »

Hello Guy

this is very helpfull indeed, thanks!
It was a critical moment, no place on earth seems to sell M10*1.25 bolts with 12.9 grade.
I will now use the flywheel bolts and reduce the flange thickness locally to 13 mm, 10mm thread depth should be enough.

I will do a post on the LSD-conversion, when its finished.
With the help on Argenta sump conversion - my pleasure.

Thomas
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

The thread penetration depends on the tensile strength of the parent material. As a rule of thumb one would use

1 x D = 10mm engaged thread minimum (known in British Standards as 'short')
1.5 x D = 15mm (medium)
2 x D = 20mm (long)

Threads are not solid things, they deflect and distort even in normal duty. The shear stress loading on the teeth of the thread is by no means uniform even with 'perfect' threadforms. The highest loaded threads are always the first 5 from the joint and they would tend to be subjected to shear stress well above the acceptable level, but owing to the deformation of thread contours and the beneficial load sharing effect that takes place, materials like cast iron which are weak in shear (here of course 'thread shear') can survive higher loads if the engagement is made longer. Steel of almost any kind has twice the shear strength of CI (for comparative purposes) and at 10mm fine you've got 8 engaged threads (approx) and you should be fine with that. I might be tempted to use Loctite on those bolts because of their critical location.

GC
TR-Spider
Posts: 132
Joined: June 23rd, 2006, 8:37 am
Location: Rekingen / Switzerland
Contact:

Post by TR-Spider »

Yes, I am tempted to use Loctite, too.

Due to the inavailability of of other bolts, I have to do changes the parts again, see scetch.

As I see it, the torque is transmitted from the crownwheel to the centerbody by the surface pressure between them, caused from the bolts.
The cover flange only has to transmit bending force between the outside bearings, no torque (well, from bearing friction...).

What is your engineering-feel about lowering the bolt surface by 2.5 mm as shown red in the scetch? This would give me 11.5mm or 9 turns then. Cover flange material is C45 steel. Or better lowering the bolt surface only by 1mm?

Thomas
Attachments
Diff-Cover.jpg
Diff-Cover.jpg (126.43 KiB) Viewed 9600 times
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

How many bolts and what is the applied torque?

GC
TR-Spider
Posts: 132
Joined: June 23rd, 2006, 8:37 am
Location: Rekingen / Switzerland
Contact:

Post by TR-Spider »

There are 10 bolts torqued with 98Nm.
Engine is 2 liter FI, approx 150 hp.
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

The main load is likely to be the cyclic shear stress on the bolt shank I think, unless you tell me there is a high bolt axial load.
The applied end load at 68 lf ft is 33kN approx allowing for 20% inaccuracy between applied torque and induced strain and assuming k factor or 0.3 with lubrication with Loctite during assembly. Assuming the crownwheel is made of the 'right stuff' which I'm sure it will be being a gear, this is probably plenty strong enough at 10mm penetration with such a rigid joint. I have done some calcs on bolt shank shear based on max achieveable mean (FW) torque and bolt pcd of 100mm and there seems no problem there with 10 bolts, just check the maths for me, I don't get a lot of spare time to do this.


Mean max 2 liter torque = 150 lbf ft
= 200 Nm
Peak cyclic torque x 3 = 600Nm

Torque = 600Nm
Thus shear force = 600/0.05
= 12000N
Shear force per bolt = 1200N
Shear stress, force/shear area = 1200/66.5
=18 N/sq mm (Mpa)
Safety factor 3 x 1 shear stress = 54Mpa

Endurance strength typical = 400Mpa for 12.9 grade steel bolt

Cyclic shear stress on bolt shank is well below endurance limit.

I personally would have no concerns about this holding up. However (and I imagine you know this) all those sharp corners have to go - major torsional stress concentrations.

GC
TR-Spider
Posts: 132
Joined: June 23rd, 2006, 8:37 am
Location: Rekingen / Switzerland
Contact:

Post by TR-Spider »

If we take your calculation of shear stress at the flywheel and transfer it to the crownwheel, I think we have to increase the torque by the gearbox rate and the diff rate. Which would mean for 1. gear 200Nm*3.6*4.1=2960Nm. Hm...

Using your calculculation formulas I get the following shear stresses:
1. gear 617 Mpa
2. gear 349 Mpa
3. gear 232 Mpa
4. gear 171 Mpa
5. gear 149 Mpa

That's the time I wish I wouldn't have started thinking about it...
But of course that is the pure shear stress calculated including a factor of 3 for safety. In reality the friction between crownwheel and flage should do its job reducing the shear significantly due to the clamping forces of the bolts, just shows how important the proper torqening of the bolts is...and another good reason to use Loctite.

I will take care for the corners to get proper radii.

Thanks for your help
Thomas
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

Ah yes, pity me, I'd forgotten the gear multiplication. Multiply everything up and see what it does to the stress vs endurance limit at k factor 3 and let me see.
We could factor in friction 'clamping force' and do full joint stiffness calcs I guess but have no chance at present, would take all day to plot it all out dynamically.


G
Last edited by Guy Croft on March 9th, 2007, 6:30 am, edited 1 time in total.
Testament
Posts: 101
Joined: June 22nd, 2006, 7:47 pm
Location: Taupo, New Zealand
Contact:

Post by Testament »

It might pay to do a simple calculation for the axles, to see what the weakest link is, if they are weaker than your diff-crownwheel bolts it makes it pretty simple.
TR-Spider
Posts: 132
Joined: June 23rd, 2006, 8:37 am
Location: Rekingen / Switzerland
Contact:

Post by TR-Spider »

It can be really painfull looking into your 15 year old school scripts and they just show you: yes, you did learn all about bolts and flanges and you have been able do calculate all of that...

A very rough estimation the transmittable torque on the crownwheel/flange says 2400 Nm, and yes...the halfshaft would not widthstand that torque ...nice.

So I take a conservative approach saying: the the Fiat 132 rear axle was designed carefully to the load both for crownwheel/bolts and halfshafts by the factory (and is said to be the toughest of the Fiat lot). The changes are not that big, so I will put things together with the 12.9 flywheel bolts, use OE torque ratings and Locktite...and drive it.

I will let you know...
Thomas
Guy Croft
Site Admin
Posts: 5039
Joined: June 18th, 2006, 9:31 am
Location: Bedford, UK
Contact:

Post by Guy Croft »

A bit of 'empirical' testing eh, Thomas, what?!

Works every time!

GC
Post Reply

Who is online

Users browsing this forum: No registered users and 37 guests