Where is a 10mm bolt used in in the 308 Transaxle?

Bolts are now available from a couple of reliable sources:

Hill Engineering/CustomFerrariParts:
Received eMAIL from 348paul (Paul Hill) with a price. He asked around was told by one of his larger customers that they hadn't needed one in 20 years. So he can't justify a larg run which would get the cost down. He was kind enough to offer to do a small run, mostly for me, at a price that was a lot better than I was getting from the US bolt companies tho.

Ferrari.uk
Malcom West posted that Ferrari.uk unexpectedly received a supply of them that they'd ordered early last fall & weren't scheduled to be delivered for a couple of more months. I wrote Malcolm suggesting Ferrari.uk consider Hill Engineering as a more responsive & reliable supplier.

CUT MY OWN:
Lastly, I received the long class 12.9 10mm cap screws from McMaster Carr. Found that even tho it's hardened, I can cut it with HSS tooling. So I can cut the threads with my lathe's carbide threading tool. This will be the 1st time I've ever cut threads, so there'd be some learning involved.

However, research indicates that rolled threads have 10% to 30% higher tensile strength than equivalent cut threads. However, both would meet the specs for the same class of bolt.

Think I'll make one and then sacrifice it by over torqueing it until I get to 95 or 100 N-m, or it breaks. If it passes the over-torque test, & I'm confident I can reliably make 8 of them, then I'll make them, otherwise will buy them.

 

Ahhh, I see. THANKS!
Each thread would stretch less for the same total elongation. Was already planning on extending the threads 6 or 7 mm into the clamped section like the original & tbakowski's bolts were made(see pix in my prior post). Does that sound like enough add'l threads?

 

Well, Spent 3 hours last night setting up the lathe (~1 hr) & cutting threads (~2hr)into 1 socket head cap screw. Took many passes with the lathe, cutting the threads.

Was advancing the cutting tool about 0.06 to 0.08 mm deeper each pass. Seemed like I'd make a couple of passes with little or no material removal, then the next pass would take a spiral of metal off the length of the thread.

To avoid breaking the tool, I would kill power & turn the lathe head by hand to cut the last couple of threads for each pass. That way the tool always stopped at the same place. After 4 or 5 passes there was a definite rise the tool would bump up against.

Under a magnifying glass the threads had a rough texture to the surface. Dodn't have a 10x1 die, so I ran the crown end of one of the nuts onto the threads. Noticably ceaned them up & smoothed them. The slots in the crown had some fine bits of metal in them & the nut's threads don't appear to have changed..

Threads still have some roughness to them tho. Think it's the nature of the metal, just not very free machining. Read somewhere that alloys good for rolled threads are hard to machine, & vice versa. Think it's true.

Will do the over-torque test sometime in the next couple of days.

ham308, Not taking, will be real happy if I can get to 105N-m! Altho frictional resistance from the rough threads may increase the torque at a lower amount of elongation.

At 2 hrs/bolt to cut the threads tho, I'm starting to ask myself how much my time is worth!!! Wish I had a heavier lathe or a thread grinder attachment w/ cooling! Hmm, wonder if I could just buy a thread grinder wheel & rig it up somehow?

 

A big piece driving this effort to replace all 8 bolts is a concern that the other bolts may also be on the way to failure. Given that they're not being torqued to their yield points, I've strongly suspected that odds are they're OK, BUT haven't been willing to take the risk.

Since the alternatives are coming down to either investing a day or 2 cutting threads in my own bolts, or else spending several hundred $, I have been racking my brain for a way to increase my confidence in the current bolts.

The word Magnaflux has been tickling the back of my mind, but I hadn't heard it mentioned in years. Last time I'd used it had to send the part out to a place offering the service. Anyway, I decided to do a quick web search & see if it was still around. It's alive & well, and apparently has gotten a lot easier to use!

What do you think, If the spotcheck kit doesn't reveal any flaws/cracks in my current bolts, especially around the head, is reusing them a reasonable risk?

Remember, that the bolts are only being torqued to 80 N-m, while the max recommended torque for a class 12.9 10mm fine thread bolt is 104N-m?

I just picked up a Magnaflux Spotcheck Jr. kit on eBAY, it's coming priority mail, so I will have the capability in a couple of days.

re: grinding wheel
Hmm, what I'd really like to find is a 60 degree diamond wheel (Better look thru my misc. stuff to make sure I don't already have one.) I have a small air die grinder that I've been intending to make into a toolpost grinder for a long time.

Hmm, Would need to be able to off-set it's axis at the thread pitch angle from the lathe's centerline. Guess I'd need to build an angle adjustment into the mount!

Maybe all I need is to mount it so it cuts the threads from the top or bottom, then could just use the cross-slide's angle adj to set the pitch angle.

 

I'm still planning the over-torque test. After this much work I want to find out how well it came out.

Since the threads are still observably rough, my machinist friend Ed recommended that I use a pretty good lubricant, maybe even an anti-sieze to overcome the thread roughness. Otherwise torsional effects on the bolt could turn out to be significant. Any result I get would be equivalent to a significantly higher dry torque as was discussed in another thread.

Have a HSS 10x1 die coming, but don't think I'll hold up the over-torque test for it as the nut's crown end did a fair amount of cleanup.

 

I slathered Permatex silver anti-sieze on both sides of the differential thru-hole, the bolt threads, and the nut's threads.

Used a Snap-On 1/2" drive click-type torque wrench.

Initially torqued the nut to spec: 60 ft-lb/81N-m to establish a starting point.

Then increased the torque in 10 ft-lb increments, readjusting the wrench each time it clicked. It seemed to be taking about 1/2 to 2/3 turn of the nut to get to the next 10 ft-lb click.

Well, I torqued & I torqued until suddenly the nut started spinning much more freely on the bolt. This was at just over 1/3 turn past 120 ft-lb. Best guess is failure occurred at ~2x Ferrari's spec:

~125 ft-lb/170 N-m

Sorry ham308 your 105N-m was a tad low, guess you had a couple of digits transposed in your calculations... ;^)

Rember, This was a lubricated test!!! Equivalent 'dry' torque would have been significantly higher! Assuming lubricated torque is 75% of dry torque for the same deformation, the dry torque equivalent would be about 220 N-m!!!! This is over 2x the max recommended torque for a 12.9 10mm fine pitch bolt!

I had to use wrenches to back the nut off. It still required a fair amount of force to get the nut back off. Once I had things apart, I took some close-up pictures (attached).

POST MORTEM RESULTS: NUT FAILED, NOT THE BOLT!!!
The bolt's threads were still pretty intact, altho they seem a bit rounded on one side of the bolt in the region the nut was over while testing. Threads on the other side of the bolt actually look pretty good, they were starting to deform, but no sign of actual thread failure.!

However, the first 4 threads or so inside of the nut were partially stripped, and clearly more damaged than the bolt's threads! These threads will now spin freely on the original bolt that I took the nut from to do the test.

The threads inside of the nut's crown section actually seemm to be in good condition & still functioned, making it hard to remove the nut from the bolt. Best guess is that they simply gave enough to slide over the bolt's threads when the nut's main threads failed.

Oh, yes, the front of the nut was visably deformed outwards, almost like it had been hit with a hammer!! IMHO, the nut is not as hard an alloy as the bolt. Still plenty tough tho!

BOTTOM LINE:
I'm not going to worry about my home-threaded bolts being up to the job after this test.

I'm also no longer highly concerned that the original nuts & bolts have had their threads damaged by the original torqueing. Will take a good look at the threads to be sure tho, but I don't expect to see any deformation. If the magnaflux results also look good, then I'll probably reuse them.
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login
Image Unavailable, Please Login

 

John,
Yes, will use my magnaflux dye kit on each of them as well as looking carefully at the threads before deciding to reuse it.

If all 7 pass magnaflux, then I'm going to conclude that it was just an isolated flawed bolt.

re:"none have stated this is a failure observed before"

When I dropped Steven's boot latch housing over at KTR, I spoke with Jim, their Ferrari expert. Said he's seen a couple of differentials with broken bolts. Didn't have time to chat much, so that's all I learned. Also, Ferrari.uk sells several sets of bolts/year & can't keep up with demand.

Yes, I'm quite pleased with the camera. It does even better than I'd hoped when I bought it! BTW, these pix were made at about 3mp resolution, not the ful 5mp! I then cropped them & had to reduce the resolution by another 10% or so to get to the 640x480 upload limiit.

ham308,
Was the number you were using is the torque value for the bolt's 'yield strength'. This is the point where the tension on the fastener causes it to be permanantly stretched by 0.2% beyond it's elastic limit, & thus will not return to it's original length. I think the max recommended torque is close to producing this tension. (Need to go back & read my Machinery's Handbook again.)

However, yield strength still a way away from the 'Ultimate Strength' which I believe is the tension that the bolt must withstand w/o breaking?

 

Had already thought the weight/balance thing thru. Wasn't worried so didn't say anything.

Which way I go is going to depend a lot on how good the current bolts turn out. If they test out good, then can just go with the bolt I got from tbakowski & weight won't be an issue. Even buying 1 or 2 from ferrari.uk isn't a wallet breaker. This approach is pretty attractive as cutting those threads is over an hour/bolt & most likely I'll end up scrapping 1 or 2 of them.

If I go with the rethreaded socket head cap screw, the weights ought to still be reasonably close. The bolt bodies are going to be identical as they're the same diameter, length & thread. That leaves the difference being the mass of the socket head vs the hex head. Which is going to be pretty close. Actually not sure which would be heavier.

Altho, most likely the only reason I'd go with the socket head is because I can't reuse the current bolts & don't choose to bite the bullet & buy more new ones from ferrari.uk. In which case all 8 would be socket head & again it's a non-issue.

ham308,
Funny thing is that they're from McMaster carr & only about a buck apiece. Got a real lecture from Ed, said that McMaster bolts are like buying stuf from Kmart or Target. OK, but not top shelf. Haven't told him about the test yet. Think he's going to be surprised also.

 

Just pulled 3 bolts (every other one) from the differential, gave them a magnaflux dye penetration test & examined the threads under magnification.

None had cracks, and the threads were in mint condition. Even ran a 10x1 dye over the threads on one & couldn't see any difference.

I'm now convinced that this was an isolated failure & the other bolts will also test out good. Am just going to replace the failed bolt. Even if another 1 or 2 show cracks I'll just replace them. Will use all new nuts since they don't seem to be as tough as the bolts based on my over-torque test.

The end of this saga is finally in sight!

 

I got lucky & picked up the set of differential bolts that were being sold on eBAY for the grand sum of $9.70!

I magnaflux tested them & all are fine. Threads looked fine also, & a die went onto all of them smoothly.

I decided to trust them a more than the bolts that came out of my differential & installed them.

They are conventional black finish bolts, 12.9 markings on the head, & rolled threads.

The bolts originally in my differential have unmarked black heads, but the bodies are appear to have been ground rathar than rolled to get the final OD. Hard to tell for sure, but the threads may have been cut rathar than rolled. One reason for thinking the threads may have been cut is that the bodies have a small diameter hole on the threaded end. Holes like that are often used for a lathe center to support the end while turning or cutting threads.

It was the absence of any grade markings, combined with a broken bolt that worried me the most about the original bolts.

As soon as my Ferrari.uk order with the the replacement differential cover studs & header gaskets come in, & my header comes back from being ceramic coated by Jet-Hot, I'll have my car on the road!

I greatly appreciate the technical help I received from tbakowsky, Kermit, ham308, Malcolm West, & the other responders. Also to Paul Hill for offering to make up bolts for me!

This project ended up being quite a crash course in fastener specifications!