Halfshaft Bolt Answers (all threads merged)

[BlackICE]

Gimbal, the design looks promising. But I have a few concerns.

1. How difficult would it be to drill and tap the additional screw holes in the axle.
2. Would the extra holes weaking the axle enough so that it may fail.
3. Machining the screw holes on the perimeter of the flange to match the spines may be difficult. We don't know if the axle's splines were machined in the same location relative to the 2 axle's screw holes.

Note if the axle is destroyed in the process the transaxle will have to be shipped to the UK! Proceed with extreme caution.

 

[Shadowman]

Gimbal, the design looks promising. But I have a few concerns.

1. How difficult would it be to drill and tap the additional screw holes in the axle.
2. Would the extra holes weaking the axle enough so that it may fail.
3. Machining the screw holes on the perimeter of the flange to match the spines may be difficult. We don't know if the axle's splines were machined in the same location relative to the 2 axle's screw holes.

Note if the axle is destroyed in the process the transaxle will have to be shipped to the UK! Proceed with extreme caution.

In his design he has the additional holes being drilled and tapped into the coupler as well as the axles however this was discussed with Jay as well. For example the OEM coupler is cast and I do not know how it would respond to drilling and tapping. If the washer is be secured to the coupler then the bolt will take all the deflection stress. As for the drilling of the axle; this is an exercise that I will not even consider as the R&R process and associated risk is far to extreme.

After looking at mine I remain somewhat convinced that the problem has to do with the random act of sloppy spline fit to the coupler when compared to anything else as such I remain on the fence whether this approach would be another patch rather than a long term resolve.

BTW there is a laser etched number on the inside of the coupler that someone should be able to associate it with something; of value is any body's guess.

Just my feelings at this moment.

Shadowman

 

[Gimbal]

Dang good point on the clocking of the holes relative to the splines. I did not think of that. I'll have to think of a solution.
Drilling and tapping the axle is an unknown at this point. As long as the axle is not too hard it should be possible. I don't think it will weaken the axle very much.
If I destroy the axle I will then have to replace it with a new one and maybe get one with one large thread in the middle, like an M12 or so!
And yes Shadowman I also wonder how different the loading will be on the axle bolts. Analogdesigner was going to play around with some FEA on this design, maybe that will shed some light on some issues.

 

[nota4re]

Gimbal, et. al,

The fundamental reason that our axles have two bolts instead of one in that Ricardo elected to keep the centerbore locator for all of the machining processes performed on the shaft. A reasonable question, therefore, is why not save the drilling/tapping of the shaft until the last step - and do it with a single tapped hole in the center. I believe the answer to that is question is that the engineers felt it was impractical to do the drill/tap step downstream of the hardening steps. My uneducated opinion is that trying to secure a "floating" flange with additional fasteners is like more lipstick on the pig. The fasteners that are there would be sufficient, IMO, if they could eliminate the "float" of the flange and therefore provide a seating surface for the backside of the washer. This would dramatically improve the contact patch and also remove the load from the existing bolts.

I think Jay is on track to try to implement such a resolution. I hope he is successful and he can no doubt sleep better with such a "fix" in place.

However, this is a very difficult fix to try to distribute to others and there's a huge amount of liability to be assumed. In attempting to distribute to others, the tolerances of the existing seating surfaces (the existing combined with elimination of the washer gap) must, by definition, be very tight. A one size fits all approach is likely not in the cards. As soon as we acknowledge this, then there's a dependency on the competency and thoroughness of the installer. Done improperly, there could still be a failure. Who's responsible?

And speaking of responsibility - as soon as non-Ford-approved corrections are made, Ford is somewhat off the hook. If, God forbid, there's a catostrophic failure with a third party solution - it could get very ugly.

This is the reason that I was willing to describe the details of what a solution could be. Those owners with the skills, connections, or means to implement their own solution are free to do so. In terms of a business proposition of offering it to others - I guess as far as Cool Tech is concerned, I'd be confident to do that as soon as my test vehicles had an accumulated mileage of maybe 200,000 miles or so with no failures. I'll be back to you when this happens :willy

 

[usmcfred]

Haste Makes Waste??

Was hastey and non-conventional development the root cause of the design problem(s) that have resulted in a number of our half shafts falling out? After reading a Society of Automotive Engineers (SAE) paper entitled The Ford Transaxle-Tailor Made in 2 Years by Glenn D. Miller of Ford and Andrew Cooper, Bob Janczak and Steve Nesbitt of Ricardo UK, I would emphatically say YES.

Let me quote directly from this paper SAE #2004-01-1260. In paragraph 2 under the OBJECTIVE section "... the two year timeline prohibited normal development procedures, while the limited number of vehicle builds and prototypes restricted the amount of testing and data acquisition available. Instead, specially developed proprietary software and advanced Computer Aided Engineering (CAE) techniques were used to predict mechanical and dynamic performance, and address assembly issues in advance of component manufacture. Vehicle usage during development was therefore limited to proving analytical calculations and assumptions, while demonstrating functionality."

In short, the design relied almost entirely on CAE in conjunction with calculations and assumptions, and appears to have almost totally lacked any significant emperical testing in prototypes before its release to production at Ricardo. This is yet another indication of the likelyhood of a flawed design and/or tolerances that ultimately lead to failure.

Clearly, Ford and Ricardo engineering must have had ample time by now to thoroughly analyze the problem and develop a fix that can be carried out in the field on ALL FGT's. Ford needs to do the right thing by issuing a TSB and recall so that we can have peace of mind and no longer be afraid to drive these great cars at night and/or long distances fearing that we will be stranded in a potentially dangerous situation, together with the hassle and inconvenience of getting an immobile GT to an unfamiliar Ford dealer. usmcfred

 

[AJK]

Without disconnecting the driven axle at the coupler can one feel/see this spline play and what would be the procedure to test?

 

[BlackICE]

Clearly, Ford and Ricardo engineering must have had ample time by now to thoroughly analyze the problem and develop a fix that can be carried out in the field on ALL FGT's. Ford needs to do the right thing by issuing a TSB and recall...

Yes, I believe Ford understands the problem and also how much it would cost to fix. The huge cost, combined with the fact that not all cars are affected by the issue has slowed Ford's response to the problem.

I have heard of 05 cars with more than 35K on them, without any problems yet. Also cars with less than 3000 miles that failed. Clearly the design, material and or, tolerances should have been better.

Remember by 2009 most of the cars will be off warrantee and Ford can forget about the issue. I don't think Ford will say anything offical unless the NTSB believes there is a safety issue, or some l-wy-rs force Ford into action.

I am surprised and thankful that so far the failure has not cause any injuries or death. I have to believe than keeping the car in control at high speed with a disconnected half-shaft, would be very hard. Just imagine if someone was doing the TX mile and when up shifting the axle disconnect at 170 mph! The axle would be flying around the engine compartment at about 200 rpm!

 

[SteveA]

I have heard of 05 cars with more than 35K on them, without any problems yet. Also cars with less than 3000 miles that failed. Clearly the design, material and or, tolerances should have been better.




I did take notice that the 05 #0015 car of Stillen Racing just ran the Targa New Zealand logging over 1700+ flat-out wide-open miles without a failure. The driver of the car was a 24hr de Le Mans winner so I don't think the GT was babied. I also confirmed this car was a test mule and not a official production model showing a total of 22000+ miles. Its halfshaft bolts look stock and unchanged.

I must confess I'm as confused as ever as what to do. I'm leaning towards leaving a "happy part" alone unless Ford tells me to do otherwise.
However obviously there is too some degree a problem.

 

[Empty Pockets]

I'm with you, Steve.
A lot of guys have decided to replace 'em as a precaution figuring it's the way to go to be safe. Well, fine...'nothing wrong with doing that if it will promote one's mental health. But, as was stated before on this thread somewhere, "Prowler's"(?) replacements (and I think they were'wired too) also broke.
So again, as I see it, what's the point of doing it unless the 'bolts break first?
You can betcher boots Ford's working on it -AS ARE A LOT OF OTHER FOLKS.
So, when a real "cure" is found that we can be pretty sure is a, "do it once", fix - that's the one I'll go with whether mine are "fine" at the time or NOT!:lol
(In the meantime, I'm assuming MINE is wunna those 35,000 mile, trouble free cars!)

 

[Shadowman]

I'm with you, Steve.
A lot of guys have decided to replace 'em as a precaution figuring it's the way to go to be safe. Well, fine...'nothing wrong with doing that if it will promote one's mental health. But, as was stated before on this thread somewhere, "Prowler's"(?) replacements (and I think they were'wired too) also broke.
So again, as I see it, what's the point of doing it unless the 'bolts break first?
You can betcher boots Ford's working on it -AS ARE A LOT OF OTHER FOLKS.
So, when a real "cure" is found that we can be pretty sure is a, "do it once", fix - that's the one I'll go with whether mine are "fine" at the time or NOT!:lol

I agree that the issue is real but also very limited

However having shared this and also having removed mine I still suggest that folks spend the few dollars and have them inspected and then replaced by someone that remains methodical throughout the process. I found some of the bolts very tight and others were loose and then add to this one of the axle splines was leaking oil as such I will remedy at the same time.

For many it will be an intersting exercise at best however for most it will promote some peace of mind.

Just my thoughts

Shadowman

 

[Gulf GT]

I went over the Jay's (analogdesigner) tonight to get a jack pad, and we talked about this for quite sometime. He had both the Ford replacement bolt set and the new ARP set from Accufab. Big difference! The washer looks at least 1mm thicker, and the design of the bolt heads are much better not to mention better materials on both. It may not be a permanent fix, but it's definitely way better than Ford's option. Who knows, if you don't happen to have a lot of spline play, it might do the trick. I have an Accufab set on the way, and plan to install it anyway as a precaution until something better comes along. I also have a replacement bolt set from Ford, and I think that will become a part of memorabilia collection!

On a side note, Jay is very nice guy, very helpful, and obviously very knowledgeable. I enjoyed our conversation. He's got some great products and ideas! If you ever get a chance to meet him, you'll like him. Down to earth car crazy perfectionist nutcase, just like the rest of us.

 

[Gimbal]

Based on the many good feedbacks lately I have come up with another idea. I'm trying to think outside the "box". The idea is to have a washer that can flex but does not premanantly deform. I know of a design where the teeth of a sprag clutch were breaking, they tried harder and harder steel and finally went to a softer steel. These softer teeth lasted much longer than the hard teeth because they were allowed to flex and therefore reduce stress.
Any thoughts?

 

[BlackICE]

Gimbal I like your 1st design combined with Notra4re/Jay's idea.

How about a your new thicker hardened washer design without the perimeter screws, using only the original 2 bolt holes in the axle. The washers would be initially made oversized in thickness. Now the tricky part, how to measure the correct thickness of the finished washer so that you get the right preload with the washer bottomed out on the axle at the correct bolt torque without any bending distortion. Once the correct thickness is determine, you would machine the washers to the finished thickness.

I would guess a 1 to 3 thou. thickness undersize would get the correct preload. But I will defer to the experts. I think this design would minimize the shear stress on the bolts, while maintaining high bolt tension.

Jay what do you think? Is it worth running a FEA on it?

Questions:

1. Are flat head screws in the correct size and material available?

2. Is the flat head design strong enough for the correct torque value without destroying the heads?

3. How to measure the correct washer thickness on each axles?

4. How to machine the washers to their final size?

 

[Gimbal]

Hi BlackIce, I'll make another layout tonite incorporating some of your ideas and Jays. Since every axle is going to be different enough I think the only way is to shim each one individually rather than custom machine the washer each time. Too many tolerance stack-ups on the axle, flange and washer.

1) There are flat heads available in size and material
2) The torque spec is not that high, either socket head flat heads or phillips would be fine.
3) Some analytical calcs should get us close enough to the right washer material, heat treat and thickness for proper preload.
4) Probably shimming the washer rather than machining to a final thickness. For myself I could machine the two I need but it would not be practical for everyone.

 

[BlackICE]

Hi BlackIce, I'll make another layout tonite incorporating some of your ideas and Jays. Since every axle is going to be different enough I think the only way is to shim each one individually rather than custom machine the washer each time. Too many tolerance stack-ups on the axle, flange and washer.

1) There are flat heads available in size and material
2) The torque spec is not that high, either socket head flat heads or phillips would be fine.
3) Some analytical calcs should get us close enough to the right washer material, heat treat and thickness for proper preload.
4) Probably shimming the washer rather than machining to a final thickness. For myself I could machine the two I need but it would not be practical for everyone.

I guess a range of pre-made washer is not practical, since the range maybe +-0.010 that would require 10 sizes if you wanted to be within 0.002".

What would be the easiest way to measure the size of the shim needed, without any real fancy tools?

Yikes! No Phillips they tend to strip easily! Torx, Hex, or 12 point is better. How do you get 20 ft-lb on a Phillips?

 

[Gimbal]

Using a feeler gauge you can measure the correct gap during the shimming operation. The right gap still needs to be determined. I'm thinking .002-.005 or so. The design could be made so that a +/-.010 gap will work.

I'll check on the torque for a phillips, a torx could be used. I'll try and see if I can easily get 20 ft-lbs on one or not.

 

[Gimbal]

OK, here is another 2 layout options. The idea is to not need any shimming. The washer flexes as needed while still applying a load to the flange. The sureface where the washer touches the flange may need machining flat but maybe not. The idea is to allow some flange movement a couple thou or so reducing stress to the bolts.

 

[AZGT]

Has the question been answered if they are snapping from stress or shearing due to play after becoming loose?

 

[Gimbal]

Hi AZGT,
No one has come forward with any broken bolts to analyze. Regardless, something is breaking the bolts.
1) They were initially tightened too much
2) The washer was too soft or too hard
3) The Flange hub spline fits too loose
4) Bolts too weak to begin with
5) If the bolts are fatiguing its probably due to the loose spline fit or due to the bending of the washer, either from the soft or hard washer.
6) I don't see a large load axially on the bolts, so I feel they are not over loaded due to the design. The CV joint should prevent that from happening generally speaking.
7) During a large power load from the engine thru the tranny axle I could see a temporary large load placed on the bolts. If this happens too often the bolts could snap mainly because of the loose spline fit and/or the washer bending.

Who knows what the fix is? I don't, but I do think a couple of things could be tried. Analogdesinger is doing some FEA on a couple of options, maybe he will show something with promise that is relatively do-able for all of us. I would be more than happy to get something special made and try out if it seems promising.

We all are still not out of the woods on this problem.

Any new news from anyone on this? Any Ford news?

Thanks a lot,
Gimbal

 

[analogdesigner]

Hi AZGT,
No one has come forward with any broken bolts to analyze. Regardless, something is breaking the bolts.
1) They were initially tightened too much
2) The washer was too soft or too hard
3) The Flange hub spline fits too loose
4) Bolts too weak to begin with
5) If the bolts are fatiguing its probably due to the loose spline fit or due to the bending of the washer, either from the soft or hard washer.
6) I don't see a large load axially on the bolts, so I feel they are not over loaded due to the design. The CV joint should prevent that from happening generally speaking.
7) During a large power load from the engine thru the tranny axle I could see a temporary large load placed on the bolts. If this happens too often the bolts could snap mainly because of the loose spline fit and/or the washer bending.

Who knows what the fix is? I don't, but I do think a couple of things could be tried. Analogdesinger is doing some FEA on a couple of options, maybe he will show something with promise that is relatively do-able for all of us. I would be more than happy to get something special made and try out if it seems promising.

We all are still not out of the woods on this problem.

Any new news from anyone on this? Any Ford news?

Thanks a lot,
Gimbal

All,

Gimbal was nice enough to supply me with copies of his drawings so we examine stresses within the various configurations. Some personal issues has delayed my FEA efforts for a few more days.

Thanks,

Jay