Transaxle problems


PeteK

GT Owner
Mark II Lifetime
Apr 18, 2014
2,268
Kalama, Free part of WA State
.
 
Last edited:

PeteK

GT Owner
Mark II Lifetime
Apr 18, 2014
2,268
Kalama, Free part of WA State
After all the posts and research, I have decided to use the Accufabracing.com FGT transaxle replacement kit as insurance for the future. Along with their magnetic oil plugs and the intake sleeve stiffener, I believe them to be better than factory parts and a necessary upgrade, keeping the original parts for the next owner.

Cheers, Rick
I thought about the extra magnetic drain plugs, but since the tranny already has a strainer and one magnetic drain plug, I don't see that adding more will do any good. BTW, I had my tranny oil changed last weekend too, and there was some fine fuzz on the drain plug, but otherwise the oil was clear. I saved a sample for lab testing.
 

Indy GT

Yea, I got one...too
Mark IV Lifetime
Jan 14, 2006
2,526
Greenwood, IN
I thought about the extra magnetic drain plugs, but since the tranny already has a strainer and one magnetic drain plug, I don't see that adding more will do any good. BTW, I had my tranny oil changed last weekend too, and there was some fine fuzz on the drain plug, but otherwise the oil was clear. I saved a sample for lab testing.

Pete, I believe Rick is talking about magnetic drain plugs for the engine oil pan (two locations). Not the transaxle.
 

Indy GT

Yea, I got one...too
Mark IV Lifetime
Jan 14, 2006
2,526
Greenwood, IN
Ahhh, I love it--a technical discussion! Indy, let me (as they say in the US Senate) extend and amplify the comments of the distinguished gentleman from Indiana.

Pete, as do I! Hopefully all our Forum owners benefit from our technical banter.

Yes, you are correct, the Ford replacements parts are still available--in fact, that's what Rich put on my car.

Smart move in my opinion. For as easy and not costly a bolt swap-out as this operation is, I just do not think it is worth the expense and aggravation of having a failure somewhere out driving around and having the car rendered undriveable. Add to this the opportunity for towing damage when the tow guy shows up having never seen a car like ours or ever tried to tow one back to a service facility. The towing instructions originally shipped with our car when delivered are quite detailed and explicit. Damage will result if the proper procedure is not followed.

As for the number of failures, I was guessing at 50% from the number of comments on the Forum, I could be way off. However, what Jamal said at the R2 in 2007 did not account for the failures since, so his statement based on testing at that time almost certainly is way low. Whatever the real number, so many Forum members have experienced the problem that it's something to be on guard against.

As an engineer, I am sure you are adverse to ever “guessing” at something when you can collect DATA. Engineers love data which is then used to develop a “plan” whether a new design (i.e. program test mules) or a redesign to an existing product due to failure analysis. Jamal is an engineer and was the 2005-2006 FGT Program Manager. Ford recognized in the 05-07 timeframe the unexpected failure rate of the axle bolts was suspiciously high. They developed a test plan to gather data (strain gauge testing of loads, HCF/LCF testing of bolts, evaluation of bolt material properties, etc.) to determine the root cause of WHY these bolts were failing. Knowing this information an appropriate fix could be developed.

What was presented to the owners at the R2 discussion was an overview of the data Ford had collected on the failure mode. No conjecture, guessing or assumptions. True, at the 2007 meeting, the number of failures after that time was not known (how could they, other than to “assume” a failure rate). But it did not matter. The infant mortality issue had been identified by warranty claims, the Ford engineers investigated and determined the root cause of the bolt failures and a “get-well” plan implemented (engineering fix “CA”). The ownership ranks were notified of the recall campaign and our cars got new bolts installed without cost. I and many others participated. If you did not participate for whatever reason, that is your problem. Ford offered (and still does offer the bolts) to make our cars right without cost to us. Hard to criticize anything there.

Of the 4038 FGT units produced I still believe your “guess” of 50% had failures is way too high. Owners experiencing the failure were vocal. Ford listened and fixed the problem. And only recently has this bolt discussion resurfaced likely as ownership transfers occur and new owners are scared into thinking their car might not have been serviced with the new bolt package.

As I noted in another thread on "I Broke my GT" I had the bolts replaced last weekend at John Bailey's gathering by the GTG. The originals were just fine, and it took about 40 ft-lbs to break them loose. Even so, I had them replaced. BUT, if you have significant miles on the car (say 10,000+) at this point, if they haven't failed, they have passed the "infant mortality" part of the failure curve and very likely will never fail. I certainly understand anyone who wants to replace them after reading about all the stories on this Forum, however (that's what I did).

Pete, you joined our owners Forum in 2014. Likely you did not purchase your GT new from Ford. How certain are you that your cars previous owner did not participate in the recall campaign and just forgot to mention this to you. If Rich/Dennis did the swap and said the bolts were OE from Riccaro, I would certainly defer to their assessment as they would definitively know early versus late fix package. If they were original bolts and still maintained their initial preload, you were lucky in my opinion. Ask the GT guys how many half shafts they have opened up with OE bolts still in tact as yours were. I would guess (and I hate doing this) there are very few. But Rich and Dennis have the data.

As for the shaft backing out and not causing damage, I can only imagine that if you're rolling along at highway speed and that part backs out, the other end of the axle is still attached to the rear wheel, which will continue rolling until the car comes to a stop, so I can easily foresee that the axle and CV joint would flop around and bang into other stuff (like the cats). Maybe it won't flop around as much as I expect. But maybe the other guys have been lucky so far.

Again, here you are “imagine” again. I have read with interest since I joined our Forum in 2006 of these failures and talked to the engineers who developed the fix. There have been very few owners who experienced the failure while motoring along. As I recall and it is certainly not scientific, many experienced the failure starting off from a stop or out of the garage. And I vaguely remember a south Florida (?) owner who experienced the failure at low speed (30-45 mph) which left a trail of oil behind the car and made some noise but really did not damage any of the surrounding structures. I just do not recall any sensational failures as you have imagined.

Thanks for the discussion!:thumbsup
 

nota4re

GT Owner
Mark IV Lifetime
Le Mans 2010 Supporter
Feb 15, 2006
4,177
As was later determined by Ford's research and testing, the problem had nothing to do with bolt strength which was offered by the AccuFab kit. The problem was with the bolt electroplating process and the ARP bolts did not have this plating process applied and thus (unknowingly at the time) the ARP bolts supplied in the AccuFab kit were also not susceptible to root cause hydrogen embrittlement.

I too cannot resist jumping into a technical discussion.

Although a flawed electroplating process on the bolts may have been identified as one of the primary causes, it was most certainly only one component of the fix contained in BOTH the Ford and Accufab Kit. Specifically, a larger, much thicker and hardened spacer is used in both kits. There is, unfortunately, some tolerance in the cars when you look at how the half-shaft coupler is attached (via the subject two bolts) to the splined output shaft of the transaxle. Specifically, the coupler sits a few thousandths off of the end plane of the output shaft. When the two OEM bolts and spacer are attached and torqued, it was possible for this spacer to cup. (When the OEM spacer is removed, the cupping is quite pronounced and easily visible.) The result of this cupped spacer is that the torque on the bolts is not symmetrical across the head of the bolts. The inside edge of the bolt, for example, may not even be in contact with the spacer. Then you add in rotational forces and loads and the result is that this uneven torque across the head of the bolt shifts and over time and usage this shifting literally un-winds the bolt. The speculation is that this put even more stress on the bolt that stayed in place and the stress (especially when compounded with an embrittlement issue) would pop the head off. At Cool Tech, of the approximate 150 cars or so that we upgraded, when we DID see problems (maybe 3-4% of the cars, max), it was more the case where we had one or both of the bolts backed out (or all the way out) and fewer cases where we had broken bolts. Further, when we had broken bolts, it was ALWAYS the case that the threads were backed out. Long story short - the fix involves the incorporating of a washer of sufficient thickness and harness factor that does not allow it to cup, along with high-grade bolts. Both Accufab (ARP) and Ford also used bolts with a broader shoulder - minimally doubling or more the surface area of the bolt in contact with the new washer.
 

nota4re

GT Owner
Mark IV Lifetime
Le Mans 2010 Supporter
Feb 15, 2006
4,177
To summarize, the kits from both Accufab & Ford resolve the transaxle issue with 3 fundamental improvements;

1. Improved spacer - does not deform when bolts at correct torque values
2. Fasteners with larger shoulders - more than double the contact patch between fastener and the spacer
3. Improved fastener strength - no more embrittlement risks
 

Indy GT

Yea, I got one...too
Mark IV Lifetime
Jan 14, 2006
2,526
Greenwood, IN
Thanks Kendall and always welcome your technical participation. And an appreciation nod to you and your son who have actually changed a significant number of these bolts for our west coast owners. You guys know your bolt history as well. And you have been around on our Forum for quite some time (one month less than I). Hopefully you and PeteK will be coming to our next rally in Detroit. Several of the engineering team members who worked on this issue will be there and we can have a great time discussing the topic with those who theorized the fix and conducted the research.

I defer to what they might tell us but my eight year old notes taken at the Fairmont do indicate early thoughts as to the washer cup participation in increasing bolt head bending as you describe in your post. The Ford team looked at 5 fixes to increase washer stiffness and selected the final thicker washer. The delay in implementing the “fix” was Ford had a difficult time duplicating the failure. They tried overtorqueing the bolts and could not get a failure in 700,000 – 800,000 cycles. They just could not induce a bolt head fatigue failure which the fracture remnants seemed to suggest (failure analysis).

The hydrogen embrittlement scenario came into view (interesting topic for a future round table discussion) and it was learned this was not a good anticorrosion surface treatment for HRC40 bolt hardness. Remember the transaxle assembly, including these halfshaft retaining bolts was completely sourced from Ricaro. They are a very competent transaxle supplier and have deep engineering roots in transmission design. The bolts really were not Ford’s design, but they accepted fix responsibility. In my opinion, this was just a quality escape on the bolt anticorrosion plate callout.

Once the embrittlement scenario was introduced, the theory then was the first (of the pair) bolt would fail due to embrittlement and then the remaining bolt would be subjected to dynamic loads beyond the initial clamp load (now that its sister bolt had failed) and it would fail in fatigue. The Ford team went through a first attempt fix (AA) which had a silver washer. Then to the BA fix with a black washer. Not sure of the washer thicknesses of these two earlier attempts, they may have been thick washers, but the final CA fix did have the thick washer design, not that it was really necessary as no dynamic loading on the bolts if both bolts retained clamp load.

It is my opinion the root cause of the axle bolt issue was the electroplating issue of a hardened steel bolt without a bake cycle leading to hydrogen embrittlement bolt failures. The thick washer, while nice to have, was (again IMO) a second order fix contributor. “Belt and Suspenders” approach, which assures success and now has proven to work fine.

Whewwww. I hope we are now done.....:biggrin
 

RALPHIE

GT Owner
Mar 1, 2007
7,278
...When mine were replaced, 2 of the bolts were cracked and on another, the head had broken off....

... BUT, if you have significant miles on the car (say 10,000+) at this point, if they haven't failed, they have passed the "infant mortality" part of the failure curve and very likely will never fail....

Pete - I disagree with you, from an experience standpoint. My car had 17,604.9 miles when the recall by Ford was performed, and that is when they discovered the 2 cracked bolts and broken head.

Thus, it could have failed at any time after that, and I feel very lucky that I had made it that far.
 

PeteK

GT Owner
Mark II Lifetime
Apr 18, 2014
2,268
Kalama, Free part of WA State
Indy and Ken, thanks for the additional responses. Keep the data coming.

Ralphie, when did you have the recall performed? That's important, because the embrittlement is a function of time, not use (well, maybe if you're using them in an acid environment, but I doubt any of us are). Did you have the recall performed back in 2008-2009 or so? "Now" is 9 or 10 years after the cars were built.

I took a very close look at the parts that came out of my axles last weekend. These are original. The washer is flat with no discernible cupping or wear on the machined surfaces, and the screws are clean with no cracks evident through a 10x jewelers loupe (maybe someday when I have nothing else to do, I'll do a dye penetrant test…). Also, the radius of the screw head to the shank is pretty tight, which probably exacerbates stress concentration and cracking. I didn't do a side-by-side comparison, but it appeared the replacement screws had a softer radius head-to-shank. The ARP screws also appear to have a softer radius from the photos I've seen.

One other thing I noted: I felt for axial looseness of the flange on the transaxle spline. It was noticeable by feel, I'd estimate about .01" of play. No signs of wear, just not a tight fit. Not as tight as, say, a steering wheel on its spline.

I'm curious why Ricardo chose to use two small crews instead of one big one, like I've seen on other CV joint flanges.
 

Specracer

GT Owner
Mark IV Lifetime
Nov 28, 2005
7,081
MA
One additional feature of accufab kit, it can be safety wired, never a bad insurance policy, but not a deal breaker if not done. I have Ford's kit as it was done at the time when Ford was covering it, but carry the accufab kit in the car as a "just in case", but doubt will ever get used. You have these doubts just after you get stranded, but those doubts have long faded.
 

HPGT

GT Owner
Jul 31, 2012
82
Little Rock, AR
When I checked my car for the recall upgrade, it was not on the list. So I ordered the Accufab kit. When I started the project I realized
the new Ford bolts were installed. I went ahead and switched out the bolts on one side - but left the other side. Thanks Specracer for
reminding me to keep the unused Accufab bolts in the car.
 

Buzz Elliott

Member
Mar 12, 2021
7
Rick,

I too welcome you to our owner’s forum and will not jump on you as a new owner asking for important information on your new purchase. But as others have indicated this forum is populated with an abundance of technical information on numerous FGT topics. And the search function works well when doing research.

Pete gives a pretty concise synopsis of the early axle bolt problems which I have slightly modified for accuracy. I would further add to GTED’s comment that verification of new axle bolts is more than peace of mind. It really is a REQUIREMENT of ownership. You do not want to experience one of these failures as it renders the car undriveable when it occurs. As you state, you are not “mechanically inclined” and live in Florida, the GT guys are your closest experts on the topic. Certainly worth a call to Rich/Dennis to discuss this topic or have them perform an inspection.


If it can be determined the axle bolts have been replaced thru Ford’s customer satisfaction program 07B49 dated 3 December 2007 either visually (see search posts) or by OASIS you are good. As others have stated, AccuFab also offered a kit using ARP bolts which are visibly different looking than the Ford bolts. During the campaign the Ford bolt replacement and labor was free to the owners whereas the ARP kit was an owner purchase/install. Either replacement option works fine. The (Ford) identified bolt problem was NOT a strength issue and had NOTHING to do with vehicle speed or how hard/easy the owner had driven the car. The bolt loads are actually very low. The hydrogen embrittlement bolt failures were due to the electroplating process applied to the original European bolts used during OE assembly of the transaxle.

Pete, I think your number is a bit high. Jamal told us owners at the R2 meeting (August 1-5 2007) that they saw about 8 failures per 1000 hrs of Ford testing. They were not necessarily alarmed at this rate but customer’s claims seemed to support a higher level and since this was a halo project, they launched into a root cause investigation. To Ford's credit they did figure out the failure mode (which was perplexingly difficult) and resourced new bolts to a domestic vendor which are included in Ford's kit. You should definitely replace your bolts.
Again good advice to go check the OASIS data base with Ford to see if a repair (or bolt replacement kit) is logged against your VIN. Ford did allow the owner to just pick up the replacement bolt kit and do the install himself. Yes the Ford (free) recall has expired but you CAN still buy the bolt kit from Ford (I believe the P/N is 4G7Z-4B490-A) and have it installed by Ford, GT Guys, etc. at your expense. Or you can buy the AccuFab kit. Either way you will have to pay the labor to install either kit.
Mostly true. Again the bolt failures have NOTHING to do with vehicle speed. The bolts do not loosen in place, they break or fail. Any failure of both bolts on either driver/passenger side drive axle will render the car undriveable and you will need to get your car towed. Early on there were reports of owners experiencing this failure (very few) where it happened while driving and although disconcerting and may have produced some anxious sounds from the engine bay, I do not recall any substantial post failure damage to surrounding parts.
Correct!
Indy GT,
Is there a recommended torque for these bolts?
Buzz
 

nota4re

GT Owner
Mark IV Lifetime
Le Mans 2010 Supporter
Feb 15, 2006
4,177
Indy GT,
Is there a recommended torque for these bolts?
Buzz

Inner bolts: Blue Loctite. 27 ft-lbs
Make sure to perform the TSB for axle flange leaks at the same time.
 
  • Like
Reactions: Buzz Elliott

Buzz Elliott

Member
Mar 12, 2021
7
After 9 years in storage, it was confirmed both bolt heads were sheared on the starboard axle and I subsequently had the complete Accufab bolt kit installed along with the leak check. At this point to accurately document the corrective maintenance action, what steps do I need to take to properly document this service for the car’s history?
 

nota4re

GT Owner
Mark IV Lifetime
Le Mans 2010 Supporter
Feb 15, 2006
4,177
Having used the Accufab kit, it will be obvious to any (knowledgeable) inspector that the update has been performed. The Accufab kits use ARP 12-point bolts to re-assemble the flange which are really easy to discern from the OEM bolts.
 

Buzz Elliott

Member
Mar 12, 2021
7
Having used the Accufab kit, it will be obvious to any (knowledgeable) inspector that the update has been performed. The Accufab kits use ARP 12-point bolts to re-assemble the flange which are really easy to discern from the OEM bolts.
Yes, it’s pretty obvious for someone who looks at it, but what I’m referring to is in regard to the Ford Official Database or other sites where Ford GT VIN numbers can be checked to confirm compliance with the Takata Airbag Recall and the transaxle bolt replacement.
Thanks,
Buzz Elliott
 

twobjshelbys

GT Owner
Jul 26, 2010
6,053
Las Vegas, NV
Yes, it’s pretty obvious for someone who looks at it, but what I’m referring to is in regard to the Ford Official Database or other sites where Ford GT VIN numbers can be checked to confirm compliance with the Takata Airbag Recall and the transaxle bolt replacement.
Thanks,
Buzz Elliott
Did you have the work done at a Ford dealership? Did you supply the parts or did the dealership order them? If done by Ford they most likely entered it into OASIS, but not guaranteed so you should ask for a copy of the OASIS report. If Ford didn't do the work it will not be recorded in OASIS.
 

Buzz Elliott

Member
Mar 12, 2021
7
Did you have the work done at a Ford dealership? Did you supply the parts or did the dealership order them? If done by Ford they most likely entered it into OASIS, but not guaranteed so you should ask for a copy of the OASIS report. If Ford didn't do the work it will not be recorded in OASIS.
 

Buzz Elliott

Member
Mar 12, 2021
7
The work was not done at a Ford Dealership. I purchased the Accufab Bolts from The Racers Edge in Costa Mesa, CA, and had the replacement work completed at Benson Bros Racing, Mound House, NV. Plan to sell my GT on BaT and wanted to see if I could register the completed work on a recognized Ford database. I have photos of the old and new parts along with the BBR Work Order so I have documentation that the replacement was completed, but wanted to see if I could get the repair recorded on OASIS or an other recognized Ford database so the work could be shown by logging in the VIN.
 

twobjshelbys

GT Owner
Jul 26, 2010
6,053
Las Vegas, NV
Just keep the receipt in the cover with the owners manual.