Page 1 of 3

LOTs of US head bolts

PostPosted: Sun Feb 26, 2017 9:52 pm
by Quart Meg Miles
In 2007 I did, what I hoped would be, my last big engine job. Liners, head fill and skim, all bearings and new head bolts (via St Wilkins, probably from QED). Two years later the head gasket failed at the back so I had to fix that again. Then in 2014 I pulled the engine to change the starter ring gear and found two of my new head bolts had sheared near the top, positions 3 & 7 after 22,000 miles!

Wilkins supplied two S/H ones and fitted the ring gear but nasty fluids leaking from the front of the gasket area forced me to take the head off again a few days ago (after another 11,000 miles) and another two bolts were broken, positions 4 & 6! New ones (I marked them) so 4 out of 10! Nobody else has ever had or heard of one head bolt breaking.

I've assembled and fitted this head a dozen times since 1970, always the same way and with the same tools and rarely changing the bolts. I even checked my torque wrench from first principles two years ago and my torquing up of the bolts is something like 20, 40, 50, 60, 65 ft-lb. Wilkins uses 25, 45, 65.

What is going wrong?

Re: LOTs of US head bolts

PostPosted: Mon Feb 27, 2017 5:54 am
by alan.barker
I have always reused original head bolts and never had a problem.
Sorry but why did you need to fit new head bolts.
When the first bolts broke for me that could point to a batch problem. Meaning the supplier should do a stock withdraw of all the same batch and carry out sample test.
Alan

Re: LOTs of US head bolts

PostPosted: Mon Feb 27, 2017 7:33 am
by el-saturn
..........being such a vital part, i'd always use new ones: even in 12.9 quality! however i am currently reusing a set of arp2000s and will not have to worry (they're the best and worth every extra cent!!) sandy

Re: LOTs of US head bolts

PostPosted: Mon Feb 27, 2017 9:01 am
by rgh0
Do you have a photo of the failed bolts and a close up of the fracture surface. From this it is normally possible to diagnose the cause of failure.

There are 3 possible failure modes and they all leave tell tale signs in the shape of the fracture service - i.e. fatigue cracking, tension failure and torsion failure.

it is extremely unusual for a head bolt to fail in service if tightened correctly. The early reduced shank head bolts could fail when being tightened due to the combination of torsion and tension loads but once tightened ok did not fail in service in my experience when only the tension loads remain. I have never seen a failure with the later non reduced shank head bolts

cheers
Rohan

Re: LOTs of US head bolts

PostPosted: Mon Feb 27, 2017 10:20 am
by alan.barker
The supplier of the new bolts need to be contacted to recall or at least do sample tests on the same batch. Otherwise they could risk to have all sorts of reclaims and get a bad image. I'm sure Miles could contact the supplier and get the message across 8)
Alan

Re: LOTs of US head bolts

PostPosted: Mon Feb 27, 2017 8:13 pm
by steveh
Have the bolts sheared under the head ?

I have recently built an engine using new QED head bolts , these have a small fillet radius under the head which i noticed interfered with the sharp corners on the original washers , assembled unnoticed this could induce fatigue cracking . The original bolts were machined with a sharp corner under the head .

Steve

Re: LOTs of US head bolts

PostPosted: Tue Feb 28, 2017 9:20 pm
by Quart Meg Miles
Thanks for your replies!

Alan, at 240,000 miles and third rebore (reliner actually, as the first rebore was to 0.030") I felt it time to use all new components. St Wilkins couldn't remember whose bolts he had supplied and as it was back in 2007 it wasn't like recalling faulty new cars, even in 2014. In any case, neither St W nor QED knew of any previous failures on twin-cams.

Steve, I think you may have the answer to the bolt which broke at the head. Why they were made with sharp, stress-enhancing fillets I can't imagine and I will inspect the new bolts and washers in that light. I wonder if the slight dishing of the head around the bolt holes might also play a part as the washers end up dished. I don't like the sunken area showing in that washer, I wouldn't have expected it to wear like that as I lubricate quite generously when bolting up. The torque to undo the bolts seems as high as the setting torque.

Rohan, pictures to aid diagnosis and it may be significant that all but one broke in the screw thread, the 2014 ones very near the shank. When I came to unscrew the thread section left in the head they required just finger rotation so weren't bottomed. But the 2017 thread break is about 7 threads down and needed drilling and a stud extractor after scratching about at it for a while. I will check that the bolts won't bottom when I get a moment.

img_1445-broken-head-bolts-2014.jpg and
The 2014 pair
img_4818-broken-head-bolts-2017.jpg and
The 2017 breaks


I don't see any fatigue patterns and am unfamiliar with torque failures. Perhaps the only change in my methods was to follow your advice to slacken off the bolts a fraction before retorquing during the tightening sequence. Now I think of it, I did change the tightening pattern a few times as I don't like the diagonal sequence specified as it would seem to twist the head. There was some discussion on this some while back. Is it significant that all four broken bolts are in the front six?

Re: LOTs of US head bolts

PostPosted: Wed Mar 01, 2017 8:36 am
by nmauduit
is there a possibility that the bolts could have been bottoming (or similarly getting stuck at the thread for some reason) ?

Re: LOTs of US head bolts

PostPosted: Wed Mar 01, 2017 10:03 am
by rgh0
An interesting set of failures

The early ones look like classic fatigue cracking failure originating in the root of the threads just above the block. Loss of bolt tension over time with creep in the head and head gasket will result in this as the loss of bolt tension results in cyclic stresses in the bolt with each engine revolution and ultimately fatigue failure.

The later head failure looks like a classic cup and cone tension failure which is unusual in this area but i guess possible particularly if initiated by the sharp radius at the corner. The later thread failure appears due to a fault in the bolt at this point with what appears to be a flaw in the wire coil the bolt was made from.

The head bolt washer with wear / recession is a concern as normally they should be hardened steel and never show wear. The washer was to soft or the bolt spent a lot of time fretting on the washer to do the wear that apparent.

If you google images like

ductile or brittle torsion failure
ductile or brittle tension failure
fatigue crack failure

you will see there images of the various failures. You need to then understand bolt stress analysis to know why the various patterns occur.

cheers
Rohan

Re: LOTs of US head bolts

PostPosted: Wed Mar 01, 2017 11:23 am
by alan.barker
Have you still got the original head bolts :wink: :wink:
Alan

Re: LOTs of US head bolts

PostPosted: Thu Mar 02, 2017 6:19 pm
by Quart Meg Miles
alan.barker wrote:Have you still got the original head bolts :wink: :wink:
Alan

I'm not sure Alan! I assume you mean the ones from before 2007, which I might have squirreled away, but I have already ordered a new set from QED which I will inspect assiduously. :evil:

[Rohan wrote: "The early ones look like classic fatigue cracking failure originating in the root of the threads just above the block. Loss of bolt tension over time with creep in the head and head gasket will result in this as the loss of bolt tension results in cyclic stresses in the bolt with each engine revolution and ultimately fatigue failure."]

Thanks for the analysis, Rohan, but how much creep is left in a head after a quarter of a million miles, but I haven't any better ideas! I have been wondering if the threads were turned rather than rolled, there are some consistent marks all the way around the threads which could result from turning but I can't see how rolling would cause them.

I haven't been able to check the thread depths yet, the block is upside down and I've got a lousy cold, but will do it in a day or two and probably 'sacrifice' one of the suspect bolts to clean them.

Re: LOTs of US head bolts

PostPosted: Fri Mar 03, 2017 10:09 am
by 2cams70
I know it's a bit of a long shot but have you checked to make sure the flat surfaces on the head where the bolts sit are parallel with the deck face on the block? Any misalignment here would impose a bending stress on the bolt. I assume your tension wrench has been correctly calibrated too!! Another thing to watch out for is that someone may have previously lubricated the threads with molybdenum disulphide or anti seize grease. This will put the bolts in higher tension for any given torque setting. Although you may have cleaned the bolt threads the threads in the block may still have this grease on them. I generally run a kerosene soaked rag wrapped around a piece of dowl to clean these threads. The threads should just have a light coating of clean engine oil on them.

Note ARP bolts (not ARP studs) are also available which is what I've decided to use on my engine rather than the QED ones which seem to be a reproduction of the earlier (weaker) style Lotus OEM bolts. The ARP lube is the stuff to use with these but note you are not supposed to use ARP lube on bolts that aren't ARP.

Re: LOTs of US head bolts

PostPosted: Fri Mar 03, 2017 12:53 pm
by rgh0
Quart Meg Miles wrote:but how much creep is left in a head after a quarter of a million miles, .


The head itself from the photos looks good with very little recession under the head bolts. The earlier half moon die cast heads appear to last longer than the later sand cast heads. I suspect the first bolts failed due to slowly losing tension with a combination of creep in the gasket and bolt itself.

cheers
Rohan

Re: LOTs of US head bolts

PostPosted: Fri Mar 03, 2017 1:22 pm
by 2cams70
The reason why I ask the question about the bolt head surface relative to the deck surface is that my head before resurfacing was found to vary in thickness by a considerable amount from one end to the other - probably a result of having been badly machined in the past. This situation would also have put the bolt head surface out of parallel with the deck surface.

Re: LOTs of US head bolts

PostPosted: Fri Mar 03, 2017 2:57 pm
by alan.barker
How much difference in thickness from one end to the other. Surely across the width of the spot face for each bolt head would be very little imho, very little to measure across the spot face.
I would say the broken broken bolts need to be tensile tested etc .
Alan