Conrod Bolt "re-use" question

[smokintr6]

The thing I don't like about reusing the factory bolts is that you didn't install them so you don't have the un-stretched length. Though if I understand the analysis above, it seems that if they take torque again, then they were likely ok.. maybe an over simplification.

I always get nervous thinking about re using a bolt when the factory manual says to toss it. The flywheel bolts specifically seem like they could be reused, but I'd hate to risk it, and I HAVE had a 944 turbo flywheel come loose. those cars use the same bolt as the 9 bolt 911 crank. But they're relatively cheap, so it seems like a no brainer to get new ones.

For my 964 engine refresh I opted for new rod bolts, and the price difference between the ARP and stock equivalent didn't seem very high, especially since you can re use the arp bolts. In this case going for ARP also seemed like the right thing to do.

Unfortunately with older cars, eventually used OEM parts almost always become better than the reproduction parts available. This has been my experience with other non German vintage cars. We are already seeing this in the 911 with the Glyco rod bearing problems recently. In theory if you could find a barn fresh carrera engine that had 30k miles on it, and had been mothballed there would be many nice parts inside that would probably exceed current quality metrics. I'm thinking the rockers, rod bearings, and main bearings right off the bat. I have a friend who races a 1970s triumph, and he was the first person that pointed this phenomenon out to me. We're lucky in some ways that Porsche is still in business, and make porsche branded replacement parts with good quality, but the trade off is astronomical pricing. As a community we are less lucky than my friend in that these cars get driven for hundreds of thousands of miles, and the cars / engines have value as a whole... so low mileage part outs are less common.

It's certainly a complicated problem. I bet I spent $5k on my refresh, and I only rebuilt the heads, replaced the head studs, ARP rod bolts, rod bearings, crank thrust bearing, intermediate shaft bearings, and I did all the work myself. I left everything in place that seemed like it could reasonably be reused.

Someone else mentioned that it was impossible to determine when parts were "worn out" when you start trying to reuse everything.... That's not quite true. "worn out" literally is short for worn out of spec or tolerance. Pistons, ring gaps, bearing clearances, etc can all be measured and have a spec in the manual. You still must apply a judgement call on condition of bearing surfaces, and rocker / cam lobe condition. but that comes with experience, and some risk that one might not be willing to take as a shop with their reputation on the line. If you were rebuilding a 350 chevy, it would be dumb to reuse any bearings, because aftermarket parts are good, and you can probably get every bearing in the engine for $150. There are almost $2,000 of bearings in a 911 engine if you get porsche branded parts. For me that's a lot of money to throw away if parts are still in spec.

[Robert Espeseth]

Originally Posted by Catorce

By that logic you'd never put any new parts into your motor. Where does the line cross between "proven performance" and "worn out" on an engine component?

As soon as you remove your rod to do new bearings, and unbolt that con rod, you are upon the horns of a dilemma. You can either do math problems, throw a science lecture that talks about bolt stretch and post calculations, or you can pay 9 bucks and put a new bolt in there.

I am being facetious but you catch my drift I bet.

All parts in an Porsche engine has well documented toleranses, and it is quite easy to find out if any part is beyond use...

[chris_seven]

Originally Posted by Catorce

Hold the phone a minute, Chris. I am sorry I threw flywheel bolts into the mix, let's focus back on the rod bolts for a minute.

Tippy, you don;t have skin in the game here because you have ARP bolts, and they are designed to be reused.

Chris, I am assuming you are a shop of some sort. I am a parts manufacturer as well. We both have similar considerations, namely, customers.

I logged into my SSF account, and I saw that my wholesale cost on Porsche rod bolts for an SC (I picked SC at random because it is a common car) were $9.90 each.

So in total, the bolts cost me $118.80 for the whole motor, again, wholesale.

You mean to tell me that your shop breaks a motor all the way down to rods coming off, and the $118.80 it would take to replace them per factory recommendations is somehow too much? I am assuming you buy wholesale as well.

My angle is this - I get all your science and compound sentences, honest, I really do, but you're going to tell me that instead of just paying the $120 you're going to throw a bunch of science calculations at the motor and send your customer packing with old rod bolts?

What's your angle here? I am trying to understand how any of this benefits you as an engine builder. It seems to only create liability in the event of a failure.

Oh, and it's not as if Porsche recommends that a lot of it's bolts be trashed - I can only think of a couple that it explicitly states not to reuse, con rod bolts being one of them.

So what gives? Honest question.

Love the info in your posts, just after the business side of this decision is all.

The discussion about rod bolts is technically interesting and I have never found any justification as to why a bolt that behaves elastically needs to be thrown away.

When questions were first asked about this subject many years ago we started measuring rod bolts to see if we could learn more about the basic reasons for this advice.

If you tighten to 20Nm + 90 degrees I believe that a 911 rod bolts remains elastic and we have never found one that has yielded.

It really comes down to 'belief' and as we don't see a problem why throw them away?

If rod bolts were subjected to fatigue loading I could see the point and would throw them away.

We would never 'send one of our customers packing' and fail to take responsibility for our decisions.

I don't know why the head would shear off a flywheel bolt and it clearly can't support the torque needed to make it release.

The real question is how it supported the load during tightening. It seems reasonable to assume that the torque needed to move the bolt has increased and to try to find out why this is the case.

Work hardened materials rarely exhibit the classic signs of brittle fracture as work hardening implies some degree of ductility which will dominate the failure mode.

I would expect a Grade 12.9 fastener to have a reduction on area at failure of at least 12% so it should noticeably deform before it breaks.

Using threadlockers on bolts which are highly stressed can be a problem as the increase in prevailing torque can be quite significant and could cause an overload when the bolts have to be released.

We don't generally thread lock flywheel bolts but I would tend to use a low strength material for this duty.

I guess my logic is that if we have the engineering capacity to make sound decisions we should use this and pass on the savings.

If I felt we were taking any risks I wouldn't take this approach.

[Catorce]

Originally Posted by chris_seven

I guess my logic is that if we have the engineering capacity to make sound decisions we should use this and pass on the savings.

This right here answered my question. While I don't agree with it, I respect your decision and of all people I have spoke to on here, you seem to be the one that has been able to articulate your position best. Hat's off to you.

[Tippy]

I know this is way off subject, and I apologize, but we do put out a lot of fear regarding our rodbolts failing. When I rebuilt old American V8's, I reused the original rodbolts several times knowing zero history without issue turning over 6K RPM when I was young and dumb. Never once a failure after beating those poor engines to death. Second, I've never seen a rod bolt failure in any engine. It seems the rod caps tear open right around the rod bolt flanges more than anything. Seen that dozens of times with the rodbolts there doing their job.

Another thing I've heard, is that you must replace crank pulley bolts once used. I've reused those on several vehicles too without issue. Even supercharged ones where there's added stress from the blower pulley.

Point is, I'm not surprised by Chris' findings. As an engineer I used to work with used to say when he read something over the top, "maybe this engineer had his pocket protector on too tight?"..... 😂 😂 😂

[Walt Fricke]

I, too, reuse my 6 bolt flywheel bolts, despite torqueing them to 150 lbs/ft. Removing them, despite the red Loctite, doesn't require heat or anything special - the impact wrench just buzzes them off.

But I have had a stock rod bolt fail, with disastrous results as you would expect. Back when you couldn't get aftermarket high strength bolts for the 66mm crank rods. I spun the engine to 8,000 RPM on the track for a couple of years when blammo. The rod bolt showed the classic necking down above and below the failure point. The crank was OK - the stretch apparently didn't last long enough to drop the oil pressure on that journal - stuff wasn't all blue or black.

I like the aftermarket rod bolts, use a stretch gauge, and write things down.

[Tippy]

^ Actual usage, love it.... 😍

[gtc]

Originally Posted by Robert Espeseth

All parts in an Porsche engine has well documented toleranses, and it is quite easy to find out if any part is beyond use...

Totally off topic, but how about pressure fed chain tensioners? That is the one part I could not find any info on when I rebuilt my 375,000 mile motor. How and when do you rebuild or replace them?

[manbridge 74]

Well if you search this forum alone you see some pictures of broken flywheel bolts. Reviewing one thread it seemed to have no real conclusion but that it might be a bad run from a subcontractor. So maybe hanging on to old bolts is a good idea. Sure hope it wasn't the same maker with regards to rod bolts!

[racing97]

Originally Posted by chris_seven View Post
I guess my logic is that if we have the engineering capacity to make sound decisions we should use this and pass on the savings.

For many years the information on Porsche engines and components was grossly incomplete and I am sure if we attempted repairs and improvements on the current models we would find it has not improved much. 40 years ago we were on our own and it took the consultation of people exactly like Chris who reason and investigate test and verify and economize our projects so that we do not have to major in the minors of engine building and if we are lucky move on to improvements that could maybe move the torque needle up if you know what I mean.
regards

[Robert Espeseth]

Originally Posted by gtc

Totally off topic, but how about pressure fed chain tensioners? That is the one part I could not find any info on when I rebuilt my 375,000 mile motor. How and when do you rebuild or replace them?

Graham, no idea about how to check them :-/ ...but quite sure there is info elsewhere in this forum.

[Mark Henry]

On thing I just started and plan to do on all my future builds is rod bolt stretch vs torque settings. So far I've only done this to a 2.7-3.0 10mm ARP rod bolt (stock rod), what I did was measure both the recommended stretch and recommended torque values and on the ARP 10mm bolt found they came out exactly the same.

I will be repeating this on a Carrillo (996) rod bolt (Carrillo rod) very shortly and plan on checking all future rod bolts that come across my bench.

Not saying I'll replace stretch with torque, or that my methods are very scientific, really only doing this for my own knowledge.
I found the results on the 10mm ARP interesting and now want to see how it compares to other fasteners.

[Walt Fricke]

Mark - the major benefit of using the stretch method is accuracy, since thread friction is simply not part of the clamping force equation - you are, in essence, measuring clamping force directly by measuring bolt stretch.

As you have seen from ARP's information, you need specified burnishing of threads, and a specified lubricant, to use torque in the manner they feel is adequate if you don't measure stretch. Any variation in this, or discrepancy in your torque wrench's accuracy, will leave you with more or less than the specified clamping pressure (which you would have to calculate, since that term is not used, but it is what makes things hold together).

Porsche maybe sort of uses a method half way in between - torque to a value, and then turn X more degrees.

All these methods have to have a suitable engineering safety factor built in. Maybe in an F1 motor lightness considerations mean leaving the slimmest margin between the forces of the running engine and the clamping force, but that can't be so in a street car motor.

I'm sure that for a street motor, torque is perfectly adequate. If building a higher performance motor than Porsche intended, it only makes sense to use stretch. The tool is cheap, and doesn't really take appreciably more time than torqueing. It's only draw back is that you can't use the stretch method if you don't have the crank out on the work bench.

[chris_seven]

This situation isn't quite as simple as it sounds.

Stretch only works for a bolt that is elastic and it fails to provide the assumed preload as soon as the stress exceeds its tensile strength. This is the point at which the bolt will begin to neck.

As I suggested in an earlier post for the ARP method to work correctly the bolt must be capable of maintain the applied load throughout the tightening sequence.

For example - if the bolt were incorrectly heat treated - which I would accept is quite rare - and it was simply tightened to stretch then you have no idea on the stress/strain condition of the material. This is the reason that ARP suggest that of their bolt deforms permanently by more than 0.001" it should be rejected.

I have to say that I would question this value as being a bit too low and as the 'uncertainty' of measurment of their stretch gauge is likely to be in the order of 0.003-0.004" I would think 0.0025" would be more realistic.

I would also not make the decision to reject the bolt until I had taken it to the stretch length 2-3 times.

If you have sufficient experience then you would be able to 'feel' that the bolt was still maintaining load but otherwise you have some uncertainty.

Torque + angle is really only a stretch method without the need for a gauge.

The assumption is that the initial torque value just makes the bolt 'snug' which is to say it is sufficient torque to eliminate thread and underhead friction and does not stretch the bolt.

The stretch is then determined by the thread helix and angle of rotation.

The same problem exists as with using a stretch gauge as you have no real way of determining the stress/strain condition after the final rotation.

The best method must be to use a dial indicating or digital wrench and to ensure that load continues to increase right up to the point at which the bolt is either sufficiently stretched or rotated through the specified angle.

The use of burnished threads and consistent lubrication has been well established for many years and simply follows the concept of being able to establish a repeatable 'nut factor' which is commonly used by the Aero Industry.

If established correctly using these concepts torque tightening does become a sensible option as the causes of scatter can be eliminated and using torque becomes a viable method.

Burnishing bolts does, however, contradict the 'use once' and throw away concept.

A good explanation can be found in John Bickford's Text Book 'An Introduction to the Design and Behaviour of Bolted Joints' which is summarised here:

Dissecting the Nut Factor | Archetype Joint

[Catorce]

Originally Posted by Walt Fricke

The tool is cheap, and doesn't really take appreciably more time than torqueing.

I'm not so sure about the tool being "cheap". My stretch gauge is an ARP one that is $200. It is cheaper than my $600 a piece Snap On digital torque and angle wrenches, but still an expensive tool.

I like the ARP gauge, but quite honestly with the Snap On, you torque the bolt to it's initial torque, and the wrench is within 2 percent accuracy at this value. Then, while still on the bolt, you press a button and the wrench goes into angle mode, and you spin to angle. Done. To me, much simpler, but it takes a pricy tool to do it.

[Mark Henry]

Originally Posted by Catorce

I'm not so sure about the tool being "cheap". My stretch gauge is an ARP one that is $200. It is cheaper than my $600 a piece Snap On digital torque and angle wrenches, but still an expensive tool.

I like the ARP gauge, but quite honestly with the Snap On, you torque the bolt to it's initial torque, and the wrench is within 2 percent accuracy at this value. Then, while still on the bolt, you press a button and the wrench goes into angle mode, and you spin to angle. Done. To me, much simpler, but it takes a pricy tool to do it.

I made my own, a hexbar off an old HF adjustable wood circle cutter, bit of lathe time to make some bits and an old starrett dial indicator I had kicking around.

[Catorce]

^Yep, not a complicated tool at all. In the past couple years however, after turning semi-pro, I slowly got rid of all my homemade tools and harbor freight stuff. It just doesn't inspire credibility when people are watching.

I had lots of frankenstein stuff too. Worked fine, most of them, like you, were made on my mill or lathe.

[Ken911]

Summit racing has a stretch gauge for 50$ I couldnt find one here on pelican so I'm assuming they dont carry one.

[Mark Henry]

Originally Posted by Catorce

^Yep, not a complicated tool at all. In the past couple years however, after turning semi-pro, I slowly got rid of all my homemade tools and harbor freight stuff. It just doesn't inspire credibility when people are watching.

I had lots of frankenstein stuff too. Worked fine, most of them, like you, were made on my mill or lathe.

Most of my stuff looks good as well, I made my own 996 engine stand, I've held the $600 ($1000 Canadian) unit in my hands, mine is better thought out and equal in build quality.
Some of my stuff looks hacky, but most of my tools even a mechanic would be hard pressed to figure out which one is the pro tool.
But then there's tools I just can't make, those I buy.

[Walt Fricke]

Catorce - I've got a Summit stretch gauge. I have difficulty seeing how the $200 ARP gauge is going to do a better job in any way, so ~$50 is cheap compared to the ARP - or a fair number of other special tools.