Rolled Thread Engine Through Bolts Available as alternative

[JabSP6]

Thirsty

 

To answer the question, No i haven't. When the valve broke i stripped the motor and sent the crank, rods, and camshaft to be tested which i believe jabiru sent to Archerfield. I also sent the engine cases into Jabiru to be bolted together and the alignment and clearances for the main bearings were checked. A new set of main bearings were fitted and Jabiru gave the cases the thumbs up. All the O rings around the Dowels were replaced but the dowels were kept. This was the recomendation from Jabiru when i rebuilt the motor. Now Jabiru give you the option of running bigger (7/16") through bolts but to do this you have to change the dowels and also drill out the mounting holes of the Barrels to accept the bigger thread. These new bigger through bolts are still only machined but should be better than the standard 3/8" bolts. The main reason i have chosen to stick with the 3/8' Rolled threads is that you don't have to strip the entire engine to replace them and will be a straight forward job that the majority of us will be able to perform without the worry of machining engine cases.

 

One very important thing i did find out before i reassembled the engine was that none of the engine components are Balanced. I had the Crankshaft with the flywheel and propshaft extension bolted on balanced at 3000 RPM. Also the rods and the piston and gudgeons balanced. The results were quite incredible. Now the engine idles, revs and cruises quite smoothly compared to before the rebuild.

 

Regards

 

Andrew

 

 

[Thirsty]

I'll keep all that in mind if I ever buy a Jab. It sounds like it's worth the extra effort even for a brand new engine to balance it all up.

 

 

[ianboag]

My 2c.

 

One problem (there may have been others) with through bolts was that they were undersize. They were done to 1A standard which is about the same fit as on rattly farm gate bolts you buy at Bunnings. The nuts were wobbly. We managed to strip a number of bolts when torquing them up. That's how we found the undersize problem. We also found bolts where the threads had pulled so they looked like Christmas trees. It is possible that this happened after the bolts had been done up - possibly they were very close to the limit anyway and a bit of thermal stress pulled them over it. If this happens then some clamp is lost, the barrels/caee start to fret and we are on the way to a fatigue failure. That's one where a bolt snaps off at the barrel/case point. I understand that is where the snap usually happens.

 

I was under the impression that the Camit bolts were rolled anyway. Just rolled to be loose. 1A=wide tolerance, 3A=close tolerance. The longer nuts were a bit of a band-aid to spread the load over more threads.

 

The new SB means that engines have to be disassembled anyway as they say you have to replace the crankcase dowels.

 

Personally I am of the opinion that the problem is not enough clamp which allows things to move and fret. The old bolts/nuts were so close to max tension that there was no scope to reef them up any plus they were likely to relax in service. As they are welded up with 620 Loctite one would never know. Incidentally - this is not all bolts. We have seen ones of older engines that had no trace of this problem and torqued up just fine.

 

IB

 

 

[Thirsty]

I keep coming back to this - if Jab knew the bolts/nuts were well under spec and that engine failures were virtually guaranteed why didn't they change something long ago? The additional cost would have been nothing compared to a reliable engine - I don't get it. Unless it's not just about the bolts/nuts/dowels but something much harder to change is really the fault. Maybe the case halves won't handle much more "clamping force".

 

 

[Bruce Tuncks]

Thanks JP6, it all sounds good to me. I haven't had a problem in 370 hours, but I reckon you can't have too much reliability.

 

Rolled threads are actually a forging, and they are stronger and more fatigue-resistant because the crystals are deformed to fit the shape, instead of being cut through leaving crystal boundaries and internal defects in bad places for cracks to initiate.

 

With regard to the effect of thermal load on the through-bolts, I calculate that if the bolts were installed at 20 degrees C, they will have an extra 17.6 MPa at 100 degrees, due to the difference in thermal expansion between steel and aluminium. This may seem small in a steel of 400MPa yield stress, but its not trivial in a marginal situation. Just another reason why engines which are run to higher temperatures have more trouble.

 

So subject to Nev's good comments about accuracy of manufacture, I'll be getting a set for sure.

 

...Bruce

 

 

[facthunter]

I wasn't aware of the loose nut to stud thread fits. A lot of looseness is just not acceptable. A thread is not meant to work that way. A thread is a strange thing if you consider how it accepts the load. The first thread takes the lions share and only when it deflects/ distorts (bends) a bit, allows the second and subsequent threads to take some load, ever reducing as the nut is made progressively thicker. ( by design.). Nev

 

 

[bushpilot]

Any alternative component not specifically approved by Jabiru should be avoided in certified aircraft used for training.. Also use of non-approved parts could affect your insurance policy; check with your insurer first..

 

 

[ianboag]

Make what you will of this.

 

When we decided to use ARP studs/bolts/nuts there were some conversations with Jabiru because we had stripped/pulled TWO sets of newly supplied through bolts. Nobody seemed to be interested. It was just after this that the longer 12-point nuts were released. They were from ARP! They looked awfully similar to what we are using. Go figure.

 

We went to 10mm through bolts because the only (correct length) off-the-shelf stuff that ARP had was 10mm rather than 3/8". They could have made a special run for a setup cost of something like $US500 as I recall. ARP make engine bolts and studs for a living - much of their stuff goes into race cars. The ARP stuff was not hugely dearer that the Camit bolts. I don't understand - Jabiru use lots of car bits in the engine but want to make their own through bolts.

 

IB

 

 

[jetjr]

I think there are still some key (and experienced) people at Jabiru who think the through bolt/nut fix is uneccessary. Not sure what response youll get from Rod even with numbers to back it up.

 

Ive recieved my new nuts and new bolts from them and the thread fit sure isnt tight. Befroe I spend the time to replace them Ill be listening to see how you go with testing and Jab's response

 

 

[facthunter]

There's only one thing for sure. If you improve the strength of the stud/bolt and still torque it to the original specification, you cannot be worse off. .

 

There is definately an extra load imposed on the stud when the crankcase gets hot, there being a different rate of thermal expansion of aluminium and steel. Any extra load in service beyond the above, could only happen if the crankcase joint opened up due to other loads occurring, eg detonation. OR if the clamping force is not sufficient. ( I doubt that this is the case here, but you have to consider ALL possibilities). Locktiting the nuts to stop them coming loose, seems weird to me, though earlier Continentals had Pal-nuts fitted. ( Think that practice has been discontinued.)..Nev

 

 

[stanzahero]

You need to be careful when applying loctite to threads before tensioning them.

 

If the loctite cures you will lose most of your torque in friction. Normally almost 80% of tension is obsorbed by friction.

 

In the automotive industry (this may be different in this case though) a small amount of oil is the best thing on the threads because it lets the nuts/studs do the job they are designed to do - hold the parts.

 

 

[Thirsty]

Actually lubricating threads before torquing is a no-no as well. If you read the Jab SB dealing with the thru bolts you'll note they say to use the high strength loctite (can't remember the number off hand) and then torque up the bolts quickly before it goes off! That's going to be quite hard to do 'cause you have to get both sides on before you can torque them. Has anyone done this yet? Hard or impossible :)

 

 

[jetjr]

I have a new set of longer throught bolts and new ARP nuts here on my desk, how can I tell if the thread is rolled or cut?

 

There is definitely movement between nut and thread, seems the same when pushed or pulled.

 

The bolts are waisted to ~ 7mm diameter ~ 40mm before thread starts

 

197mm long overall

 

 

[facthunter]

What is the total waisted length? (roughly). on each bolt.

 

Lubricated changes the torque figure. I like the idea of a prescribed amount of turn after contact or a torque figure. Say torque to 10 ft/lbs and the 90? degrees more turn. DON'T USE MY FIGURES.. They are hypothetical. I'm talking the principle.

 

A simple TEST for serviceability of a stud might be to torque it to the figure required plus 30% and see if the stud returns to its previous length, afterwards. Don't do this on the engine. Put the stud through a piece of waterpipe with the ends faced off in a lathe. This is the minimum overtorque that I would expect it to carry. IF they break torquing to the standard figure, that would be a big concern. Nev

 

 

[ianboag]

What is the total waisted length? (roughly). on each bolt.Lubricated changes the torque figure. I like the idea of a prescribed amount of turn after contact or a torque figure. Say torque to 10 ft/lbs and the 90? degrees more turn. DON'T USE MY FIGURES.. They are hypothetical. I'm talking the principle.

A simple TEST for serviceability of a stud might be to torque it to the figure required plus 30% and see if the stud returns to its previous length, afterwards. Don't do this on the engine. Put the stud through a piece of waterpipe with the ends faced off in a lathe. This is the minimum overtorque that I would expect it to carry. IF they break torquing to the standard figure, that would be a big concern. Nev

I would agree with that. Lubed or unlubed makes huge difference to the tension. Jabiru say you do it dry. I've never seen that anywhere else. Conti and Lyc say to use engine oil on the threads. ARP supply their own lubricant to be used on their stuff. Go figure.

 

When we did some of the testing you describe, the bolt did not permanently stretch - the threads did. 30% over was about where we stripped one even. So I would also check that the bolt threads do not permanently distort ("pull") with the 30% overtorque. With the new longer nuts they probably won't.

 

If you are bothered about the nuts being loose, you could check (google knows) what 1A thread tolerance means and what 2A and 3A tolerances are considered appropriate for. Call Jab (or Camit) and ask to speak to someone technical. Ask whether the bolts are made to a 1A, 2A or 3A tolerance .....

 

Cheers IB

 

 

[facthunter]

Ian , the recommended torque figures , for the different engines, would take into account whether the threads were lubed or NOT lubed and be adjusted accordingly. Nev

 

 

[Bruce Tuncks]

While a straight tensile test would be the ultimate, it would be hard to organise. I reckon that torquing to failure would be good enough, provided that the friction of each nut were the same. If they were both lubricated with an extreme-pressure grease, then this should be the case.

 

I'll be interested in the results...

 

Here's a thing I would have found hard to believe but it really happened to me:

 

My Glasflugel Mosquito glider had through-bolts holding the alloy wheel-halves together. The main cyclic stress could only have been from daily temperature change.

 

Well one day one failed. Is was a 6mm dia socket-bolt. Wow, thought I, a freak thing, so I just replaced that one bolt.

 

Then 3 weeks later another failed on the same glider, and yet another, the third one being from another Mosquito based in Sydney. These were all German socket-bolts but I don't know the details. Perhaps they were over-tightened at the factory. But it took 20 years at one cycle per day.

 

Bruce

 

 

[Bruce Tuncks]

Hi Andrew, how are the new bolts going?

 

I looked up a textbook from my days in engineering school (Faires, "Design of Machine Elements") and it implied that rolled threads should be nearly twice as good as cut threads for the tensile strength range of the steel used.

 

regards, Bruce.

 

 

[turboplanner]

Provided they are a tight fit with the nuts

 

 

[JabSP6]

Bruce

 

I finally completed testing on the new rolled thread bolts that i received from TJ over in West Aus.

 

I used that exact same conditions and equipment for the test that i had previously tested the standard Jabiru bolts and Jabiru nuts. The Rolled Thread Bolts and Nuts did not deform in any way up to a tension of 50 ft/lb. At 55 ft/lb the tension wrench clicked but it started to feel as though something wasn't quite right. When i then tried to go to 60 ft/lb the Bolt itself broke just on the inside of the nut through a section of the thread. When i then picked up the nut with the broken section of thread the thread screwed straight out of the nut with ease. This suggests that the thread of both the bolt and the nut are perfect and the process of rolling these threads is much stronger than the standard jabiru bolt and nut. The bolt material itself has yeilded between 55 & 60 ft/lb. I then carried out the test 3 more times with similar result of the bolt itself breaking anywhere from 55 ft/lb to 65 ft/lb. Each time the thread section came out of the nut easily using just my fingers to wind it out. This is a huge improvement on my initial tests of the standard Jabiru bolt and nut threads which where failing between 25 & 35 ft/lb by distorting the thread.

 

I will be fitting these new rolled thread bolts the next time i need to do any overhauling of the motor.

 

JABSP6

 

 

[Guest JRMobile]

Andrew, Thanks for the feed back on these new bolts - great work.............:thumb_up:

 

Cheers John

 

 

[Bruce Tuncks]

Thanks Andrew, I will be getting a set to go with the 12 point nuts which Jabiru supplied for the mod.

 

It is not an easy job to replace those nuts, the pipes have to be removed and you have to acquire a long crow's foot extension.

 

So it makes sense to replace the through-bolts at the same time. As with you, I don't have a sense of urgency as my engine has always run well. But you can never have too much reliability.

 

There is a really good chapter in Schwaner " Sky Ranch Engineering Manual" titled "Fasteners and Failures" in which I found that all the same issues are well-known to overhaulers of Continental and Lycoming engines.

 

cheers, Bruce

 

 

[facthunter]

In a straight-out tension failure I would have liked it to have failed at somewhere other than the base of the thread. The stud should have portions reduced to below the root diameter of the thread which distributes the STRAIN along a large portion of the stud and not at the first thread. If you are looking at a fatigue failure this would be the preferred approach.

 

Some would say how can weakening it make it stronger? Well you are preventing it being weak in a small spot, which guarantees it will fail there, fairly quickly. Nev

 

 

[Bruce Tuncks]

That's right Nev, I've seen Lycoming connecting-rod end bolts built like that.

 

Here's something I found hard to understand... when the bolts are clamping something much stiffer than the bolts themselves, there is NO extra load put into the bolt until the forces are enough to open the clamped joint. So there should be no fatigue on those bolts if there is no opening of the joint.

 

But when you have aluminium clamped with steel bolts then the temperature effects will cause fatigue as I found once.

 

Now what torque should the new bolts have? The latest maintenance manual says 30 ft-lb, but my old Tech manual says 35.

 

I'm inclined to go for 35 because of the clamping theory and because the bolts are stronger.

 

cheers, Bruce

 

 

[bushpilot]

I'm doing the Jabiru engine building and overhaul course at Bundaberg this week.. There is a lot more to replacing through bolts and nuts than some might consider.. Will report more fully next week.