Head studs

I needed to make some special head studs to fit the alloy 920 barrel. The barrels require a longer stud but the stud must be reduced from 3/8 to 5/16" and fitted with the small diameter type 5/16" nuts.
The length is needed because the cylinder is different than the stock item. The reduction in diameter is due to weld build up in the bolt recess areas. I did that prior to boring for the 920 cylinders.

While sorting out the details of that I noticed that the Andover supplied studs do not utilize all of the thread available in the head. I have another head here with original looking studs and they are the same.
They hang on in the aluminium head by just 8 threads.
I made the special studs with a longer threaded section.
They catch by 12 threads, 50% more holding power.
With just 8 threads trying to hang onto hot cast aluminium, it's no wonder the stock studs pulled out on my 850!
I might see if the 5/16 studs which thread into the cylinder top could benefit from a similar re-think.

The stud at the left is the stock item from Andover, the one on the right is made to fit the 920.

Glen

what gr of steel did you use?

My Dad, a former mechanical engineer, told me make a threaded section longer does not greatly improve loading limit. As load grows, the bolt/stud begins to stretch at the threads nearest the top of hole, so its those first few hole threads that deform initially, followed by the next deepest until all fail. Since not all take same amount of load at same time, doubling length does not double load limit. Making the diameter larger or using courser threads make a much great difference.

Tornado said:

My Dad, a former mechanical engineer, told me make a threaded section longer does not greatly improve loading limit. As load grows, the bolt/stud begins to stretch at the threads nearest the top of hole, so its those first few hole threads that deform initially, followed by the next deepest until all fail. Since not all take same amount of load at same time, doubling length does not double load limit. Making the diameter larger or using courser threads make a much great difference.

Click to expand...

I don't think that is correct.
As an example , Vincent uses very long threaded sections in the alloy crankcases to bolt the engine up and hang it from the UFM. That's connection is all that holds the engine and the bike together.
If a few threads were as good as 60+? then Phil Irving would have done it that way. I'm sure that if that connection had been done with just a few threads, it would have ripped out very quickly.
As it is that connection holds the entire bike and engine together without failure for lifetime.
So that and my own experiences tell me that 12 threads in alloy will hold a lot more than 8.

Glen

850commando said:

what gr of steel did you use?

Click to expand...

304 SS

304ss only has a tensile yeild strength 0f 75-90ksi, i think id would pick something with a higher yield...

CNW uses the same for the base studs.
The stud won't come close to breaking at 30 ft lbs. The aluminium thread is the weak point by a mile. It is only hanging on by a very narrow margin of a few ft lbs when everything is perfect. Jim Comstock demonstrated that in a video.

Glen

Tornado said:

My Dad, a former mechanical engineer, told me make a threaded section longer does not greatly improve loading limit. As load grows, the bolt/stud begins to stretch at the threads nearest the top of hole, so its those first few hole threads that deform initially, followed by the next deepest until all fail. Since not all take same amount of load at same time, doubling length does not double load limit. Making the diameter larger or using courser threads make a much great difference.

Click to expand...

Tornado - without wanting to dispel your father's views...
He's correct that the "tightest" few threads will initially take the load, and if the load is high enough, start to yield. That yielding will, though, bring more threads into play and so on until all threads are at work - some to a greater degree than others, but all will play their part.
So... the simple answer is "the more threads the better"....and yes - I am a nerdy mechanical engineer with 40yrs experience,

The strength ratio for good steel alloys (not 304ss!) to aluminium alloys is at least 4:1.
Therefore you will need at least four times the thread-base area in the aluminium alloy as the core area of the stud to ensure that the stud will fail before the aluminium threads pull.
...and that's without taking in to account aluminium's greater propensity to fail at elevated temperatures.
I agree - use all the threads you can!

Glen, you gonna slim the shanks down? The increased stretch helps thread life and gasket sealing, according to the theory at least !

Fast Eddie said:

Glen, you gonna slim the shanks down? The increased stretch helps thread life and gasket sealing, according to the theory at least !

Click to expand...

Thread life and head life - reduces the compressive forces at elevated temperatures and so stops the "creep" squashing of the head (often misinterpreted as head gasket failure)

Since it is on the bench why not put in inserts as per JC? You do get quite an increase in holding power.
Your long studs are a step forward but I bet not by a whole lot. Do it now before you have to do it later.
Guess where my head is? :-(

Nigel, with the SS following the thermal expansion of the Al. much closer than steel plus a little built in stretch with SS , I won't bother with the waisting idea.
It's a good one for high strength steel bolts though.
When you think about it, there's a bit of irony there- we seem to love the idea of high strength fasteners , so we use them then chop them down in size to lower the strength to allow more stretch.
I don't worry at all about the 304ss being too low in strength. I think of them as studs into soft cheese. The soft cheese is the problem.
I might do a Jim type torquewrench test to see how much 12 threads hold in hot Al. vs 8.

Glen

When using inserts you can also get more threads/grip with time inserts than with helicoils because the helicoil has a tang at the bottom that interferes with the stud depth (the stud will not screw in as deep with a helicoil). You can go with a .620" length time insert and have longer thread engagement than with a helicoil. Whatever advantages someone may think a helicoil has dissappears when you can get more thread engagement with the extra depth of a time insert (they also stay in place and don't have any creep out problem that can happen with helicoils).

worntorn said:

CNW uses the same for the base studs.
The stud won't come close to breaking at 30 ft lbs. The aluminium thread is the weak point by a mile. It is only hanging on by a very narrow margin of a few ft lbs when everything is perfect. Jim Comstock demonstrated that in a video.

Glen

Click to expand...

not to start a pissing match, when i checked CNW data their SS was rated at 150ksi, and last thought, RobSS is correct on the load redistribution, although as rule of thumb we always ensured that we had as much thread engagement as the dia of the bolt/stud.

I have 2 left over CNW stainless studs here, so I tested one to break point.
These are from the barrel mount kit. This alloy Barrel uses 6 through bolts instead of 4 as per standard, so the two weren't needed.
This is a 3/8 X5/16" stud that would get torqued to 20 ft lbs.
It let go at 39 ft lbs, so all kinds of margin there.
I also had a spare 3/8" stud from those I made on Saturday.
With the extra cross section, it should require about 1.45 times the force to break.
That would put it at around 56 ft lbs.
It let go at 67, a fair bit higher than the CNW item, not that it's needed! Were not bolting up a D9 final drive here.
Those studs will get torqued to 30 ft. Lbs, so there is ample strength in the studs.
As Jim Comstock observed, the problem with the head connection is high tensile low expansion steel bolts meeting low tensile high expansion aluminium. When irresistible force meets immovable object it's never a good thing!
His solution is to reduce the bolt diameter to introduce more stretch, less overall strength ( almost 50% @ 5/16"Dia)
Another solution is to use a material with some built in stretch and also greater thermal expansion. SS304 works for that. There is more than ample holding strength for this job.
This is what they are made of.
My main point isn't about the SS , it's the simple observation that the stock studs do not utilize all of the holding power available in the head.
The extra four threads could easily make the difference between success and failure, and as we have seen of those connections over the years, they love to fail.
I will do a comparison of torque to pull out for 8 threads and same for 12 threads.

About 20 years ago I had problems with the 3/8 head studs so I installed 5/8" deep timeserts and reduced shank studs that would stretch instead of pulling out threads. See below (with .003" thin head gasket and copper wire).


This was the first reduced shank 3/8' Norton head stud to my knowledge, it worked perfectly and these studs have been in my bike ever since. The 3/8" studs are the ones that locate the head so it is centered on the bores. Now I make them so that the outermost diameter of the triangular section stays at 3/8" diameter. This way the shank can be reduced along its entire length and still locate the head on the cylinders.

Its reduced shank rod bolt technology but its triangular (ARP studs shown below).

Thats one way to overcome the problem and it looks like a good one.
I think all that is needed for the threaded connection, if they haven't already failed, is to use the entire depth, almost 3/4" on the two RH 10 heads I have here.
Running a die down the stock studs to add a few threads in the head insertion side would do that.

Glen

you've done your homework. you get an A+ (no mojo for thumbs up), the stretch bolts work.

The stretch is a great idea. Thermal expansion matching Al + stretch is perhaps even better, especially with an alloy cylinder.
The thermal coefficient for 304 ss is a near match for aluminium casting alloys like A132.
Im quite concerned about the expansion of the alloy barrel.
There are not very many 100,000 mile iron barrel 920s out there, or even 30,000 mile 920s.
Add the alloy barrel into the equation and things get worse. Apparently, with standard high tensile cylinder hold down bolts ( not necessarily waisted)holding against the aluminium there was a tendency to rip the threads out of the crankcases with repeated heating cycles. I can see why.
That should not occur with the bolt expansion matched to that of the cylinder.


Glen