Torque settings for uprated studs and other design features
Discussion
It's mid season and the car has been going well, after early car related issues ironed out and the first races missed it's recently taken three wins on the trot.
Originally the head (when a standard road car) was held down by M10 bolts, I upgraded to M10 ARP studs & nuts as it's all I could get in the time/budget permitted. However as things have progressed more air has been stuffed in to stay ahead of the competition and the head has started to lift slightly. The owner/driver is of course liking the smell of victory so more time and funds are available.
There is a short break so I've commissioned some more custom M10 studs made from a stronger material, larger nuts, bigger washers (the others were showing signs of sinking in).
The ARP studs had a wasted shank (down to 9.25), the new ones will be 10mm all the way down.
Head is aluminium, block cast iron.
It isn't ARP who are making them, the manufacturer doesn't get involved with torque settings.
On the next rebuild I will probably go to M12, but there isn't time to do that right now.
So my question; is there there anyone out there who can calculate the torque settings for the nuts? Paid job of course if needed.
Or would I get a spare and do a sample, torque the fastener until it starts to stretch beyond its elastic limit or something breaks, then reduce the torque by X amount?
I'm kind of favouring the latter as I don't know what i'm dealing with in that (in order of likeliness):
1.The stud could stretch or break
2.It could pull out of the block
3.The nut could break.
Originally the head (when a standard road car) was held down by M10 bolts, I upgraded to M10 ARP studs & nuts as it's all I could get in the time/budget permitted. However as things have progressed more air has been stuffed in to stay ahead of the competition and the head has started to lift slightly. The owner/driver is of course liking the smell of victory so more time and funds are available.
There is a short break so I've commissioned some more custom M10 studs made from a stronger material, larger nuts, bigger washers (the others were showing signs of sinking in).
The ARP studs had a wasted shank (down to 9.25), the new ones will be 10mm all the way down.
Head is aluminium, block cast iron.
It isn't ARP who are making them, the manufacturer doesn't get involved with torque settings.
On the next rebuild I will probably go to M12, but there isn't time to do that right now.
So my question; is there there anyone out there who can calculate the torque settings for the nuts? Paid job of course if needed.
Or would I get a spare and do a sample, torque the fastener until it starts to stretch beyond its elastic limit or something breaks, then reduce the torque by X amount?
I'm kind of favouring the latter as I don't know what i'm dealing with in that (in order of likeliness):
1.The stud could stretch or break
2.It could pull out of the block
3.The nut could break.
Edited by Evoluzione on Wednesday 13th July 11:12
One of my favourite topics. In essence it's all very simple - but beset by such vagaries as to render most torque calculations hopelessly imprecise and essentially useless.
For a given fastener size and the same thread pitch the clamping load is proportional to the torque applied and the clamping load wants to be a set percentage, usually 90% for a non-stretch fastener, of the yield strength. So in principle desired torque is proportional to the stud material tensile strength. So your first approximation is just to increase torque in that ratio assuming you know the tensile strength of the ARP and new material. That should take you to the same % of yield strength as the ARP studs used.
However you say the ARP studs were wasted. Well I hope not, that's a lot of money down the drain, but assuming they were waisted, down to 9.25mm diameter, then you can further increase torque by the area ratio so another 17%.
But, lubrication values, material coatings and thread surface finishes alter the coefficient of friction so much that the torque might as well just be a rough guess. Torque wrench accuracy is yet another buggeration factor. Fastener stretch is a better method but you can't measure that with a blind stud and in any case you wouldn't know how much the stretch should be! It's generally only used on through bolts for conrod big ends.
So let's go back a bit to a few basics. First off the studs will never pull out of an iron block first provided there's sufficient thread engagement. 1.5xD should be plenty but 2D would be nice. Hand tight is fine. You don't have to torque studs into a block. You will of course have specified rolled threads on the studs and not cut ones hopefully.
I always do up the nuts on critical fasteners three times to burnish the threads and standardise the friction values. You don't need the full torque, 90% will do with plenty of lube.
Then with the threads well lubed you should be able to feel the critical torque value as the stud starts to go beyond its elastic limit and it won't take any more torque. Don't be too afraid of a bit of permanent stretch. It's better to be just nudging into the non elastic area than being well short of it.
Being pragmatic, if you can apply a decent skelp more torque with the same type of oil as you used on the ARP studs, say 20%, then you should have that much more clamping load and be back into safe operation. Whether you use every ounce of the stud's yield strength is probably less important. If you want to really do that then you'll have to chart the torque versus tightening angle to see where the elastic limit is. It'll probably differ stud by stud though given the vagaries of different bits of metal.
Tell me the tensile strength of the old and new material, thread pitch, the ARP suggested torque and the lube type for that torque and I'll have a look at the numbers for you.
Pumaracing
For a given fastener size and the same thread pitch the clamping load is proportional to the torque applied and the clamping load wants to be a set percentage, usually 90% for a non-stretch fastener, of the yield strength. So in principle desired torque is proportional to the stud material tensile strength. So your first approximation is just to increase torque in that ratio assuming you know the tensile strength of the ARP and new material. That should take you to the same % of yield strength as the ARP studs used.
However you say the ARP studs were wasted. Well I hope not, that's a lot of money down the drain, but assuming they were waisted, down to 9.25mm diameter, then you can further increase torque by the area ratio so another 17%.
But, lubrication values, material coatings and thread surface finishes alter the coefficient of friction so much that the torque might as well just be a rough guess. Torque wrench accuracy is yet another buggeration factor. Fastener stretch is a better method but you can't measure that with a blind stud and in any case you wouldn't know how much the stretch should be! It's generally only used on through bolts for conrod big ends.
So let's go back a bit to a few basics. First off the studs will never pull out of an iron block first provided there's sufficient thread engagement. 1.5xD should be plenty but 2D would be nice. Hand tight is fine. You don't have to torque studs into a block. You will of course have specified rolled threads on the studs and not cut ones hopefully.
I always do up the nuts on critical fasteners three times to burnish the threads and standardise the friction values. You don't need the full torque, 90% will do with plenty of lube.
Then with the threads well lubed you should be able to feel the critical torque value as the stud starts to go beyond its elastic limit and it won't take any more torque. Don't be too afraid of a bit of permanent stretch. It's better to be just nudging into the non elastic area than being well short of it.
Being pragmatic, if you can apply a decent skelp more torque with the same type of oil as you used on the ARP studs, say 20%, then you should have that much more clamping load and be back into safe operation. Whether you use every ounce of the stud's yield strength is probably less important. If you want to really do that then you'll have to chart the torque versus tightening angle to see where the elastic limit is. It'll probably differ stud by stud though given the vagaries of different bits of metal.
Tell me the tensile strength of the old and new material, thread pitch, the ARP suggested torque and the lube type for that torque and I'll have a look at the numbers for you.
Pumaracing
FordPrefect56 said:
One of my favourite topics. In essence it's all very simple - but beset by such vagaries as to render most torque calculations hopelessly imprecise and essentially useless.
For a given fastener size and the same thread pitch the clamping load is proportional to the torque applied and the clamping load wants to be a set percentage, usually 90% for a non-stretch fastener, of the yield strength. So in principle desired torque is proportional to the stud material tensile strength. So your first approximation is just to increase torque in that ratio assuming you know the tensile strength of the ARP and new material. That should take you to the same % of yield strength as the ARP studs used.
However you say the ARP studs were wasted. Well I hope not, that's a lot of money down the drain, but assuming they were waisted, down to 9.25mm diameter, then you can further increase torque by the area ratio so another 17%.
But, lubrication values, material coatings and thread surface finishes alter the coefficient of friction so much that the torque might as well just be a rough guess. Torque wrench accuracy is yet another buggeration factor. Fastener stretch is a better method but you can't measure that with a blind stud and in any case you wouldn't know how much the stretch should be! It's generally only used on through bolts for conrod big ends.
So let's go back a bit to a few basics. First off the studs will never pull out of an iron block first provided there's sufficient thread engagement. 1.5xD should be plenty but 2D would be nice. Hand tight is fine. You don't have to torque studs into a block. You will of course have specified rolled threads on the studs and not cut ones hopefully.
I always do up the nuts on critical fasteners three times to burnish the threads and standardise the friction values. You don't need the full torque, 90% will do with plenty of lube.
Then with the threads well lubed you should be able to feel the critical torque value as the stud starts to go beyond its elastic limit and it won't take any more torque. Don't be too afraid of a bit of permanent stretch. It's better to be just nudging into the non elastic area than being well short of it.
Being pragmatic, if you can apply a decent skelp more torque with the same type of oil as you used on the ARP studs, say 20%, then you should have that much more clamping load and be back into safe operation. Whether you use every ounce of the stud's yield strength is probably less important. If you want to really do that then you'll have to chart the torque versus tightening angle to see where the elastic limit is. It'll probably differ stud by stud though given the vagaries of different bits of metal.
Tell me the tensile strength of the old and new material, thread pitch, the ARP suggested torque and the lube type for that torque and I'll have a look at the numbers for you.
Pumaracing
Thankyou.For a given fastener size and the same thread pitch the clamping load is proportional to the torque applied and the clamping load wants to be a set percentage, usually 90% for a non-stretch fastener, of the yield strength. So in principle desired torque is proportional to the stud material tensile strength. So your first approximation is just to increase torque in that ratio assuming you know the tensile strength of the ARP and new material. That should take you to the same % of yield strength as the ARP studs used.
However you say the ARP studs were wasted. Well I hope not, that's a lot of money down the drain, but assuming they were waisted, down to 9.25mm diameter, then you can further increase torque by the area ratio so another 17%.
But, lubrication values, material coatings and thread surface finishes alter the coefficient of friction so much that the torque might as well just be a rough guess. Torque wrench accuracy is yet another buggeration factor. Fastener stretch is a better method but you can't measure that with a blind stud and in any case you wouldn't know how much the stretch should be! It's generally only used on through bolts for conrod big ends.
So let's go back a bit to a few basics. First off the studs will never pull out of an iron block first provided there's sufficient thread engagement. 1.5xD should be plenty but 2D would be nice. Hand tight is fine. You don't have to torque studs into a block. You will of course have specified rolled threads on the studs and not cut ones hopefully.
I always do up the nuts on critical fasteners three times to burnish the threads and standardise the friction values. You don't need the full torque, 90% will do with plenty of lube.
Then with the threads well lubed you should be able to feel the critical torque value as the stud starts to go beyond its elastic limit and it won't take any more torque. Don't be too afraid of a bit of permanent stretch. It's better to be just nudging into the non elastic area than being well short of it.
Being pragmatic, if you can apply a decent skelp more torque with the same type of oil as you used on the ARP studs, say 20%, then you should have that much more clamping load and be back into safe operation. Whether you use every ounce of the stud's yield strength is probably less important. If you want to really do that then you'll have to chart the torque versus tightening angle to see where the elastic limit is. It'll probably differ stud by stud though given the vagaries of different bits of metal.
Tell me the tensile strength of the old and new material, thread pitch, the ARP suggested torque and the lube type for that torque and I'll have a look at the numbers for you.
Pumaracing
They were wasted, they're going in the bin
Old ARPs were 8740 which is 160,000 PSI yield strength.
New ones will be 13-8PH H950 which is 204,885 PSI yield.
Increase of 28% on those numbers alone + the extra diameter.
M10 x 1.25
ARP Ultra torque lube (blue packet)
ARPs were torqued to 85Nm
Out of interest, why would ARP reduce the shank diameter from 10 to 9.2?
FordPrefect56 said:
I find it hard to reconcile your quoted 8740 tensile at 160,000 psi with what ARP say for it (180,000 to 210,000) and with the lubed torque setting of 85 Nm (63 ft lb).
I haven't quoted a tensile number.There is a tensile figure and Yield figure, tensile is when it breaks, Yield is the point of no return (elastic limit).
What I have quoted back there is the two Yield figures of the materials, what you have quoted is tensile which is why it's a much bigger figure.
Edit:
Tensile strength of 13-8PH H950 is 220,000
The old ARP 8740 is 180,000 to 210,000
I wonder why ARP can't be accurate, It's not going to be very precise if we use their figures in an equation....
Edited by Evoluzione on Wednesday 13th July 16:31
Evoluzione said:
I haven't quoted a tensile number.
There is a tensile figure and Yield figure, tensile is when it breaks, Yield is the point of no return (elastic limit).
What I have quoted back there is the two Yield figures of the materials, what you have quoted is tensile which is why it's a much bigger figure.
Sorry. I asked for tensile but got yield so my tiny brain got confused when I didn't spot that. Yield is really neither here nor there if tensile is much higher. You can pull the studs into non-elastic stretch and get a bit more out of them. I still see 8740 as about 180,000 psi yield and over 200,000 tensile. Anyhoo. Do them up f**king tight and see what happens.There is a tensile figure and Yield figure, tensile is when it breaks, Yield is the point of no return (elastic limit).
What I have quoted back there is the two Yield figures of the materials, what you have quoted is tensile which is why it's a much bigger figure.
FordPrefect56 said:
Evoluzione said:
I haven't quoted a tensile number.
There is a tensile figure and Yield figure, tensile is when it breaks, Yield is the point of no return (elastic limit).
What I have quoted back there is the two Yield figures of the materials, what you have quoted is tensile which is why it's a much bigger figure.
Sorry. I asked for tensile but got yield so my tiny brain got confused when I didn't spot that. Yield is really neither here nor there if tensile is much higher. You can pull the studs into non-elastic stretch and get a bit more out of them. I still see 8740 as about 180,000 psi yield and over 200,000 tensile. Anyhoo. Do them up f**king tight and see what happens.There is a tensile figure and Yield figure, tensile is when it breaks, Yield is the point of no return (elastic limit).
What I have quoted back there is the two Yield figures of the materials, what you have quoted is tensile which is why it's a much bigger figure.
"f**king tight" That's not very accurate....
stevieturbo said:
Evoluzione said:
Out of interest, why would ARP reduce the shank diameter from 10 to 9.2?
I'm near sure I read somewhere this actually allows the stud to clamp more ? Or perhaps more evenly somehow ?The thread forming process causes the material to move about, the material from the bottom of the thread 'squidges' up towards the top of the thread, as it 'Squidges' it also becomes denser and work hardens resulting in a much stronger thread.
The ones the OP is having made with a 10mm shank will start as a piece of 10mm dia bar and it will be turned down to about 9.15mm dia +/- a bit less than f
k over the area where it will be rolled.Page 4 shows the pre-rolling sizes required depending on thread pitch.
http://www.gewinde.ch/files/Eichenberger-Rolled-Th...
The 9.25mm shank is not the minimum cross sectional area anyway, the minimum area is the tensile stress area (58mm^2 for M10) basically at the root of the threads (approx) that is the governing geometric area. As a start point id prorata the torque from the ratio of yield stress and make sure i recheck the torque a few times after heat cyling the head.
Tango13 said:
ARP don't 'reduce' the shank of the stud, they started with a 9.2mm dia bar because it's easier and thus cheaper.
The thread forming process causes the material to move about, the material from the bottom of the thread 'squidges' up towards the top of the thread, as it 'Squidges' it also becomes denser and work hardens resulting in a much stronger thread.
The ones the OP is having made with a 10mm shank will start as a piece of 10mm dia bar and it will be turned down to about 9.15mm dia +/- a bit less than fk over the area where it will be rolled.
Page 4 shows the pre-rolling sizes required depending on thread pitch.
http://www.gewinde.ch/files/Eichenberger-Rolled-Th...
There are many studs with quite a considerable difference in stem diameter vs thread diameter though, and there is definitely a deliberate reason for them doing it. I just dont recall fully what it was.The thread forming process causes the material to move about, the material from the bottom of the thread 'squidges' up towards the top of the thread, as it 'Squidges' it also becomes denser and work hardens resulting in a much stronger thread.
The ones the OP is having made with a 10mm shank will start as a piece of 10mm dia bar and it will be turned down to about 9.15mm dia +/- a bit less than f
k over the area where it will be rolled.Page 4 shows the pre-rolling sizes required depending on thread pitch.
http://www.gewinde.ch/files/Eichenberger-Rolled-Th...
stevieturbo said:
There are many studs with quite a considerable difference in stem diameter vs thread diameter though, and there is definitely a deliberate reason for them doing it. I just dont recall fully what it was.
less chance for relaxation as theres more stretch to induce the preload. bolts with big neck down are often TTY bolts and they can "tune" the bolt by varying the geometryAlso the bolt sees a lower amplitude of fatigue stress when it is more flexible (longer and smaller diameter) but the mechanical properties need to be superior to be able to develop the right amount of preload
Inline__engine said:
less chance for relaxation as theres more stretch to induce the preload. bolts with big neck down are often TTY bolts and they can "tune" the bolt by varying the geometry
Also the bolt sees a lower amplitude of fatigue stress when it is more flexible (longer and smaller diameter) but the mechanical properties need to be superior to be able to develop the right amount of preload
I didnt mention bolts, I mean proper studs, not TTYAlso the bolt sees a lower amplitude of fatigue stress when it is more flexible (longer and smaller diameter) but the mechanical properties need to be superior to be able to develop the right amount of preload
The reason you aren't going to be able to nail down an exact yield or tensile strength for 8740 is it has a huge range depending on heat treatment and exact metallurgical composition. Anyhoo, ignoring all that and checking the specs for your new material in H950 condition online I come up with 98 Nm using ARP lube.
stevieturbo said:
I didnt mention bolts, I mean proper studs, not TTY
i realize that, but its the same primciple, often TTY neck down more but thats just an observation. if you induce a given preload force the thinner diameter stretches more and that is better joint. but you also need a certain amount of preload and that requires a certain minimum CSAEdited by Inline__engine on Thursday 14th July 01:48
Edited by Inline__engine on Thursday 14th July 01:50
FordPrefect56 said:
The reason you aren't going to be able to nail down an exact yield or tensile strength for 8740 is it has a huge range depending on heat treatment and exact metallurgical composition. Anyhoo, ignoring all that and checking the specs for your new material in H950 condition online I come up with 98 Nm using ARP lube.
Thanks for that, ARPs site says 95NM to, so not so far away. They also say "Recommended torque is equal to 75% of the fastener's yield strength." Which is useful.I think ARP are being deliberately vague and evasive over their materials to protect their interests, they will know what they have used on these studs, but the chances of me finding out are pretty slim.
Edited by Evoluzione on Thursday 14th July 13:55
I had T&Kprecision make me some studs, and i ordered enough extra so i could sacrifice a few to some loading tests, measuring torques, stretch and load (i borrowed some load cells from work.....) in order to characterise them properly. I also sacrificed a spare head and block to measure installed clamp load and estimate the head stiffness......
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff