Bolt torque discussion

-
krazykuda said:
The clamp load is more than the load seen by the fastener... If the load on the bolt exceeds the clamp load on the fastener, then you will get failure/breaking of the bolt...

The bolt acts kind of like a spring, the clamp load is what keeps it together...
Click to expand...
I saw a video by one of the major head gasket manufacturers and they were saying they had made a custom
Machine that monitors how much a cylinder head lifted off the block as the piston went through its cycles. Yes thats right lifted.
Although the head appears to be stationary during engine operation, the head actually lifts
A very small amount unloading the head gasket. If this unloading is excessive, the gasket will fail. The elasticity of the bolts allows this to happen
The sbm with its 4 bolts only per cylinder was particularly prone to this at high compression ratios.
The reason the Mls style head gaskets were a big help to the troublesome sbm is the spring steel can move with the head but still retain a seal and spring steel is noted for its ability to retain its shape.
So a certain amount of bolt stretch during operation is considered normal.
 
krazykuda said:
You have to be careful on what type of oil that you put on them... different oils have different lubrication properties and can affect the clamp load for a particular torque...

80% of the torque value is influenced by the friction of the fastener....
Click to expand...
And the finish surface on the threads makes large difference too, as well as any plating.

What most users do't realize with the ARP's is that they are not going up by 40-50% or in bolt tension/clamping force by going from, say, 85 to 130 ft lbs torque. Because the ARP lube is so much slicker than oil, the torquing friction is a lot lower and it takes lot more tension in the fastener on on the threads to get the torque friction up. The actual force achieved is somewhere in the 100% higher range over what the stock fastener torqued to 85 ft lbs with oil would achieve. That is a very rough number but you are in the ballpark with that %.

IIRC (and this is going back a while): The threads in the cast block are considerably stronger than the steel bolts/studs. Again, IIRC, this is due to the material properties of the cast iron.
 
Duane said:
I saw a video by one of the major head gasket manufacturers and they were saying they had made a custom
Machine that monitors how much a cylinder head lifted off the block as the piston went through its cycles. Yes thats right lifted.
Although the head appears to be stationary during engine operation, the head actually lifts
A very small amount unloading the head gasket. If this unloading is excessive, the gasket will fail. The elasticity of the bolts allows this to happen
The sbm with its 4 bolts only per cylinder was particularly prone to this at high compression ratios.
The reason the Mls style head gaskets were a big help to the troublesome sbm is the spring steel can move with the head but still retain a seal and spring steel is noted for its ability to retain its shape.
So a certain amount of bolt stretch during operation is considered normal.
Click to expand...
Yes, the combustion pressure against the head is taken by the bolts so the idea that the external forces don't end up on the bolts is incorrect. With the extra peak force on the head from combustion, each head bolt around a cylinder will see peak tension increase by a few thousand pounds, and then go back to just the tension from torquing. That increase in tension will stretch the bolts a tiny bit each combustion cycle before they relax back to the original length.

The solution to getting less movement each combustion cycle is to use a thicker bolt/stud, or one with a higher Young's modulus, also known as the modulus of elasticity. That is what ARP is doing with different ('stronger') stud/bolt materials.

Even then, you need more clamping force because the head is going to bow between the fasteners each cycle. More torque on a stock bolt might trap the gasket better, but can you an only get so much improvement before it reaches its limit in other ways.
 
Duane said:
One of the previous posters said to think of a bolt or stud as a spring. If you were trying achieve a certain amount of tension from that spring, a longer spring would need more stretching before it got to the same tension as a shorter spring.
Click to expand...
Yes, good analogy. Do you think this is what was being sought with the higher torque? ==> Achieving given amount of strain (stretch proportional to length), which directly relates to stress, which relates directly to tension. Just trying to close the circle on what that was about....
 
nm9stheham said:
Yes, good analogy. Do you think this is what was being sought with the higher torque? ==> Achieving given amount of strain (stretch proportional to length), which directly relates to stress, which relates directly to tension. Just trying to close the circle on what that was about....
Click to expand...
I believe this is what the poster was after. It was only 10 ft lbs difference but different none the less and I believe that the one longer stud of the longer ones again he went another 10
 
Duane said:
I believe this is what the poster was after. It was only 10 ft lbs difference but different none the less and I believe that the one longer stud of the longer ones again he went another 10
Click to expand...
OK tnx. If he had no other info on the stud/thread behavior, then this really is just a pure guess and who knows what he got for tension. But if he put a dial indicator on each stud and made the stretch per unit length (strain) equal on each stud.... or had some other knowledge that would tell him it would behave that way...... then he did what he intended.
 
I did some rod bolt torquing today and that is where I throw everything I think about bolts out the window. Now we go by the book. Here was my dilemma. I bought a used set of Crower Rods and assembly. No other marketing on the rods bolts other than Crower so I measured the bolts and that are around 1.800 long- 7/16. The specksheat says 65 pounds .004-.006 stretch with 20-50 oil. The next bolt down calls for 75 foot pounds .005-.007 stretch with their lube. I decide on aiming for a hair over .005 with 20-50 oil. 72 foot pounds is my number for this build.
 
nm9stheham said:
Yes, the combustion pressure against the head is taken by the bolts so the idea that the external forces don't end up on the bolts is incorrect. With the extra peak force on the head from combustion, each head bolt around a cylinder will see peak tension increase by a few thousand pounds, and then go back to just the tension from torquing. That increase in tension will stretch the bolts a tiny bit each combustion cycle before they relax back to the original length.

The solution to getting less movement each combustion cycle is to use a thicker bolt/stud, or one with a higher Young's modulus, also known as the modulus of elasticity. That is what ARP is doing with different ('stronger') stud/bolt materials.

Even then, you need more clamping force because the head is going to bow between the fasteners each cycle. More torque on a stock bolt might trap the gasket better, but can you an only get so much improvement before it reaches its limit in other ways.
Click to expand...

This is correct I was remembering my notes wrong; I found them dated from 6 years ago so it's been a minute. The part I was thinking of is how a fastener is preloaded (torqued) it causes external forces to be distributed evenly among the parts and all the fasteners at once; for example you have 4 head bolts torqued down but one is actually less than the others due to an assembly error or whatnot, the other 3 bolts will end up taking most of the external forces (combustion pressure) and you'll have greatly reduced clamping loads on the parts. Not really anything new to what has already been said in the thread so far lol.
 
pittsburghracer said:
I did some rod bolt torquing today and that is where I throw everything I think about bolts out the window. Now we go by the book. Here was my dilemma. I bought a used set of Crower Rods and assembly. No other marketing on the rods bolts other than Crower so I measured the bolts and that are around 1.800 long- 7/16. The specksheat says 65 pounds .004-.006 stretch with 20-50 oil. The next bolt down calls for 75 foot pounds .005-.007 stretch with their lube. I decide on aiming for a hair over .005 with 20-50 oil. 72 foot pounds is my number for this build.
Click to expand...

Now I feel like I need a bolt stretch gauge... I'll probably get one before my next engine build which hopefully won't be for a couple more years I'm pooped from my last one
 
E8D499EA-4DBD-4BFA-BAD7-F12B474F90F4.jpeg
MopaR&D said:
Now I feel like I need a bolt stretch gauge... I'll probably get one before my next engine build which hopefully won't be for a couple more years I'm pooped from my last one
Click to expand...


Very nice tool once you get used to setting it up and using it.
 
pittsburghracer said:
I did some rod bolt torquing today and that is where I throw everything I think about bolts out the window. Now we go by the book. Here was my dilemma. I bought a used set of Crower Rods and assembly. No other marketing on the rods bolts other than Crower so I measured the bolts and that are around 1.800 long- 7/16. The specksheat says 65 pounds .004-.006 stretch with 20-50 oil. The next bolt down calls for 75 foot pounds .005-.007 stretch with their lube. I decide on aiming for a hair over .005 with 20-50 oil. 72 foot pounds is my number for this build.
Click to expand...
If you don,t mind can you describe your procedure on how you use the stretch gauge on a 7/16 bolt. Do you set the gauge
To zero and then torque with gauge removed or gauge in place and use a box end wrench.
I found I had to change my procedure with 7/16 bolt versus
a 3/8 bolt because of the torque requirement too high to get with a box end wrench.
 
I just tore down engine to change a tight bearing and engine had not been run. I had used arp lube on the arp studs and nuts. The lube had turned hard and flakey. I don't like that inside engine especially by crank. Has anyone else seen this.
 
pittsburghracer said:
I did some rod bolt torquing today and that is where I throw everything I think about bolts out the window. Now we go by the book. Here was my dilemma. I bought a used set of Crower Rods and assembly. No other marketing on the rods bolts other than Crower so I measured the bolts and that are around 1.800 long- 7/16. The specksheat says 65 pounds .004-.006 stretch with 20-50 oil. The next bolt down calls for 75 foot pounds .005-.007 stretch with their lube. I decide on aiming for a hair over .005 with 20-50 oil. 72 foot pounds is my number for this build.
Click to expand...
FWIW.... Brings up some points to consider..... Stretch is the best method to get the desired fastener tension BUT the accuracy of the measurements for fasteners of this size becomes the problem..... .005" +/-.001" is a +/- 20% accuracy, plus any errors from angling of the gauge off-axis. Not as consistent as you might think. Torquing has more accuracy/resolution in a good wrench and operator reading it, but then the friction variables come in to effect the results, despite better resolution on the equipment.

If you pick the stretch method, the torque really just becomes a check on whether things are getting way out of whack.
 
a couple of observations
on head lifting- on the turbo Rambler project we used German Bellevue washer/ 0 rings and the heads did dance around (I have no remembrance of where Barney got them)
on the SBC we used necked down bolts for the short ones so we could get same stretch as the long ones- these are still available

on the cracked heads are you using hard AN style washers with the bevel to spread the load?
in aluminum blocks we used a larger thread in the aluminum than on the nut end
bolts can be too stiff and thus loose clamp easier- you do need the spring/ stretch effect
using a grad 8 or better bolt in a light load where a 5 ws specified is almost sure to give you a too stiff/ hard bolt
to make a grade 8 work you have to torque it to grade 8 torque which just does not work on say valve bodies (warp)or water pumps (strip) etc-- use the correct bolt
 
nm9stheham said:
FWIW.... Brings up some points to consider..... Stretch is the best method to get the desired fastener tension BUT the accuracy of the measurements for fasteners of this size becomes the problem..... .005" +/-.001" is a +/- 20% accuracy, plus any errors from angling of the gauge off-axis. Not as consistent as you might think. Torquing has more accuracy/resolution in a good wrench and operator reading it, but then the friction variables come in to effect the results, despite better resolution on the equipment.

If you pick the stretch method, the torque really just becomes a check on whether things are getting way out of whack.
Click to expand...
I like to think of the stretch method as taking the wrench error and thread friction out of the equation.
 
nm9stheham said:
Yep, I am pretty sure I understand all that KK; tnx. Did you mean 'maximum tension (stress) with minimal stretch' at the end of paragraph 2?

What I wanted to try to understand is the 'why' behind thinking that the longer head bolts needed to being torqued to a higher value, as Duane brought up. What is the objective of more torque with a longer head bolt? Are they trying to get to a more consistent stress/tension/clamping value? And are the looking at proportional stretch (strain) to judge the right value of torque, rather than just torque?
Click to expand...

The longer bolt for the most part doesn't need more torque than the rest of them as it would take a significant amount more length to make much difference... Yes bolt length can affect the torque on a bolt, but not as much as the thread diameter, pitch, size of the bolt (diameter), and grade of the bolt (hardness of the metal)... It would take much more amount of length to require the bolt to be torqued to a higher amount for the same diameter of bolt...

The clamp load on the bolt is what holds the joint together... The load on the bolt acts in the opposite direction of the clamp load... The clamp load holds the joint together until it sees a load/force higher than itself... Once the load on the bolt exceeds the clamp load, then the joint will separate and that's when the parts break... It's like push-pull... The clamp load is pushing the joint together, the force on the bolt is trying to pull it apart...
 
nm9stheham said:
Yep, I am pretty sure I understand all that KK; tnx. Did you mean 'maximum tension (stress) with minimal stretch' at the end of paragraph 2?

What I wanted to try to understand is the 'why' behind thinking that the longer head bolts needed to being torqued to a higher value, as Duane brought up. What is the objective of more torque with a longer head bolt? Are they trying to get to a more consistent stress/tension/clamping value? And are the looking at proportional stretch (strain) to judge the right value of torque, rather than just torque?
Click to expand...

yes, I went back and edited a few things in my posts as there were a few typos and I didn't have time to read and correct them earlier...
 
nm9stheham said:
And the finish surface on the threads makes large difference too, as well as any plating.

What most users do't realize with the ARP's is that they are not going up by 40-50% or in bolt tension/clamping force by going from, say, 85 to 130 ft lbs torque. Because the ARP lube is so much slicker than oil, the torquing friction is a lot lower and it takes lot more tension in the fastener on on the threads to get the torque friction up. The actual force achieved is somewhere in the 100% higher range over what the stock fastener torqued to 85 ft lbs with oil would achieve. That is a very rough number but you are in the ballpark with that %.

IIRC (and this is going back a while): The threads in the cast block are considerably stronger than the steel bolts/studs. Again, IIRC, this is due to the material properties of the cast iron.
Click to expand...

Yes, surface finish affects the surface friction of the threads, which also will influence the torque on the bolt...

The coating/finish on the bolt also affects the surface friction... The surface friction is along the threads and the under side of the bolt head... This is what is 'dragging' as you try to turn/torque the bolt...

We used the standard phosphorus and oil on the connecting rod bolts in our factory... We used a special coating on the nuts that was more like a wax than an oil... It was dry to the touch but had a slight white 'haze' on the nut... If an engine was made and they found a defect in it and then tore it down and reused the parts, sometimes the connecting rod nuts would get dirty and need to be cleaned before we could reuse them... The cleaning would strip off the coating and they had to be re-coated... They tried to use different oils in place of the wax coating on the nuts, but it didn't work... The oil made the nuts much slicker and then they would fail in the torque multiple that torqued the connecting rod bolts... A 'torque multiple' is a machine with more than one spindle that can tighten more than one nut at a time... We tightened four nuts for two opposite cylinders at the same time with these machines... When we tried to re-oil the nuts our self in house, they would reject in the multiples more often... The repair loop for the connecting rods could only handle about 10 engines max... If we got a rash of rejects in a short amount of time, it would fill the repair station and then the next reject would stop the line as it could not get into the repair loop as the line was programmed not to let a reject pass by the repair loop...

So what we did then was send all the connecting rod nuts back to the supplier to be cleaned and re-coated with the wax so we could re-use them... We would also batch run the recycled nuts so we could keep track of them if they started to reject in the multiples too much... If they started rejecting too much, we would then remove them from the assembly line and use fresh boxes of brand new ones...

That is why I don't recommend using any extra lube on the nuts and bolts... The different oils can throw off the torque and clamp load... All fasteners at the factory are used as the supplier ships them in and all torque specs are made for those conditions... Only in the case like ARP where they supply the lube with the bolts and that is what they used to develop the torques with should you use lube on a bolt...
 
Last edited:
Duane said:
I like to think of the stretch method as taking the wrench error and thread friction out of the equation.
Click to expand...
Yes. I'd be curious to know how stretch measurement errors compare to the friction errors form torquing.
 
krazykuda said:
We used the standard phosphorus and oil on the connecting rod bolts in our factory... We used a special coating on the nuts that was more like a wax than an oil... It was dry to the touch but had a slight white 'haze' on the bolt... If an engine was made and they found a defect in it and then tore it down and reused the parts, sometimes the connecting rod nuts would get dirty and need to be cleaned before we could reuse them... The cleaning would strip off the coating and they had to be re-coated... They tried to use different oils in place of the wax coating on the nuts, but it didn't work... The oil made the nuts much slicker and then they would fail in the torque multiple that torqued the connecting rod bolts... A 'torque multiple' is a machine with more than one spindle that can tighten more than one nut at a time... We tightened four nuts for two opposite cylinders at the same time with these machines... When we tried to re-oil the nuts our self in house, they would reject in the multiples more often... The repair loop for the connecting rods could only handle about 10 engines max... If we got a rash of rejects in a short amount of time, it would fill the repair station and then the next reject would stop the line as it could not get into the repair loop as the line was programmed not to let a reject pass by the repair loop...

So what we did then was send all the connecting rod nuts back to the supplier to be cleaned and re-coated with the wax so we could re-use them... We would also batch run the recycled nuts so we could keep track of them if they started to reject in the multiples too much... If they started rejecting too much, we would then remove them from the assembly line and use fresh boxes of brand new ones...

That is why I don't recommend using any extra lube on the nuts and bolts... The different oils can throw off the torque and clamp load... All fasteners at the factory are used as the supplier ships them in and all torque specs are made for those conditions... Only in the case like ARP where they supply the lube with the bolts and that is what they used to develop the torques with should you use lube on a bolt...
Click to expand...
 
krazykuda said:
We used the standard phosphorus and oil on the connecting rod bolts in our factory... We used a special coating on the nuts that was more like a wax than an oil... It was dry to the touch but had a slight white 'haze' on the bolt... If an engine was made and they found a defect in it and then tore it down and reused the parts, sometimes the connecting rod nuts would get dirty and need to be cleaned before we could reuse them... The cleaning would strip off the coating and they had to be re-coated... They tried to use different oils in place of the wax coating on the nuts, but it didn't work... The oil made the nuts much slicker and then they would fail in the torque multiple that torqued the connecting rod bolts... A 'torque multiple' is a machine with more than one spindle that can tighten more than one nut at a time... We tightened four nuts for two opposite cylinders at the same time with these machines... When we tried to re-oil the nuts our self in house, they would reject in the multiples more often... The repair loop for the connecting rods could only handle about 10 engines max... If we got a rash of rejects in a short amount of time, it would fill the repair station and then the next reject would stop the line as it could not get into the repair loop as the line was programmed not to let a reject pass by the repair loop...

So what we did then was send all the connecting rod nuts back to the supplier to be cleaned and re-coated with the wax so we could re-use them... We would also batch run the recycled nuts so we could keep track of them if they started to reject in the multiples too much... If they started rejecting too much, we would then remove them from the assembly line and use fresh boxes of brand new ones...

That is why I don't recommend using any extra lube on the nuts and bolts... The different oils can throw off the torque and clamp load... All fasteners at the factory are used as the supplier ships them in and all torque specs are made for those conditions... Only in the case like ARP where they supply the lube with the bolts and that is what they used to develop the torques with should you use lube on a bolt...
Click to expand...
Thanks very much for sharing all that KK... fascinating. (Well, sorta.... LOL) Question: How did the torques 'fail' when using the oil vs mfr supplied wax and how was that detected? Too many fractional turns to get to a torque or ??? Or did the multiple end up with a twisting force on it? Just curious.

In the field, for most of us who will continue to use torque wrenches, it seems like consistently using a given oil on the threads is the only real thing to do. So if the FSM says 'light oil', what is that 'light oil'?

BTW, I learned/was taught to torque thing dry, or at least wiped clean of any obviously liquid oil... so that is another matter.
 
nm9stheham said:
Thanks very much for sharing all that KK... fascinating. (Well, sorta.... LOL) Question: How did the torques 'fail' when using the oil vs mfr supplied wax and how was that detected? Too many fractional turns to get to a torque or ??? Or did the multiple end up with a twisting force on it? Just curious.
Click to expand...

We used torque and angle to measure the clamp load... If we tried other oils, they would most likely fail in the multiple for angle...

We set a torque spec that we start measuring angle at, let's say 40 ft * lbs, so at 40 ft *lb the angle is now "set" to zero and we measure from there.. We would typically see angle run in the range of about 70° - 90° and we set a range that angle would be for the final torque... With "foreign oil" the angle readings would be higher and outside our angle range that was programmed in the machine...
 
Last edited:
nm9stheham said:
In the field, for most of us who will continue to use torque wrenches, it seems like consistently using a given oil on the threads is the only real thing to do.
Click to expand...

the service manual torques are for using a torque wrench... If you use the torque that is recommended in the service manual, you will be fine...
 
You must log in or register to reply here.
-

Back
Top