Guys, I have a very basic question: in a bolted joint, does using a threaded stud beyond a certain length gain me anything in strength? Specifically, I have an aluminum racing engine block with threaded stud holes 2.7" deep for 1/2-13 SAE studs holding the aluminum heads. ARP doesn't make such a stud off-the-shelf, and I wonder if all that threaded length is necessary. The block mfgr is out of business so can't ask him directly. Thanks-
Eng-Tips is the largest engineering community on the Internet
Intelligent Work Forums for Engineering Professionals
-
Congratulations waross on being selected by the Tek-Tips community for having the most helpful posts in the forums last week. Way to Go!
Stud grip length vs joint strength 3
- Thread starter Jsgarage
- Start date
patprimmer
New member
You get to a point where the holding strength of the thread is greater than the strength of the stud. A little extra length to cater for inaccuracies in thread fit or fatigue of the aluminium can't hurt, but much more is no benefit.
Others here might have a formula for calculating that, but the difficulty will be knowing the strength of the grade of aluminium used.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
Others here might have a formula for calculating that, but the difficulty will be knowing the strength of the grade of aluminium used.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
-
2
- #3
The rule of thumb for aluminum is that the engaged thread length should be 2.5-3X the thread diameter. So a 10mm thread diameter would typically have 25-30mm of engagement in the block. For steel it's 1-1.5X and iron is typically 1.5-2X.
More threads won't buy you anything because the block threads will shear before the threads on a high tensile stud due to the fact the first ten threads carry most of the tensile load.
More threads won't buy you anything because the block threads will shear before the threads on a high tensile stud due to the fact the first ten threads carry most of the tensile load.
- Thread starter
- #4
Lou Scannon
Automotive
"More threads won't buy you anything because the block threads will shear before the threads on a high tensile stud due to the fact the first ten threads carry most of the tensile load." I.e., a domino type failure; the first ten threads fail, followed by the 11th, etc.
-
1
- #6
Jsgarage,
For an optimum thread engagement length with a steel stud into an aluminum substrate, you should have enough threads engaged so that the pull-out strength of the stud into the aluminum is equal to (or greater) than the tensile strength of the stud. The pull-out strength of the stud is based on the root shear strength of the aluminum it's threaded into. The root shear area is basically the dimension across the base of the thread crest section times the total helical length of the engaged threads. The simple shear strength is just then load/area.
If you wish to be more accurate in your analysis you should also apply the appropriate knock-down factors (Kt) for stress concentrations.
For your example, assuming your block is cast from A356-t6 aluminum, then it has a shear strength of about 17ksi. If your tapped holes are 1/2-13 UNC x 2.7 inch, then the aluminum threads would have a simple pull-out strength of just over 32,000 lbs if all the threads are loaded. And if you assume that your 1/2-13 UNC ARP stud has a tensile strength of 180ksi, then its simple tensile strength (P/A) is somewhere around 23,000 lbf. If you want to equalize the strengths of your stud and head, then you can safely reduce your thread engagement to about 1.94 inches.
Hope that helps.
Terry
For an optimum thread engagement length with a steel stud into an aluminum substrate, you should have enough threads engaged so that the pull-out strength of the stud into the aluminum is equal to (or greater) than the tensile strength of the stud. The pull-out strength of the stud is based on the root shear strength of the aluminum it's threaded into. The root shear area is basically the dimension across the base of the thread crest section times the total helical length of the engaged threads. The simple shear strength is just then load/area.
If you wish to be more accurate in your analysis you should also apply the appropriate knock-down factors (Kt) for stress concentrations.
For your example, assuming your block is cast from A356-t6 aluminum, then it has a shear strength of about 17ksi. If your tapped holes are 1/2-13 UNC x 2.7 inch, then the aluminum threads would have a simple pull-out strength of just over 32,000 lbs if all the threads are loaded. And if you assume that your 1/2-13 UNC ARP stud has a tensile strength of 180ksi, then its simple tensile strength (P/A) is somewhere around 23,000 lbf. If you want to equalize the strengths of your stud and head, then you can safely reduce your thread engagement to about 1.94 inches.
Hope that helps.
Terry
YvesLLewelyn
Mechanical
As ever, I tend to agree with TB. It is notable that if you add "a bit to be on the safe side" to his 1.94 inches it is not a lot different to the original 2.7 inches. I would be inclined to stick with the original length - quite possibly the makers of the block knew what they were doing.
On a different point - I know from personal experience that it is very difficult to have a stud made that is anywhere near as strong, tough and unbreakable as an ARP (or some other good brand) stud would be.
Maybe you can find a good-quality bolt that will do the job. I did this - the really extra good bolts I used were $7 each when a normal very good bolt was about 50 cents at the time.
On a different point - I know from personal experience that it is very difficult to have a stud made that is anywhere near as strong, tough and unbreakable as an ARP (or some other good brand) stud would be.
Maybe you can find a good-quality bolt that will do the job. I did this - the really extra good bolts I used were $7 each when a normal very good bolt was about 50 cents at the time.
The problem with theoretical and actual thread load length however is that the load is not spread uniformly across the threads as I mentions above. As a result the highest loaded threads strip and then the load transfers to the next threads and they also strip. In a perfect world if the load was spread relatively uniformly across the thread length, then you could always get increasing strength with longer threads but in reality this does not happen.
In regards to bolts vs. studs there are other issues besides clamp force. Bolts impart much more block distortion in typical thin wall castings compared to studs which generate primarily a tensile loading. In soft alloys like aluminum, bolts can easily gall and damage the threads in addition to causing inaccurate tightening torques and variations in clamp loads.
In regards to bolts vs. studs there are other issues besides clamp force. Bolts impart much more block distortion in typical thin wall castings compared to studs which generate primarily a tensile loading. In soft alloys like aluminum, bolts can easily gall and damage the threads in addition to causing inaccurate tightening torques and variations in clamp loads.
TrackRat,
I agree with your points about load distribution within the threads. I was just trying to simplify things.
Generally, when you exceed a thread engagement beyond about 5P, the thread pitch tolerance accumulation (especially with a UNC thread) pretty much guarantees that you will not have equal load sharing in the thread engagement. With a 1/2-13 UNC thread engaged over 1.94 inches, that means around 25 thread pitches. Not only will those 25 threads not load share, you'd be lucky to get the stud to thread all the way down 1.94 inches without seizing up, due to the pitch errors.
As a rule of thumb, you should never have a thread engagement beyond about 1.5 or 2.0D. That would mean about 13P max for this application. If you need to equalize the root shear and tensile strengths of the base material and stud, then you should use a larger diameter thread on the end of the stud going into the aluminum.
Regards,
Terry
I agree with your points about load distribution within the threads. I was just trying to simplify things.
Generally, when you exceed a thread engagement beyond about 5P, the thread pitch tolerance accumulation (especially with a UNC thread) pretty much guarantees that you will not have equal load sharing in the thread engagement. With a 1/2-13 UNC thread engaged over 1.94 inches, that means around 25 thread pitches. Not only will those 25 threads not load share, you'd be lucky to get the stud to thread all the way down 1.94 inches without seizing up, due to the pitch errors.
As a rule of thumb, you should never have a thread engagement beyond about 1.5 or 2.0D. That would mean about 13P max for this application. If you need to equalize the root shear and tensile strengths of the base material and stud, then you should use a larger diameter thread on the end of the stud going into the aluminum.
Regards,
Terry
Terry-
I think we're on the same page here. I was trying to clarify that increasing the thread length beyond 2.5-3X in aluminum was not going to increase grip strength because the load is not evenly distributed. In this particular case going beyond 3X 1/2", i.e. 1.5" of thread, does nothing to increase thread strength. I'm surprised to see designers who use 2.7" deep threads when this does not make for a better design and can in fact cause block distortion or issues in other areas.
I think we're on the same page here. I was trying to clarify that increasing the thread length beyond 2.5-3X in aluminum was not going to increase grip strength because the load is not evenly distributed. In this particular case going beyond 3X 1/2", i.e. 1.5" of thread, does nothing to increase thread strength. I'm surprised to see designers who use 2.7" deep threads when this does not make for a better design and can in fact cause block distortion or issues in other areas.
YvesLLewelyn
Mechanical
Rather than try to calculate and speculate on the thread length of studs - what is the stud diameter/thread length on some well-known alloy-block engines like the Chev LS series V-8's and the Rover V-8's etc. Probably a good indication of what is needed.
There's no guessing required. That was the point in the above discussion. The info. supplied has been tested and confirmed. That's why I find it odd that some designers would use thread lenths much longer than usable. Doing so increase costs, machine time, tooling wear, etc. It makes me wonder if they bothered to research the subject or if they just drew the deepest thread they could fit in the space assuming more is better? <LOL>
YvesLLewelyn
Mechanical
TR - My intention was that it would be interesting to compare calculated and actual thread lengths used by manufacturers.
And yes, - I am a great believer in "(a bit) more is better". Proper engineering calculation and design and then being slightly generous in material thickness, bolt diam., thread length etc.
Unless absolute minimum lightness etc. is required (as in F1 for example) "a bit more" probably is better.
I recently saw on TV the front wheels and suspension come off an F1 car when the drivers braked - I bet the designer thought to himself "I should have used a bit of extra material on those suspension arms". I am sure the calculations said they would be strong enough.
And yes, - I am a great believer in "(a bit) more is better". Proper engineering calculation and design and then being slightly generous in material thickness, bolt diam., thread length etc.
Unless absolute minimum lightness etc. is required (as in F1 for example) "a bit more" probably is better.
I recently saw on TV the front wheels and suspension come off an F1 car when the drivers braked - I bet the designer thought to himself "I should have used a bit of extra material on those suspension arms". I am sure the calculations said they would be strong enough.
A simple tensile tests tells you what the grip strength is for various thread depths and materials. It's very easy to conduct. That's why I was surprised that some designers haven't bothered to investigate this information. The 2.5-3X depth in aluminum gives you an extra safety margin, so going beyond that is a waste of time and energy. They are probably operating on the principle that the thread load is uniformly distributed and thus more thread length = better, but of course this is not true.
patprimmer
New member
This al sounds a bit variable to me.
The claim that only so many threads contribute because of accuracy of the threads must vary according to the accuracy of various thread forming processes and also with the stress strain and elongation at yield points for the materials and for temperature changes and various co efficients of thermal expansion and rate of application of load and of fatigue strength and on any cold flow effects and even on the extra bit of load sharing that comes from gap filling compounds like Loctite.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
The claim that only so many threads contribute because of accuracy of the threads must vary according to the accuracy of various thread forming processes and also with the stress strain and elongation at yield points for the materials and for temperature changes and various co efficients of thermal expansion and rate of application of load and of fatigue strength and on any cold flow effects and even on the extra bit of load sharing that comes from gap filling compounds like Loctite.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
YvesLLewelyn,
If you read my post of June 21st, I gave a design approach for determining the optimum number of engaged threads with a steel stud and aluminum substrate. While my suggested analytical approach was somewhat simplified (for the sake of brevity), it certainly was not rote speculation. And even a simplified analytical approach is certainly better than copying some other engine's fastener installation. Whose loading conditions, fatigue life, materials, factors of safety, tolerances, stress concentration factors, etc. you likely know nothing about.
I hope that Jsgarage considers the suggestions presented to him and takes the time to gain a greater understanding of the engineering principles involved in threaded fastener joints. That's the purpose of this website.
Regards,
Terry
If you read my post of June 21st, I gave a design approach for determining the optimum number of engaged threads with a steel stud and aluminum substrate. While my suggested analytical approach was somewhat simplified (for the sake of brevity), it certainly was not rote speculation. And even a simplified analytical approach is certainly better than copying some other engine's fastener installation. Whose loading conditions, fatigue life, materials, factors of safety, tolerances, stress concentration factors, etc. you likely know nothing about.
I hope that Jsgarage considers the suggestions presented to him and takes the time to gain a greater understanding of the engineering principles involved in threaded fastener joints. That's the purpose of this website.
Regards,
Terry
patprimmer
New member
A test is only as good as the variable it controls and how well it replicates in use conditions.
I think that is a valid reason to allow some safety factor.
That factor in my opinion should be determined by the extent and importance of the variables.
I do agree, do tests, but also consider what extra to allow.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
I think that is a valid reason to allow some safety factor.
That factor in my opinion should be determined by the extent and importance of the variables.
I do agree, do tests, but also consider what extra to allow.
Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &
for site rules
YvesLLewelyn
Mechanical
I don't disagree with any of the calculated values or the logic behind them.
However - the original problem was a practical one. Jsgarage asked if he needed to use the original length 2.7" (68mm) threaded length. If the original engine used a thread length of 68mm - I personally would continue to use 68mm - you could obtain suitable studs somewhere.
If I had an engine rebuilt by someone who told me that he used 40mm long studs instead of the original 68mm because the "calculations" said they would be OK - I don't think I would be very happy. The original maker of the engine may have discovered (for whatever reason) that they needed to use 68mm of thread length.
There would also be the awkward problem with a shorter stud that they couldn't be securely tightened against the bottom of their holes - meaning they would be a problem to tighten/untighten easily.
However - the original problem was a practical one. Jsgarage asked if he needed to use the original length 2.7" (68mm) threaded length. If the original engine used a thread length of 68mm - I personally would continue to use 68mm - you could obtain suitable studs somewhere.
If I had an engine rebuilt by someone who told me that he used 40mm long studs instead of the original 68mm because the "calculations" said they would be OK - I don't think I would be very happy. The original maker of the engine may have discovered (for whatever reason) that they needed to use 68mm of thread length.
There would also be the awkward problem with a shorter stud that they couldn't be securely tightened against the bottom of their holes - meaning they would be a problem to tighten/untighten easily.
- Status
Similar threads
- Locked
- Question