Internal Thread Stripping - Revisited.... 1

[MikeLiljenquist]

We have a stainless steel ring, that has 30 threaded through holes. Into these holes we place set screws, and the set screws are used to apply a load to an inner ring...used for clamping another machine part.

The ring is stainless, and the set screws are nickel plated. These screws are replaced about two times per week. For the past year we have seen a problem with the internal stainless threads stripping. The screws are 3/8" Diameter. We are applying a maximum of 160in-lbs of torque each time we install these.

So, my questions are:

1. Is there a limit to how many times you can use a threaded hole? Fatigue?
2. Could reducing the torque, by doubling the number of screws solve the problem?
3. Would a different plating or screw material help solve the problem?
4. Any tips on how to solve this problem would be appreciated.

Thanks,

Mike

 

[Metalguy]

Without knowing which SS you have, and how hard it is, and what amount of thread engagement you have, it makes it hard to give a meaninful reply.

So I'll guess-you have soft 304 or 316 SS, and enough threads (at least one full diameter). Your torque is low, so that leaves galling as the number 1, 2, 3, 4, and 5 top guesses!

Try a thread anti-seize compound, or use hardened heli-coil inserts or something like that.

 

[MikeLiljenquist]

I appreciate your help.

Material Type is 420 stainless, HRc38 and more than one full diameter of thread engagement. Does this change anything? Ni Anti-Seize was being used "once-in-a-while."

Thanks again,

Mike

 

[Metalguy]

Yes, it sure does change things a bunch! And now I don't have any good ideas about the cause. I'd make sure that the torque is really what you say it is.

I had on case where a front wheel came off a security guards truck on his way home. He though some union people were trying to kill him-was pretty shook up.

I was investigating, and found out he had just rotated the tires/wheels around, doing it himself. He even had looked up the wheel-nut torque. Swore he had used a torque wrench.

Everything pointed to insufficient torque-all of the other nuts were very loose. I asked him to bring the torque wrench to the shop where we were examining/testing the truck. He called his wife who immediately brought it over. Problem was immediately obvious-he used a small wrench calibrated in in/lbs.!!!

I tried hard to keep some of the sarcasm out of my report!

 

[TVP]

I am looking at some galling data from Carpenter in their excellent reference Carpenter Stainless Steels, and it appears that Type 420 is not much better than annealed Type 304 with respect to Threshold Galling Stress, which is obtained from Carpenter lab testing. You definitely need to use a lubricant with these fasteners, and may want to consider using a different fastener/coating combination. Since the torque is so low, you may want to consider using a more galling-resistant alloy such as Gall-Tough from Carpenter (I swear I am not a salesman for them). You can obtain more information on this alloy from their website:

http://www.cartech.com/

Click on Technical Information then Alloy Name and enter Gall-Tough

 

[strokersix]

How about a solid replaceable thread insert instead of coiled wire insert? I don't recall the proper term for these but you would have to drill out and tap a larger size in the ring to accomodate the inserts. They could potentially allow the use of longer set screws too.

Mike

 

[MikeLiljenquist]

Thanks again for the suggestions.

One more piece of information: The screws are used to load a ring that presses a piece of sheet metal. The sheet metal is stretched about 1300µm. It is the force of the tension screws that provides the stretch.

I checked on the use of heli-coil replacement threads. It seems that we are having trouble with them also. Maintenance believes the heli-coil failure is due to being used in through vs. blind holes.

About two years ago the number of set screws were reduced from 60 to 30. So, we installed one of the rings using 60 screws. The thought here was that by reducing the torque, we could extend the life of the threads. I was somewhat suprised that the final torque readings were only reduced by about 20% from 100in-lbs to 80 in-lbs. Does this make sense? It seems that it should be closer to a 50% in torque?

The internal threads have been used over 1300 times. With approximately the same load being applied each time. The screws are replaced every 5th installation. Over time, will threads just wear out? Especially considering WD40 or no lubricant was used for about 80% of the current thread life?

Thanks again for your help.

Mike

 

[TVP]

Yes, threads will wear out if no lubricant is used, and the materials are particularly susceptible to galling, and the number of insertions/re-insertions is high.

 

[diamondjim]

I think I would contact the loctite
people and see if you should be using
some special compound if you are centain
that they are not failing by gauling as
was suggested earler. You would always
use the same torque on the screws and
would increase your safety factor. Is the
problem how you tighten the screws, ie. do
you tighten 3 at 120 degrees and then
move one 180 from the first and then 120
degrees etc. If you progressively just
go around the ring, you are imposing different
forces on the shaft or an imbalance.

 

[MikeLiljenquist]

These are torqued in series. Each revolution of the ring we stretch the sheet metal about 100µm. So we are going around about 12-13 times. The starting screw has the highest torque value during the tensioning process.

I will contact Loctite on Monday. Thanks for the help again.

Do you think increasing the number of screws will help solve the problem, or is the torque already so low that reducing it further will not help much?

Thanks,

Mike

 

[diamondjim]

I really am not understanding what you
are doing. Are the setscrews in an
external collar locking an inner ring
as your first implied? If so, why are
you stretching an outer ring? Are you
trying this method to expand the outer
ring? Using the crisscross method of tightening,
you should only have to torque in 1/3
increments to reach the final torque value.
For example if your final torque was 36 ft-lbs
torque the first sequence to 12 ft-lbs,
the second to 24 ft-lbs, and then 36 ft-lbs
for the last time around again using a criss
cross pattern.

 

[MikeLiljenquist]

diamondjim,

These are set screws for tensioning an ID saw blade. I was being cautious, because I am not sure what the company rules are as far as talking about specific processes. But, at the same time we need to try and solve this problem.

The blades are tensioned by turning the set screws and measuring the runout of the blade. When the specified runout is reached, they proceed to the next screw in series.

We stretch the blade about 1250µm initially, and then after several cuts, retension in 50µm increments. This continues over the life of the blade. The blade is only about 120µm thick.

The blades are clamped in the main "ring", and inside the main ring, is a tension ring. The set screws go through the main ring, and contact the tension ring. The tension ring is the pressed into the blade, and the blade is "stretched."

I apologize for being vague before...

Does it make sense that measured torque values would only fall by 20% using 2X the number of screws?

Would reducing torque by 20in-lbs, contribute much to thread life extension?

 

[Metalguy]

"Would reducing torque by 20in-lbs, contribute much to thread life extension?"

Probably would help, even tho your orig. torque is low. But you are using set screws, so nearly all of the friction generated by that torque goes into the threads. With hex-head bolts, up to 90% of the friction is right under the bolt head-relatively little on the threads. You lose a little bit to the contact point, but the diameter is probably small.

Since your thread friction is actually far higher than it would first appear (at least to me), it is very important to *always* use a good lube on them-something WD40 surely is not.

 

[diamondjim]

First time around at 1/3 the torque value
Second time around at 2/3 the torque value
Third time around at full torque value.
Assuming you are using a calibrated torque wrench
and tightening in a crisscross manner.

4 times around maximum at 1/4 intervals might be ok
as well. Is the collar so thin than you are getting
false value?

 

[kalvin]

you need to carefull that you apply the loctite evenly around all the threads or you will have a condition of "side loading" thereby forcing the set screw to one side of the tapped hole and reducing the thread strength by half. You may want to look at something like Spiralock (

http://www.Spiralock.com

) which has more even thread load distribution and reduces fatigue.

 

[Grunchy]

I too think this is probably a fatigue problem. With galling surface metal plugs up the thread form and the screw eventually just freezes in the tapped hole. If you lubricate the fasteners you can virtually eliminate that problem. If the problem is stripping as you say, then I do not think galling is an issue.

I agree with kalvin, check out spiralock. Regular threadform fasteners concentrate all their stress on the first few engaged threads even if the threads are 1d long, and what you wind up with is a fatigue failure that progresses through the tapped hole a few threads at a time. Eventual failure is guaranteed. Spiralock-type threads spread the load much more evenly over the threads, I bet you would at least double the life of those tapped holes.

 

[diamondjim]

Sorry about the confusing multi posts.
It seems some of my posts go to limbo
and then appear a day or too later.
I think vibration is your biggest killer
and I would torque as high as possible so
the the load does not go from loaded to
unloaded. I wonder if double setscrews would
help. Are you certain that these do not loosen?
Without having a drawing and only relying on
your info. Maybe even stretching oftener would
ensure having load on the setscrews at all times.
Much of the other advice seems plausible as
well. The spiralock for one time fastening,
I like, but not certain how it works for multi
tightening. But certainly try all of the options
cause what you are doing now isn't working. I
assume you have several machines. What kind of
points are on the setscrews? Are you turning
the setscrews angularly or by torquing? Are you
replacing the setscrews after so many tightenings?
There are so many unknowns which make this interesting
and not simple. I am still trying to imagine if
each point in creating a temporary flat from screw
to screw to understand the difference in torque
when doubling the number of screws. Please keep
us posted what works. The fact you are stripping the
internal threads kind of implies that maybe you
should be using helicoils as the material seems
to be not hard enough to withstand the torque
forces. Is the blade getting hot and expanding
and loosening the torque values. The every two
weeks represents how many tightenings? Is the blade
acting as a borer or a debarking action? I am really
leaning toward the helicoil idea because of the cyclical
vibration. Have you used any of these before?
We use these on machine gun applications and seem
to hold up well in aluminum parts which are relatively
soft and see much vibration.

 

[MikeLiljenquist]

We currently use helicoils for thread repair/replacement. The saws we use cut silicon ingot.

The set screws are m10x1, not the standard 1.5. The set screws we use have a non tapered flat tip, so the diameter is the same as the minor diameter I believe...Nickel Plated, I inspected several of the new screws and there are visible surface imperfections, pits, etc. Maybe this is also a contributor to our problems. Heat seems to be very minimal, and very localized.

Two weeks represents about 12 retensions of the blade.

I went to the Emhart Teknologies website to look up the helicoil info, and there is a Screw-Lock Helicoil that looks promising.


Our application is for use in thru-holes. Our maintenance department believes that this is why the helicoils do not appear to be effective. Currently, helicoils are only lasting approximately 2 tensionings.

Are there any good text/reference books on this subject?

As of today, we are using a small amount of anti-seize on all of the tension screws. Threaded holes are being cleaned with a nylon brush during blade changes.

Your help is greatly appreciated,

Mike

 

[sbra]

Hi There,
If as you have said you are getting Galling or
Thread Stripping then you should LOOK at the thread itself.
First is the thread in the Stainless Steel made by
Thread Cutting or Thread Forming.

Thread Cutting is making thread using a TAP.

Thread Forming is when thread is Formed using Form Tap.

The Thread made by the Forming method is superior in
Strength--- Surface Finish in comparison to cut thread
threads.


Also the Screws you use Should be looked at.
Have these made by Forming Method and if made from
a Material that can be Hardened and Tempered this will
if done properly will be VASTLY SUPERIOR to any standard
bolt no matter what Coating it Has which unless you like
Colours most coating (except for corrosive preventative)
has LITTLE ADVANTAGE when the BASE METAL UNDERNEATH IS
NOT SUITABLE for the job it is required.
Probably chat to some TOOL & DIE SHOPS and this is
standard practice for them as alot of this work is done
by these overhere in UK.
Using a lubricant is okay bur Basically your Threads must
be a good finish (flawless...no pitting... complete form...
not too coarse --for M10 a 1mm pitch is good as you stated)
and if both are approx 40--45 HRC in Hardness and must be
STRESS RELIEVED before Hardened (if this is done) you should get good result's.
Regard's
SBRA.

 

[diamondjim]

Another thought if the heli-coils are
not working is keenserts which are
solid inserts and the ods are much larger
than helicoils which may be a plus.