Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

Bolts Clamp Load: no threads per inch effect?!

687 views
Skip to first unread message

Ralf B. Lukner

unread,
May 10, 1997, 3:00:00 AM5/10/97
to

I have a question about the relationship between tightening torque for
bolts and the clamp load:

-----------------------------------------------------------------------------
Why doesn't the number of threads per inch affect the observed bolt clamp
load for a given torque? Intuitively, a bolt with a large number of
threads per inch will have a higher clamp force for a given (wrench)
torque than a bolt with a small number of threads per inch and the same
imposed torque.
-----------------------------------------------------------------------------

Example formula:

Preload caused by torquing is given by (e.g., Parmley, Standard Handbook
of Fastening and Joining, 1997):

T = K D P,

where T = installation torque, lb-in
K = torque coeficient (e.g., .15 or .20)
D = nominal bolt diameter, inches
P = clamp load objective, lb

Thus, according to this example formula, which agrees with the results
from "extensive work": for a given reading on the torque wrench, the
estimated clamp load would always be the same if only the number of
threads per inch is changed. Any insight into why the number of threads
per inch does not need to be considered would be appreciated.

--Ralf

Robert B. Price

unread,
May 10, 1997, 3:00:00 AM5/10/97
to

Ralf B. Lukner wrote:
>
> I have a question about the relationship between tightening torque for
> bolts and the clamp load:

Big snip

See the article in the 8 Ma7 97 issue of Machine Design mag.

--
| Robert B. Price, C.Mfg.E. |
| Automation*Gears*Machinery | "Sapere aude"
| Consulting Engineering |

Dan Timberlake

unread,
May 11, 1997, 3:00:00 AM5/11/97
to

Are you thinking of the threads acting as springs in parallel?

A screw thread is a single inclined plane. More threads (taller nut)
reduce the contact stress somewhat, but it's still the first few threads
that carry the load.
Although threads act like inclined planes, so the normal force resolves
into clamping and bolt-twisting forces, the clamping force results from
stretching the bolt. It takes a certain amount of force to stretch 10 mm
steel bolt 50 mm long an extra 0.152 mm. Whatever you do to stretch the
bolt this amount will require this force. A common method in FEA of
simulating a torqued (stretched) bolt is to specify a low temperature for
all the bolt elements. Similarly, steel construction handbooks specify
two ways to torque bridge bolts. Either many foot pounds at
ambient temperature, or fewer foot pounds if you heat the bolt up with a
torch before tightening. When it cools, it's tight.

Martin Green

unread,
May 12, 1997, 3:00:00 AM5/12/97
to

I think Ralf must be right about the number of threads. Certainly, with
more TPI, you would need less wrench torque to acheive the same bolt
tension. The reason it doesn't appear in the formula T=KDP must be because
the effect is buried in the "torque coefficient", K. Engineering formulas
are like that.

Marty Green

Ralf B. Lukner <luk...@che.utexas.edu> wrote in article
<lukner-1005...@bingham.che.utexas.edu>...


> I have a question about the relationship between tightening torque for
> bolts and the clamp load:
>
>

----------------------------------------------------------------------------

dcon...@siohio.com

unread,
May 12, 1997, 3:00:00 AM5/12/97
to

Hello,

I think you guys are confusing wrench torque with wrench motion.

With more TPI you need more nut rotation to achieve the same bolt
tension (i.e., if you achieve a certain torque by applying 90 degrees
of bolt rotation (past hand-tight) with a 10 TPI bolt, you would
expect that for a 20 TPI bolt it would take about 180 degrees of bolt
rotation past hand tight to achieve the same torque) This is because
the amount of rotation is inversely proportional to the TPI for a
given applied bolt axial displacement.

However, given that you apply the correct amount of rotation for your
TPI to achieve a specified torque, the bolt tension is independent of
both the TPI and the amount of rotation that was required to achieve
it.

The torque formulae don't have a TPI factor because TPI only affects
the amount of rotation required to achieve a given torque. How much
of the torque is transferred into bolt elongation and how much is
wasted thru friction is independent of TPI.

Have a nice day!

Dennis

On 12 May 1997 14:13:40 GMT, "Martin Green" <test...@pangea.ca>
wrote:

Israel Kehaty

unread,
May 12, 1997, 3:00:00 AM5/12/97
to

In article <lukner-1005...@bingham.che.utexas.edu>,
luk...@che.utexas.edu says...

This formula is correct as long as the number of threads is large enough so
that the strength of the threads againt shear (and bending) of the thread is
larger that the tensile strength of the bolt shaft.

The formula above calculates the "P - clamp load" based on the breaking
strength of the bolt in tension.

For example if you use a steel bolt in an aluminum thread the thread
length is detemined by the aluminum because the thread in the aluminum is
weaker that the thread of the steel bolt.

If you use too much threads than the friction torque between the bolt and
the hole may be larger due to accumulated errors in pitch and maybe
straightness of the bolt or the the threaded hole.

Martin Green

unread,
May 14, 1997, 3:00:00 AM5/14/97
to

I replied to this posting last week, and now I think that the answer I gave
was wrong.
If there were no friction in the threads, the bolt tension would certainly
reflect back into the torque wrench. But then there wouldn't be much point
in tightening down the bolt, because it would just unwind itself when you
let go of the wrench.

How hard do you have to push to get a heavy box to slide up an inclined
plane? It turns out the answer doesn't depend (well...hardly at all) on the
slope of the incline, unless the tangent of the slope is large compared to
the co-efficient of friction. I guess it's the same with the torque on a
bolt.

Martin Green

Ralf B. Lukner <luk...@che.utexas.edu> wrote in article
<lukner-1005...@bingham.che.utexas.edu>...

Hugh Hunkeler

unread,
May 16, 1997, 3:00:00 AM5/16/97
to

Ralf B. Lukner wrote:
> I have a question about the relationship between tightening torque for
> bolts and the clamp load:
> Why doesn't the number of threads per inch affect the observed bolt clamp
> load for a given torque? Intuitively, a bolt with a large number of
> threads per inch will have a higher clamp force for a given (wrench)
> torque than a bolt with a small number of threads per inch and the same
> imposed torque.

The bolt with more threads per inch has more threads contributing to the
axial load, BUT the bolt with less threads per inch has a greater pitch
angle to convert torque (in the x-y plane) to axial load (in the z
plane). So, they cancel out in the torque-vs.-axial load equation. The
threads per inch will have an effect on the torque at which the threads
strip, and certainly on the number of turns that the bolt must undergo
to become tight.

***Remove ".nospam" from address***

Erkki Niemi

unread,
May 18, 1997, 3:00:00 AM5/18/97
to

Ralf B. Lukner wrote:
>
> I have a question about the relationship between tightening torque for
> bolts and the clamp load:
>
> -----------------------------------------------------------------------------

> Why doesn't the number of threads per inch affect the observed bolt clamp
> load for a given torque? Intuitively, a bolt with a large number of
> threads per inch will have a higher clamp force for a given (wrench)
> torque than a bolt with a small number of threads per inch and the same
> imposed torque.
> -----------------------------------------------------------------------------

>
> Example formula:
>
> Preload caused by torquing is given by (e.g., Parmley, Standard Handbook
> of Fastening and Joining, 1997):
>
> T = K D P,
>
> where T = installation torque, lb-in
> K = torque coeficient (e.g., .15 or .20)
> D = nominal bolt diameter, inches
> P = clamp load objective, lb
>
> Thus, according to this example formula, which agrees with the results
> from "extensive work": for a given reading on the torque wrench, the
> estimated clamp load would always be the same if only the number of
> threads per inch is changed. Any insight into why the number of threads
> per inch does not need to be considered would be appreciated.
>
> --Ralf

Hello!

Because friction takes 85 % of the total torque, the difference in the
remaining 15 % does not matter much.

Regards, Erkki

0 new messages