Calibrating torque wrenches, was dist. rattle

auditude at auditude at
Wed Jan 28 12:44:08 EST 2004

Bernie Benz <b.benz at> wrote:
> > From: <auditude at>
> > 
> > Okay, those are nice technical terms.  But I guess I didn't see where you
> > answered my question.  What do you Do Yourself to fix a torque wrench that
> > doesn't work, just replace it?
> That's about all that you can do with the clicker type.  The beam type
> always "works".  Just apply your correction factor.

Or have it recalibrated like I did would be one other option, instead of replacement.  

I'm surprised your not applying your DFI IIAB (did I spell that right?) rule of thumb here.  Assuming I like the convenience of the click type, I only fixed what was broke, maybe a spring or adjustment or something inside my inaccurate torque wrench, while preserving all the other parts that were still good.

On the beam type, yeah, those laws of nature are pretty reliable.  I guess inside a click type torque wrench there are also laws of nature at work, but with more parts to fail or wear, and proper unloaded storage also a factor over the simpler beam/pointer style.  Like the expanding pellet thing inside the 3 prong MFTS.  It seems (is?) more reliable than some integrated circuit.  The KISS principle at work. =)
> >Do you apply a correction factor to all its measurements?
> Sure, as necessary.
> > 
> >> If one does not know what he has paid for, it most probably was not worth
> >> the price.
> > 
> > Since my torque wrench now works better than it did when I sent it in, what
> > does that last statement have to do with this thread?
> What does "works better" mean?

Well, like you wrote above, the pointer type always "works", so by "works better" I probably mean the same as you, only more of it.  As in more accurately, easier, and maybe even more precisely (Depending on what was wrong inside of it.  The report that came back from Washington Calibration didn't mention how many samples they took and the resultant measurements, so I don't know if the measurement error was consistent and repeatable).

In the case of my torque wrench, I don't have the numbers with me, but let's say for example the measurement was off by -10 ftlbs at 50ftlbs applied and -25ftlbs at 100ftlbs applied or something like that.  When I got it back, it was more like -0.5ftlbs off at 50ftlbs applied and +1.0ftlbs at 100ftlbs applied.  They changed something inside it to make it more accurate, and maybe more precise too since things are supposedly brought back into specification.

Since the measured torque is now closer to the actual torque than it was before the service, to me it works better.  I'm not sure how else I can explain that phrase to make it work better. =)

> > There is a such a thing as "core competency" you know, where you outsource
> > things that can better be done efficiently and effectively by someone else.
> > Nowhere in this concept is there a requirement to fully understand the methods
> > used.  The end result is what's relevant.
> > 
> > My not knowing how they made it better is only a problem when I try to DIY the
> > calibration of my torque wrench.  I was waiting for you to explain how you Do
> > That Yourself, and I guess I still am.  If it's not possible or practical,
> > just say so.
> One method, I put the 1/2" drive end in the bench vice, stand on the
> bathroom scale and push up or down on the wrench handle, the diff in
> indicated weight being the applied force.  Knowing the length of the handle,
> you know the applied torque.

Okay, so you are talking about verifying that the tool is in calibration.  I was calling that calibration checking.  I thought you were offering a DIY method of calibrating or recalibrating.  By that I mean taking some action that would bring the torque wrench back into an acceptable range of accuracy from not.

I suppose if you include yourself in the measurement system, then even with your calibration checking, as long as you apply your correction factor, you are in effect recalibrating your measurement system.  By having the torque wrench serviced, I was recalibrating a tool to be used in my measurement system, so that I wouldn't have to apply any correction factors (which would be a challenge with a click type wrench).

Using the video camera example again, if it wasn't calibrated and thereby made to work better, then one might have to wear tinted glasses when viewing all the recorded  movies to get a feel for what it looked like in real life.

I guess what counts is if the operator is able to apply the correct torque to a fastener.  Also, if money is being spent on calibration services that there is still enough left over for food, shelter, entertainment, emergencies, and retirement.  And/or that the spender feels like they got fair value for their money.  I have more peace of mind knowing that my wrench is more accurate than it was, even if it is not as accurate as a beam type.  

If I end up assembling some connecting rods or something like that, I'll get a beam type torque wrench, or even better, do it by measuring bolt stretch.  Most likely I'd have it done, altho' I'd like to learn (not necessarily on my motor!)



More information about the 200q20v mailing list