Torque Wrench Calibration

[RRoller123]

Working through a leakage problem I have uncovered, and Tim A sent me this useful paper on calibration of the torque wrench. I used to have the QA Lab do this when I was working, but now retired, so doing this today here in Frankenstein's Laboratory. Have to be absolutely sure of this most critical of variables in solving these problems.

You folks may find this useful. The only comments that I have so far is that I have a woodworker's bench vise with wooden contact plates, that distort when the wrench is put in the vise, so I have to remove the plates to get to bare metal. That will hold it fine. And keep it as perfectly horizontal as possible.

https://helicoptermaintenancemagazine.c ... tIzMVIyBUU

Pete

[18Fiatsandcounting]

Good suggestion Pete, and in fact it was a good reminder for me to go check my calibrations (they were pretty close). Rather than a weight, I use a bathroom scale and place the gripping part of the torque wrench on the scale and pull the scale upwards until the wrench clicks, at which point I read off the weight on the scale. From there, you have to do some math to account for the distance from the ratchet to the point where the handle rests on the scale (unless it's exactly 1 foot), but it's straightforward.

This method assumes your scale is accurate, but I would imagine a modern digital scale is within a few percent which is likely just fine for automotive torque wrenching. The advantage is that you can check the torque wrench all across its range.

-Bryan

[RRoller123]

I just finished this calibration effort, and the results were surprising, but in a good way. I bought this Craftsman Torque wrench in about 1980, maybe a little earlier!!!

For my setup, I am extremely focused on the critical Head Bolt torque, so the numbers run as follows:

Run the wrench setting up to the desired setting from a lower setting, and do this for each test. Always leave the Wrench set at the loosest (lowest) setting, to avoid spring fatigue.

38.325 pounds of total weight, hung by a string. Measured with an accurate Postal Scale.
19.20" from center of rotation. Out at the end of the handle.
Wrench held horizontal in vise, with a supporting dowel to keep it from rotating.

This should yield 61.32 Foot-Pounds of Torque.

My wrench, with this setup, indicated ["clicked"] at a setting of 63 Foot Pounds. Not too bad, only off by 2.6%.
So my Wrench reads a little too high, the actual torque is slightly lower.

BUT with lower readings, the wrench was off by a little more.
41.89 Foot Pounds actual torque, required a setting of 45 Foot Pounds on the wrench, and error of 6.9%.

Picture of the setup:

[18Fiatsandcounting]

Looks like I have the same Craftsman torque wrench as you, and mine has stayed pretty accurate all these years. At least to my satisfaction, which I consider 5% at the higher torque readings. I also turn it back to the loosest setting immediately after use, especially on the higher torque settings. I don't recall ever "maxxing" it out, but I have gone up to 130 ft lbs. Not much need to go higher on a Fiat, except for the infamous crankshaft nut on certain model years, at 250 ft lbs or so. At that point, I just tighten that %*&#@ as tight as I can get it, and call it a day.

-Bryan

[RRoller123]

It has been very good for over 40 years for me, too. Upon closer inspection I realize that the thing actually can't be calibrated, at least I can't see any way to get at the innards, so I will just use it with a slightly higher indicated reading. Maybe a couple of ft-lbs at 50 and above, and maybe 3 or 4 at around 35 and below, that should do it for now.

The dowel had no negative impact , the head freely slipped along it, and it just kept it from tipping and creating an angular
error.

[18Fiatsandcounting]

Now if you had 2 torque wrenches and wanted to test them against each other, plug one of them into a 1/2" drive that has a socket that fits on the other torque wrench's 1/2" drive, put one torque wrench on tighten and the other on loosen (I think), and they both should click at the same time working against each other. If they are both calibrated, of course.

I suppose you could get creative and do this with all manner of combinations of 1/2", 3/8", and 1/4" drives.

-Bryan

[DieselSpider]

I guess I should be thankful that I still use an old fashioned beam and pointer style torque wrench. If the pointer is rests at 0 then its calibrated. I use that one to test the clickers which can be all over the place even when new. Sometimes simple is better.

[RRoller123]

Those still require calibration just as any measuring instrument does. Nothing is made permanently accurate.

[DieselSpider]

Those still require calibration just as any measuring instrument does. Nothing is made permanently accurate.

Calibration usually involves zeroing the pointer on those. My 1/2" drive non-ratcheting vernier scale model has passed calibration tests now for close to 50 years.

[RRoller123]

Excellent!

[DieselSpider]

Excellent!

I look at it as: Simple many times is really the best.