DIY "Click" Torque Wrench testing.
 

A guy over on a Sprite bulletin board I frequent suggests this method as a way to check the accuracy of a "click" type torque wrench. Sounds like a valid method assuming the torque wrench is set to be accurate at 12 inches from the center of the drive. What do you guys think? Here is what he wrote:

"Click type wrenches are accurate enough for routine engine builds - I check mine weekly by securing the drive square in a vice and with the wrench horizontal, hang a 28lb scale weight on the wrench handle and slide it out till the wrench clicks - then measure the distance of the weight from the centre of the drive square - simple maths then calculate the correct setting - eg if set to 28 ft lbs, the weight should be exactly 12 inches along the handle when the wrench clicks."

Makes sense to me. I'm sure an engineer will point out that it does not take into account the weight of the handle past 12 inches, but it would be close enough for me. I like it.

Re: DIY "Click" Torque Wrench testing.
 

the torque wrench should be accurate enough at any distance the force is applied, assuming the arm is rigid.

the actual torque is very simply calculated... force x moment arm (distance from center of drive).

for example, if by this method the wrench clicks at 13 inches instead of the expected 12, then the wrench is actually applying 28lb x 13in = 364in-lb = 30.3ft-lb.

yes, the weight of the torque wrench could be considered negligible.

Could orient the wrench and lever arm in the horizontal plane with a level and use a cable and pulley to transfer the weight at 90 degrees. You'd have friction in the cable though.

Assuming the wrench was off, say reading 32lbs when it should be reading 28lbs is there any way to calibrate the wrench yourself?

Mine (Crafstman) came apart and I recalibrated it this way. I think there was a thread about Crafstman torque wrenches that came apart. It's just a nut down inside the handle, but it is a biotch to get at.

calibration
 

normally there is some sort of nut or screw used to change the preload on a spring. once you figure out the mechanism used on your particular wrench, then the calibration method is as you would expect: simply adjust until the click happens at 12 inches.

I'm not near a Torque wrench right now but is 12 inches at the center of the knurled portion of the handle?

not necessarily.

I've got a clicker type and a bar type. The bar-types rarely go out of spec.

So, couldn't you join the two, say with a 10mm socket in one and a 10mm allen in the other, and then put opposing force on them watching the bar-type one for when the other clicks?

Just a thought. Never tried it.

i can't see why not!

i've always found the bar type to be a little difficult to use. just curious: how often do you find it useful? what is the torque range?

My bar type goes up to about 60 ft/lbs. My clicker goes up to 150 ft/lbs. The bar type is good for smaller bolts with lower torque settings. The clicker is good for bigger things.

I've never fully understood how to use the bar type. I just apply pressure up to the displayed torque value and wait for the nut/bolt to stop turning. Seems accurate enough, but not idiot proof like the clicker.

the bar type is very simple to use in theory and you are likely using it correctly. as you say, it isn't idiot proof, so keep it away from me.

btw, it will stay calibrated indefinitely unless the indicator gets bent or the material properties of the lever arm somehow change (by exposure to very high temperature for example).

The correct method to read a beam-type torque wrench is to read the scale as the fastener rotates; the torque to overcome initial friction resistance is higher than when the fastener rotates.

Sherwood

rolling torque