Checking torque wrench accuracy: Methods?

My new block and rotating assembly is supposed to be from the machinist this week, and I'll be putting it together soon, and I was concerned about my torque wrench accuracy. I may entertain buying a new one, but I'd like to make sure that's even necessary. I currently have a Craftsman click type wrench, along with a couple bending beam type wrenches. I'm not sure how to check these out for accuracy. Suggestions?
I was entertaining making a (balanced) fixture with a pivot and a (weighted) moment arm to produce a ft.-lb load, but I don't have an accurate way to weigh out the mass that I would attach to the moment arm. The pivot point would have a hex shape to allow putting a socket on it, along with the torque wrench. Pulling on the torque wrench would lift the moment arm weight off its resting point. Too much work?

Like I said, just looking for suggestions here.

Thanks,
Mike

Put a couple sockets face to face on a piece of hex stock and use one of the beam wrenches to torque the clicker until it clicks.

For what it's worth, I've read repeatedly that good engine builders prefer a good beam wrench to a clicker.

I was trying to torque some heads with the engine in the car and I could not easily read the scale on my old beam wrench and apply the torque to the wrench at the same time. So, I bought a micrometer style torque wrench and it, like most tools, was made in China. When I got it out of the box I coupled the two wrenches together and checked the torque of the new wrench. It was dead on. I torqued the heads and re checked the settings, still dead on.

I have re-checked it a few times (same method) in the last few years and it is still accurate.

The beam style is hard to beat as a calibration device. I prefer to use them over the click style but there are situations where they are useful.

Mocked up a cheap setup that ought to work out good enough. I'm using some bearings that I had laying around,and some half-inch drive sockets that fit the bearing I.D. Just need to get some hex stock tomorrow to do the actual coupling between the two sockets, and cut/drill a baseplate to tie the bearing carriers together. I don't know what the strength/yield point is of hex stock material, but it ought to be easy to find out if I see the beam torque reading plateau.

That is one Hi-Tech setup!! I just used a 1/2 drive 8 point 1/2 socket. Maybe I need to rethink my calibration technique..

TimAT said:

That is one Hi-Tech setup!! I just used a 1/2 drive 8 point 1/2 socket. Maybe I need to rethink my calibration technique..

Click to expand...

I just wanted to be able to read the beam pointer accurately without stuff moving around or slipping out of my hands. This ought to do it, and it won't cost me much.

Is there going to be unmeasured force going into the bench through the bearing ?

DWncchs said:

Is there going to be unmeasured force going into the bench through the bearing ?

Click to expand...

There shouldn't be. I brushed off the mental cobwebs (as best I could) and did the usual freebody diagram from Statics 101 class. The bench and the bearings do experience some forces, as I'm tugging hard on the beam torque wrench, but nothing in the setup moves. Because nothing is moving, the forces must inherently be balanced (equal, but opposite) to satisfy the laws of physics.
And, as you can see, in light of the fact that there is no rotational motion despite the force applied by me, the torque sensed by both wrenches must be balanced also (equal and opposite). With equal torque inputs, the wrenches should then read equal numbers if they are both calibrated correctly (or, I suppose, if they're both equally off calibration LOL).

Pretty dry description I admit, but unfortunately that's how my mind is wired. :amused:

Well, I got the two bearings bolted to a piece of plate steel, and found a suitably strong hex shaft to connect the two sockets, and did some testing today. I went in 20 ft-lb increments (or 20 lbs-ft for the anal types) up to 90 pounds, and both wrenches matched. I set the clicker torque wrench to a particular number for each test, and then pulled on the beam wrench, and took note of what the reading was when the clicker wrench snapped. Very good correlation between the two. I'm rather impressed that the Craftsman torque wrench has held steady all these years, given how much use it's had.