Part 3: Torque Wrenches
By Joseph C. Dille
BMWMOA #24754
It is important to have a reliable, accurate torque wrench to properly tighten fasteners to specification. A torque wrench is probably be one of the most expensive hand tools in your collection. In this, the final installment, I explain the differences between the two common types of torque wrenches and explain how to use them. I also share data I obtained by testing a bunch of wrenches.
There are two common types of torque wrenches for home shop use; the "beam" type and the "clicker" type. The beam type torque wrench is shown in Figure 1c and is the least expensive torque wrench. The beam wrench works by the beam bending in response to the torque applied as shown in Figure 2c. This type is very simple, reliable, and accurate, and there is little that can go wrong with it when used properly. When tightening a bolt, make sure to only apply force in the center of the handle. This allows the beam to bend in the manner it was designed to indicate the correct torque. Do not over torque the wrench or the beam may bend permanently. Do not drop the wrench because rough handling can bend the pointer arm or pointer. If the pointer is bent, it can be bent back to the center without affecting accuracy. If the beam is bent it cannot be bent back.
Figure 3c shows the clicker torque wrench, which is sometimes called a digital wrench. A clicker torque wrench works by preloading a "snap" mechanism with a spring to release at a specified torque. When the mechanism releases the ratchet head it makes a "click" noise as shown in Figure 4c. The torque is set by rotating the handle until the desired torque is shown in the window. Older clicker wrenches have a micrometer style scale along the handle instead of a window. The clicker wrench is much easier to use because it is easy to set the desired torque and just pull until you feel the click. The ratchet head also makes it easy to use in confined spaces. It is good practice to set a clicker wrench to its lowest setting before putting it away to prevent the spring from taking a set. Avoid rough handling and dropping because it can damage the mechanism. Do not use the torque wrench to loosen tight fasteners since this may damage the calibration.
I always wondered about the accuracy of torque wrenches, so I made my own torque wrench tester. The tester consisted of a lever arm that lifted a series of weights off the floor. The torque tester is shown in operation in Figure 5c. By changing the position of the weights on the lever arm and changing the weights, I could obtain torques from 3 to 105 ft-lb. The weights were barbell weights that I determined the exact weight using a digital shipping scale. I then calibrated the lever arm by using a precision electronic torque wrench and then back calculated the lengths using my known weights. A known torque exists when the arm is horizontal and the weights are lifted off the floor. I made a table of lever arm lengths and weight combinations so I could determine the applied torque in any situation. I estimate the accuracy of my home-built instrument to be +/-3% of the calculated torque.
Using the torque tester is straight forward. The pivot tube is placed in a vise and the desired weight is placed on the holder. The lifting chain/cable is then adjusted so the lever arm is parallel to the floor and located at the desired length per the torque table. The torque wrench is then inserted into the tester so it is approximately even with the lever arm. For beam wrenches, the wrench is rotated until the weights come off the floor and the indicated torque and applied torque are recorded. For clicker wrenches, the procedure is a little different. The wrench is set for torque slightly below the calculated applied torque. The wrench is then inserted in the tester and rotated slowly until it clicks. The wrench is set for the next higher torque increment and tried again. Successively higher torque settings are tried until the weights can be lifted from the floor without the wrench clicking. The highest setting where the wrench still clicked was recorded along with the applied torque. I learned that it was important to rotate the wrench slowly to avoid premature clicking caused by the inertia of the weights.
Once complete, I decided to use my new toy to check a bunch of torque wrenches to see how good they really were. I asked my friends in the Mac-Pac* to bring their torque wrenches to one of our wrench sessions at Bruce's garage. I was able to test a total of 13 wrenches, 3 beam, and 10 clickers. The results of my testing is shown in Table 1c and Figures 6c and 7c. I found the clickers to be much more repeatable than I expected. The beam type and clickers were both fairly accurate and linear over their range. The bottom line is torque wrenches, even inexpensive ones, were quite good.
| Owner | Size | Type | Range ft-lb |
Maximum Error | Average Error | ||
|---|---|---|---|---|---|---|---|
| ft-lb | % | ft-lb | % | ||||
| Joe | 3/8 | Clicker | 5-75 | -7.3 | -12.3% | -5.0 | -11.0% |
| Joe | 3/8 | Beam | 2-50 | 1.3 | 0.3% | 0.6 | 0.2% |
| Roger | 3/8 | Clicker | 2-21 | -1.6 | -13.9% | -0.8 | -8.6% |
| Bruce | 3/8 | Clicker | 2-21 | 1.0 | 5.3% | 0.8 | 1.7% |
| Mike D. | 3/8 | Clicker | 10-80 | 1.2 | 0.6% | 0.2 | 0.1% |
| Wayne | 3/8 | Clicker | 10-75 | -2.8 | -6.8% | -1.6 | -4.0% |
| Joe (new) | 1/2 | Clicker | 10-150 | -4.3 | -4.7% | -1.7 | -2.4% |
| Joe (old) | 1/2 | Clicker | 10-150 | 4.7 | -10.7% | 0.2 | -0.6% |
| John | 1/2 | Clicker | 10-150 | -3.3 | 20.0% | -1.0 | 1.1% |
| Mike D. | 1/2 | Beam | 5-150 | 7.8 | 28.1% | 4.9 | 12.7% |
| Ron | 1/2 | Beam | 5-150 | -4.3 | -6.8% | -2.5 | -4.3% |
| Bruce | 1/2 | Clicker | 25-250 | -1.8 | -3.4% | -0.8 | -1.3% |
| Mike B. | 1/2 | Clicker | 10-150 | 2.7 | 3.9% | 1.2 | 2.0% |
If you wonder about the accuracy of your wrenches, you can get them calibrated by a Snap-On tool dealer or a local metrology lab. Griot's Garage can also calibrate torque wrenches on a mail order basis for $25 plus shipping.
Torque Wrenches and Accessories
From time to time people have asked me if it is OK to use an extension with a torque wrench. The answer is yes. Using an extension or reducer with a torque wrench does not affect the accuracy. Others have asked if it is OK to use a universal joint with a torque wrench. The answer is NO. Universal joints change the torque as the drive angle increases. I checked this out with my torque tester. The results are shown in Figure 8c. Don't use universal joints with torque wrenches.
Torque extensions are sometimes required to tighten fasteners in locations where the torque wrench will not fit such as the drive shaft flange on older airheads. Figure 9c shows the extension I made for this purpose. To work correctly, one must understand how the position of the extension affects the torque as shown in Figure 10c. There is a formula for relating actual bolt torque to the wrench torque based on the length of the wrench and extension and the angle between the two. It is best to keep the two at right angles so the torque will be the same.
I wish to thank fellow BMW riders:
- Roger Albert
- Brian Curry
- Bob Spena
- Wayne Woodruf
for their help editing the article and providing encouragement. I hope this article has removed some of the mystery from the little things that keep our bikes together. I will close with a list of useful fastener-related references.
Fastener References
The following links are dead or otherwise unavailable since Joe wrote this article, but may be discoverable in the Wayback Machine.
Thanks for visiting!
Back to Part 2, "The Nuts and Bolts of Bolting"
