TORQUE WRENCH TESTING?
- Thread starter Jeffs88
- Start date
-
Sponsors (?)
ID89GT
10 Year Member
You need 4 things to check the accuracy of a torque wrench.
1.) A vise to clamp the square drive end of the torque wrench so that it is stable and isn't moving all over the place.
2.) A suitable weight that is within 2% of the weight marked on it.
3.) A ruler or tape measure.
4.) A coat hanger or wire strong enough to hold the weight.
Mount the square drive of the torque wrench in the vise and tighten it down securely. Measure from the center point of the square drive to the line in the middle of the torque wrench handle or handle pivot. Multiply the measured distance by the known value of the weight.
Example:
14 inches x 50 lbs = 700 inch pounds.
Divide by 12 to get foot pounds: 700/12 = 58.33 foot lbs.
If the torque wrench has a setting dial, set it to the value you just calculated. Attach the wire or coat hanger that has the weight to the pivot point, or line on the handle. Listen for the click or beep, or watch for the flash or the indication on the dial. It should match the calculated value within 2%.
That’s the simple part. Unfortunately I have no idea how to adjust anything but the bending beam torque wrenches, so you are on your own from this point on.
1.) A vise to clamp the square drive end of the torque wrench so that it is stable and isn't moving all over the place.
2.) A suitable weight that is within 2% of the weight marked on it.
3.) A ruler or tape measure.
4.) A coat hanger or wire strong enough to hold the weight.
Mount the square drive of the torque wrench in the vise and tighten it down securely. Measure from the center point of the square drive to the line in the middle of the torque wrench handle or handle pivot. Multiply the measured distance by the known value of the weight.
Example:
14 inches x 50 lbs = 700 inch pounds.
Divide by 12 to get foot pounds: 700/12 = 58.33 foot lbs.
If the torque wrench has a setting dial, set it to the value you just calculated. Attach the wire or coat hanger that has the weight to the pivot point, or line on the handle. Listen for the click or beep, or watch for the flash or the indication on the dial. It should match the calculated value within 2%.
That’s the simple part. Unfortunately I have no idea how to adjust anything but the bending beam torque wrenches, so you are on your own from this point on.
That's interesting...I'll try it. I guess it will click too early if it's out of calibration in one direction or not click at all if off the other direction....hmmmm.
As usual, you come thru with a reasonable answer.....Thank you.
Jeff
What brand is it? I believe you can just walk into a Sears and hand them your Craftsman wrench and they will get it calibrated for you. Not sure how much it costs.
jrichker's post makes perfect engineering sense, but I see a couple issues:
1. The weight of the wrench itself is ignored
2. It only measures the calibration of the wrench at one specific value. Torque wrenches need to be calibrated for accuracy all over their operating range.
jrichker's post makes perfect engineering sense, but I see a couple issues:
1. The weight of the wrench itself is ignored
2. It only measures the calibration of the wrench at one specific value. Torque wrenches need to be calibrated for accuracy all over their operating range.
I't a Husky...not an incredibly expensive piece. I've used it mostly over the years torqing lug nuts and such. I just rebuilt my T-5 a month or two ago and used it for that as well. Over this winter I'm going to be swaping on a set of GT40P's and explorer intake as well as rebuilding my Diff. so I thought it would be a good time to make sure it's gonna give me the correct values BEFORE I get started.....
Who sells Husky? Home Depot? Maybe drop into a Home Depot and talk to somebody in the tool department and see what you can do about getting it calibrated.
You're splitting hairs...
If the torque wrench weighs 3 lbs, and its maximum load is 100 ft/lbs. the 3 lbs only represents 3% of the load. That plus the 2% tolerance for the weight puts the test accuracy at +/- 5%, which is very good. If you wish to be more precise, add the weight of the wrench to the the test weight. It won't be exact, but it will reduce the overall error percentage.
The Husky/ Home Depot Model # 39104T, Internet # 100058237, Store SKU # 749342, click style wrench is 250 ft/lb max load and weighs 6.32 lbs.
6.32/250 = 0.02528% + 2% for the weight = +/- 4.25% accuracy.
Using different weights is just as feasible as a 50 lb weight. The only limitation is how many weights you have that only have a 2% variance from the weight marked on them.
You could just buy a beam style torque wrench, tighten a bolt in something to a specific torque then use your click style to confirm and then return the beam style once your testing is complete. If the bolt moves a little (couple degrees) and clicks, then it should be fine. If it instantly clicks and doesnt move keep increasing the torque setting on the wrench till it moves a couple of degrees and clicks. If this increase in the setting is to far off for you, then you have your answer.
He's right. The beam style torque wrenches only require you to bend the pointer so that it reads zero when at rest. Other than that, they never go out of calibration.
That's Probably the best way to test it then. Is it true that you should never use a torque wrench to loosen a bolt. only to tighten?
ghostang93
Member
That's the way the AF teaches. Of course they also want the torque wrenches calibrated every 3-6 months. The AF also teaches that you should never use a torque wrench at their lower and upper values due to calibration. I want to say the lower/upper 30% but I just can't remember. Most of the guidelines just aren't practical for home use though.
You can, but the idea is to never continue to apply force after the "click". The torque wrench is not designed to take additional loading beyond that point. If it were true that you couldn't use a torque wrench to loosen, then you also shouldn't tighten left hand threads.
In other words, the wrench is designed to work in both directions, just not past the "click".BTW:
http://www.tegger.com/hondafaq/torque_wrench/index.html
Boydster
Active Member
We have a torque wrench at work that we use all the time. It's a 1" drive, max 800 ft/lbs, about 6' long. We break loose the same nuts that we torque with it. Nothing wrong with it and it won't hurt the calibration. Ours get calibration certification annually and believe me, if wrenches were constantly coming back out of whack, they'd be letting us know.
However, after spending a lifetime with torque wrenches, I agree with Nik. I wouldn't use it past the click.
stang&2Birds
Founding Member
This is how you test a torque wrench: 
http://home.jtan.com/~joe/KIAT/kiat_3.htm
View attachment 161412
Note, read the article.
http://home.jtan.com/~joe/KIAT/kiat_3.htm
View attachment 161412
Note, read the article.
stang&2Birds
Founding Member
Yup. That's because in real life, springs are not 100% linear.
It's easy to see way at the high-end of the range. As the spring compresses more and more, the coils start to touch. In real life, some coils will start to touche before others. When the coils start to touch, then you loose the "spring action".
Also, there are a number of metallurgical reasons why springs are not linear at full extension or near compression. Many first semesters Mech Engir metallurgy books cover the basic aspects. It gets much more involved depending on how much accuracy, repeatability, and usage life is required.
stang&2Birds
Founding Member
Here is something that I quickly found:
From: http://lpsa.swarthmore.edu/Systems/MechTranslating/TransMechSysElem.html
Aside: Behavior of springs in the real world.
In the real world, springs are not completely linear. For example the spring shown at the top of this section might have a force-displacement curve that looks more like the one below.
View attachment 161410
This graph has two distinct non-linearities. There is a slight curvature to the graph that indicates that it is increasingly difficult to stretch (or compress) the spring as you move away from the relaxed (no applied force) state of the spring. There is also a hard non-linearity as the spring is compressed enough that the coils actually touch and prevent further compression.
Even though the spring is non-linear, a linear analysis can still be useful. Note that over a small interval (say around 4 cm of displacement) that the curve can be approximated by a straight line (i.e., a linear approximation).
From: http://lpsa.swarthmore.edu/Systems/MechTranslating/TransMechSysElem.html
Aside: Behavior of springs in the real world.
In the real world, springs are not completely linear. For example the spring shown at the top of this section might have a force-displacement curve that looks more like the one below.
View attachment 161410
This graph has two distinct non-linearities. There is a slight curvature to the graph that indicates that it is increasingly difficult to stretch (or compress) the spring as you move away from the relaxed (no applied force) state of the spring. There is also a hard non-linearity as the spring is compressed enough that the coils actually touch and prevent further compression.
Even though the spring is non-linear, a linear analysis can still be useful. Note that over a small interval (say around 4 cm of displacement) that the curve can be approximated by a straight line (i.e., a linear approximation).
jetmech807
10 Year Member
I believe most click wrenches have lead in the bottom of the handles. We melted the lead out of one once and there was a nut inside. I would assume that would be how to adjust it. I wouldn't get into all that myself.
Sent from my HTC EVO 4G
Sent from my HTC EVO 4G
