Torque Wrench

[jackellis]

I have probably spent too much time on this little information-gathering task but it's certainly been educational. I just bought a "new" (aka closeout?) CDI 10-75 ft-lb wrench for $70 shipped (reputable Ebay seller). Apparently they're a pretty good brand (owned by Snap-On) and the calibrated range is suitable.

This is going to be an occasional use tool - mostly for emergencies since I do not typically do my own maintenance - so for that reason it has to stay in the airplane. I did enough reading this morning to feel pretty comfortable that just about any brand is likely to work well and give good service if it isn't dropped or abused. My research also suggests it's more important to get a tool that has the right range of calibrated values and can be adjusted finely enough than it is to buy a premium brand tool. There were some 20-100 ft-lb Snap-ons listed on Ebay that could only be adjusted to the nearest 10 ft-lbs for about the same price I paid for the CDI.

Next project - build a calibration rig with some weights.

[Daryl]

A few years ago I was installing the manifold on my car, old mechanic neighbor happened by and was supervising me. I got the torque wrench out and was about to tighten the bolts when he said "Forget about the torque wrench, about two grunts tight ought to do it for those".

[Tom Slick]

CDI is what I meant, not PDI. We used them and Snap-on in the USAF, they are good wrenches, especially at the price you paid.

[woodturner]

I just bought a "new" (aka closeout?) CDI 10-75 ft-lb wrench for $70 shipped (reputable Ebay seller). Apparently they're a pretty good brand (owned by Snap-On) and the calibrated range is suitable.

As another poster mentioned previously, it's important to store click type torque wrenches properly. Often, that means storing at the zero ft-lbs setting and never turning the knob below zero (because that changes the settings). However, I would suggest reading the manual for the particular wrench you have and following the manufacturer's directions carefully.

For infrequent use, it shouldn't be necessary to recalibrate the wrench very often. It's relatively expensive to get a wrench calibrated, but you can do a reasonable approximation with the weight method.

[unknown poster]

One additional thing to keep in mind when using a click type wrench is that they are usually only accurate in the upper part of their range. My Craftsman wrenches are spec'd to be within a certain range of accuracy, but only in the upper 80% of their range. That means if you have a wrench that measures from 20-100 ft lbs, it is only accurate from 36-100 ft lbs.

[Mr__Bill]

A down and dirty way to test the accuracy of your torque wrench at a given torque is to test it against a known accurate one. You just need a sq to sq in the size of the wrench and match one against the other. It helps to have two people, one to watch each side and then when the known good wrench is at, say, 30 flb note what your wrench reads. Then use that reading to torque to what you need, just remember that you can't interpolate and expect to be accurate, if your wrench reads 32 at 30 then 64 may not be 60. It's better to have the thing accurate but if you are only going for one torque......

Bill
over here in the monsoon

[JimD]

Bill,

Interesting thought. If I combine it with Woodturner's comment maybe I should use my older beam type to check my newer inexpensive click type. I could probably rig up a connection using 1/2 to 3/8 back to 1/2 adapters (or even use my 1/2 inch beam to check my 3/8 click since there is overlap in the ranges).

I tend to believe even inexpensive torque wrenches are much safer than the more typical "tight enough" that I have more typically used with things like spark plugs. Any form of lubrication of the threads (like by anti-seeze) makes a big difference in the stress in the threads. Aluminum heads got me to start using a torque wrench and anti-seeze - I am afraid of my old methods which I used on older steel heads.

For really large bolts and critical applications, we do not use torque in the nuclear industry. We measure the elongation or stress in the bolt. What you really want is a certain clamping load and the torque on the fastener is only a relatively crude way to understand if you are getting what you want. A key variable is friction (which makes lubrication very important). For one application involving smaller bolts, we use automated torque wrenches that look like oversized powered screwdrivers. They do a three step sequence measuring the torque for each step and give you a printout when they are done.

Jim

[LCHIEN]

there's always the old fashioned way...
a fastener torque of 30 ft pounds implies it takes a force of thirty pounds applied at a distance of 1 foot to turn the fastener.
Take a force gage (calibrated) and hook it to a point exactly 12.0 inches from the fastener center and pull at 90 degrees to the line thru the handles and center of fastener.
When the wrench starts to turn, read the force gauge and the torque wrench to see if they agree. A point other than 12 inches (i.e. the end of the handle) can be used when you appropriately scale the force.

[Black wallnut]

Since you brought it up.... I've always read that torque figures were for clean and dry fasteners. If you add lubricant then it takes less torque to reach the same bolt stretch or put simply tightness. I'd be real interested in reading evidence that refutes this.

[JimD]

I don't think it's safe to assume all torque values are for clean and dry fasteners but it is completely true that lubrication makes a difference. If the lubrication (or lack thereof) is not specified, I would normally assume it is implied by the context. A fastener inside the engine would typically be assumed to be lubricated. Something like a wheel fastener, I would assume unlubricated. If it makes a big difference, it should be explicitly specified. So if you assume wrong but did not fail to note that it was specified, you should be OK.

There are also torque sequences that help to make the tension more consistent. For instance, a torque, untorque, retorque is generally more consistent than a simple torque. And the materials of the threads also makes a difference.

Jim

[woodturner]

I've always read that torque figures were for clean and dry fasteners. If you add lubricant then it takes less torque to reach the same bolt stretch or put simply tightness.

In most applications, torque specs are assumed to be "dry" (unlibricated) specs unless specified as "wet" (lubricated) specs.

For example, automotive fasteners are almost always specified as "dry" (unlubricated) specs.

The problem that can arise is that friction in dry fasteners can vary significantly, particularly in older equipment. For example, for the same torque value an old spark plug will be stretched less than a new spark plug, due to corrosion and other sources of friction that build up over time.

As a practical matter, the difference between the wet and dry specs are often minimal.

When I change plugs, what I do is use the antiseize but torque the fastener to the low end of the range. So a spark plug specified at 30 +/- 5 ft-lbs dry torque I would torque to 25 ft-lbs with the anti-seize. Were a wet spec provided, the 25 ft-lbs would likely be within the range of the wet spec.

[JimD]

I found this website interesting for lubed versus unlubed and the effect of plating of various materials on torque:

http://www.raskcycle.com/techtip/webdoc14.html

I found a similar source that recommended reducing torque 20% if you use anti-seeze. 5 ft-lbs would be almost 20% of a 30 ft lb spec. The same source that recommended the 20% reduction also recommended using lubricated fasteners and lubricated torques. That agrees with other information I have seen. The variability in the clamping load for a given torque is less with lubrication.

Jim

[RAV2]

Harbor Freight had the 1/2" click on sale for $10 with coupon the other day - lifetime warranty and plastic case.

I also have the Craftsman beam type from about 25 years ago but it is a challange to use if you are not able to clearly look at the needle during torquing.

I'll bring both into work this week and see how accurate they are against our torque testers.

[cabinetman]

I've got both a click type and a beam type. I prefer the click type as the beam type seems a bit vague with the pointer.
.