Torque wrench recommendation?

Well, I guess we can agree to disagree here Alan. But I find the approach you mention to be much less sophisticated in terms of the actual physical properties involved in the field. I'm a skilled technician so what I practice is not "back yard mechanics" - that was false modesty.

There's no luck involved. It's experience, both in an academic engineering sense, and a skilled mechanic sense in the field. The fact is, I've made it a life long study to understand the physics involved generally, and more specifically, the precise physical characteristics of fasteners, and how they interact with the surrounding material. If you do this, then you have a keen sense for the large range of variables that go beyond just the obvious ones such as metallurgy, temperature, galvanic conditions, lubricants/CoF, etc. Add on top of that the obvious variables in the field such as changing metallurgy, temp cycles, wear cycles, surface deformation, foreign particulate in the thread, etc. and a static torque setting looks pretty darn unsophisticatd.

My company makes the world's premier sub-surface sensing devices - mag, x-ray, nano are just a few of the technologies which we have pioneered in different applications. Our pipe line sensors scan most of the world's important pipelines for example. We actually do most of the standards setting in this regard. Our sensing devices are used extensively by Boeing and US Airforce for the purpose of flaw detection in critical aerospace parts. Why? Because our expertise materially exceeds theirs. We also happen to make the world's most precise thread gauges in part because there is some technological overlap here. So, rest assured, I'm not talking from the perspective of some dim witted shade tree mechanic "lucky enough" to have some smart engineer doing my thinking for me. What's important is to know that a static setting is not appropriate for all (not even most) field settings and environments, and to go get the knowledge and experience to know this.
 
Last edited:

Seymour Snerd

Lifetime Supporter
Cliff --

If I understand your correctly your argument is that:

  1. In real field situations, fastener conditions vary enough from those that were the premise for the design engineer's torque specification that the correct value no longer falls with the tolerance of the original specification (typically a range of 20%). I'll accept this temporarily if you'll supply directly supporting and credible literature citations.
  2. As an aside, apparently manufacturers such as Toyota, Ferrari and Porsche don't know this, or for some other reason continue to publish fixed fastener torque specifications in service manuals even though to do so is pointless because of #1. Can you explain this at all?
  3. You personally can set fastener torque precisely to that new setting, using only your own tactile sense over the entire rage of ~10-200 ft-lb as is commonly found with automobile fasteners. Could you explain how you've achieved that level of accuracy, precision and consistency well enough to satisfy #1, and more importantly how you know that you've done so?
  4. Only cylinder head bolts/nuts are an exception to all of the above. Why?
  5. Your company has advanced expertise regarding fastener design and behavior to the point where you resort to false modesty on the subject. Why would someone in your position have to track down and pay for a calibration expert for your torque wrench with all that at hand?
 
Last edited:
Measuring bolt stretch is the most accurate method of correctly tightening a fastener. This is the preferred method for rod bolts, where both ends of the bolt are accessible for measuring bolt stretch. Measuring applied torque with a calibrated wrench runs a distant second in terms of accuracy, because there are so many other variables introduced into the process, namely friction between the fastener and the material being fastened. Hence the need for lubricants to minimize the effect of friction in the threads and under the bolt head.

As for hand-tightening to a given torque spec, the only thing I know for sure is when I've gone a wee bit too far.
 

Howard Jones

Supporter
We calibrate torque wrenches that vary from 25 inch pounds to 600 ft pounds once a year here where I work. I asked the cal guy, they come out with their equip in a van, to do mine a few years ago and he said that both of them were right on. Both are Proto's. One is a 0-150ft/lbs and the other is a 0-60 ft/lbs if I remember right. Anyway I asked how may of the companys were out of cal. He said none. They are usually either broken or in cal. I'm talking about more than a couple hundred wrenches. I think it is pretty unusual for a torque wrench to fall out of calibration unless it is broken or abused.

The main trick is to keep them clean, dry, used within the middle 80% of it's range, and set to zero when not in use. A good wrench should last forever if it's treated like it's your money.
 

Seymour Snerd

Lifetime Supporter
A good wrench should last forever if it's treated like it's your money.

I'm glad you said that. When I took my HF one apart and realized how the "clicker" type work (a short stiff spring against a detent, with variable preload on the spring) I couldn't think of a "wear mechanism", except something that would take place in the first few uses as the mechanism beds in and settles. Presumably the manufacturer takes care of that.

As for the beam type; how would they drift significantly unless used incessantly?

BTW I did read somewhere recently a supplier suggesting storing them at a low but not zero setting, since (it was said) the latter could induce inaccuracy.
 

Seymour Snerd

Lifetime Supporter
Just as an update to this topic:

I today took delivery of a Northern Tools 3/8", 0-99 ft-lb. digital torque wrench.

(Northern Industrial Digital Torque Wrench — 3/8in.-Drive | Torque Wrenches | Northern Tool + Equipment)

To my surprise it came with an individual calibration sheet showing its accuracy to be more like +/- 1% as opposed to the specified 2.5%.

I took this occasion to compare it to all my other ones, as follows:

  1. Recently professionally re-calibrated Sears Craftsman 44598 "Torque Meter" (see post #15, it's *extremely* accurate). Had not been used since calibration.
  2. 10-yr old never calibrated Sears Digitork click type 1/2" wrench (that had the famous slipping number window problem that I fixed myself rather easily).
  3. 30-yr old Sears beam type 1/2" wrench never calibrated.
  4. 15-yr old KD tools 240 in-lb click-type wrench never calibrated.
  5. Harbor Freight "Pittsburgh 807" wrench I paid $20 for a few years ago and that on first use was so obviously and wildly wrong that as a lark I took it apart and recalibrated it against #1. Doing that was almost as easy as the 1-hr round-trip it would have taken to get another one that might have been just as bad.
The upshot is that all of them are within a 2-3 ft-lbs of each other in the 20 ft-lb + upper ranges, and all within the tolerance of torque specifications commonly encountered in automobile repair documents. In particular the new electronic wrench and the calibrated Sears "meter" (#1) tracked each other very closely above 10 ft-lb.

So if I were buying my first wrench today it would be the Northern Tool 152585.

One thing I learned in this exercise is that the click-type wrenches vary significantly in the tendency to apply over-torque in follow-through after the click. Obviously this somewhat a function of operator skill, but some were quite a bit more difficult to control after the detent has released; this has to do with the length and force-slope of the "dead zone" after detent release, and is probably also affected by the friction and lubrication in the detent mechanism. Something to play with if you're ever comparison shopping for one of those.

This is easy to evaluate if you connect the wrench-under-test to an electronic one in "peak-hold" mode.

Of course whether it's an issue in a real-world wrenching situation depends to some extent on the fastener's static-vs-dynamic torque characteristics, since at the moment of detent release the fastener is no longer moving and static torque is probably dominant, which would mitigate or eliminate the impact of the effect.
 
Last edited:
No mention of properly lubricated threads?

I would bet that the difference between dry and lubricated threads is greater than the error in a cheap torque wrench.

Just my opinion.
 

Seymour Snerd

Lifetime Supporter
No mention of properly lubricated threads?

I would bet that the difference between dry and lubricated threads is greater than the error in a cheap torque wrench.

Just my opinion.

Yes, the difference in fastener tension between dry and lubricated is on the order of 20%.

But that's a question of proper usage, not tool accuracy or qualityi.
 

Ron Earp

Admin
Probably doesn't need a better one. Really. Grassroots Motorsport Magazine (either that or the SCCA's SportsCar) did a test of torque wrenches, from the lowly $19 Harbour Fright unit up to a $XXX Snap On piece. While The Snap On wrench was nice, the low end and mid-range units got the torque values right too and if I recall correctly the cheap HF was the MOST accurate. At least for the test they ran it looked like a decent wrench could be had for not much coin.

Ron

Good info to have and it looks like you've duplicated the GRM test that I referred to above. Big money does not have to be spent on torque wrenches. They are useful and I wouldn't assemble my engines without one, despite what others on the thread may do. Without a torque wrench I can snug things up tight, damn tight, and fooking tight, but I have no idea what the actual torque value will come out to be. Probably more depends on how I was feeling at the time than anything else.

R
 

Pat Buckley

GT40s Supporter
It always amazes me to watch those who should know better torquing a fastener incorrectly.

I'm talking specifically about not allowing the final sweep of the process to end up at the desired value. Too often I see the wrench stopped at perhaps 80 or 90% of the swing and then restarted.
 
I still have & use for the majority of engine assy the Warren & Brown 'clicker' (120ft lb max)that I bought in 1970, I also have a Norbar ( 40 ft lb max ) that is used mainly on auto trans assy ( bought in the early eighties). The Norbar has an adjustable spring tension clicker that has to be let off after use for storage to remain consistant. Changing torque wrenches every few years for the sake of change seems foolish to me, we would have to develop a new 'feel' for the new equipment. The biggest variable in torqueing fastners is me (or you), if your shoulder is a bit stiff or your working in a confined area there is no way you can 'judge' even torque values, so I dont go along with Cliffs ( I only use them for head bolts) statement above..... plus if he had used it a bit more regularly on his intake manifold bolts.........:)
 
the other thing about bolting alloy parts together with a torque wrench is the bolts are done up evenly so less likely to warp, important with modern Japanese / Korean cars and there light weight alloy wheels, it is amazing how easy it is to buckle one if you don't take a little care.
 
Good question and heck if I know. The last time I saw one calibrated it was at Sears, but that was more than ten years ago. I bet there are some interwebz articles about performing your own calibration.

Around here there are several places that deal with it for industrial businesses (typically oil and gas stuff). It costs around $60 for a torque wrench to be recalibrated and recertified.

Here's one that does it

N-Tech, NDT, NDE, Inspection Equipment, Non destructive testing, magnaflux, Borescopes, Fiberscopes, Videoscope, Remote Visual, Calibration
 

Chris Duncan

Supporter
Going to have to side with Cliff on this one. The only place I torque is head bolts and wheel lug nuts. If you've been wrenching long enough you can feel the right torque. That said it can't hurt to torque stuff if it's critical and you're not sure of your skills.

Most of the major car manufactures have gone to a rotation degree spec instead of a static torque on the head bolts, so that sort of argues that a static torque spec is not as accurate.

As far as torque wrenches the beam type are by far the most economical if you are in a seldom use scenario, they are also the most dependable, no moving parts.

The Snap-On clickers are my favorite for something used more often. You can leave them on any setting, no zeroing required. Most Snap-On trucks have a calibration gauge and they will do it for free.
 

Seymour Snerd

Lifetime Supporter
If you've been wrenching long enough you can feel the right torque.......

Most of the major car manufactures have gone to a rotation degree spec instead of a static torque on the head bolts, so that sort of argues that a static torque spec is not as accurate. .


I'd be interested to know what procedure you've gone through that convinced you that you are always applying correct torque without using a torque wrench. How did you measure the actual torque you applied once your hands and arms became so finely calibrated? And how do you know that you continue to do so?

BTW I'll be glad to conduct a blind test the next time either of you is in my neighborhood. I'll call out various torque figures while watching a calibrated meter that you can't see, and measure how much you exert. It'll make a great Youtube video.

Yes, some manufacturers are specifying a more time-consuming and complex tightening procedure because it's more accurate. Duh, why else would they put professional mechanics through that? And how does that support or even relate to the issue of proper use of a torque wrench in other torque-specified fastener tightening situations found in the very same shop manual?
 

Pat Buckley

GT40s Supporter
I would include rod bolts and main cap bolts to the list of fasteners that I for sure use a torque wrench on.

As for using the "arm guage" - I do think that you can repeat past performance on a pretty consistent basis -

I do not use a torque wrench to tighten spark plugs for instance.

If you can say that you do, then good for you, but if you don't then I think you can agree that this isn't a black and white discussion.

Well, do you?
 
Back
Top