Torque Tolerance

[BoomerRoadie]

Hi all.

I just posted this question in a different forum, so you may see it again, but I like this forum a lot and have always got really good feedback from you, so this is more of a "why not post it here" kinda thing. I'm not fishing for answers.

This is a bit long, but work with me here...

I have a general question about torque tolerance. For the moment, I'm not speaking of any particular bolt or nut or whatever you're torquing . But I imagine all car specifications for torque values should have (or even publish) the allowable tolerance or acceptable range the torque can be set to and it will be effective and safe. As an example, on the car I have, 1997 Boxster, the oil drain plug is stated at 39 ft./lbs. Great! That's probably the nominal value (middle of the range), but what is the range?

Digging a little deeper, I'd imagine that just about every single bolt or screw has its own allowable tolerance and there is no one size fits all, but I want to stay general just for a moment longer...assuming published torque settings have a tolerance, generally speaking are we talking about a couple % either side of nominal or are torque setting tolerances measured even tighter, say a quarter of a % high and low of nominal?

To continue the previous example in a bit more detail, let's say the oil drain plug wants 39 ft./lbs. +/- 3.9 ft./lbs. In this case the tolerance would be 10% of the nominal value. I'm just making up numbers now to tease out the point.

Here's the deal...if I set my crappy torque wrench that has a published accuracy range of +/-4%, and I set it to 39 ft./lbs, and I use it with good technique I will hear the "click" as low as....well wait a second...we have to first understand what +/-4% actually means. 4% of what???

So here's a little more detail.
If I have 3/8" TW with a published range of 10-80ft./lbs., the actual range of the tool is 70, and in this case the accuracy is +/-4% of 70. This yields an accuracy of 2.8 ft/lbs. So going back to my previous example when I heard the "click", the actual torque applied to the oil plug was as low as 36.2 ft/lbs. and as high as 41.8 ft/lbs. I have no idea where it actually is, but I can be sure it's somewhere within that range. Considering, again from my made up numbers from above, the oil plug has an allowable tolerance of 3.9 ft/lbs. the true range of acceptability is from 35.1-42.9 ft/lbs. If this were all true, my crappy torque wrench is not so crappy after all as I hit the tolerance pretty well; I'm right in there.

So first, is my logic correct?
Second, what is a good general rule of thumb for specified torque tolerance, 1%? 10%? Yes I know some things are more sensitive than others, but rule of thumb here...
Third, are the torque settings that are published, published with a tolerance? If not, what is the rule of thumb for those?

Ultimately, this conversion, and understanding the math and data can effectively help you save a TON of money on buying your next torque wrench. Do you need to pay for 2% accuracy or is 4% just fine?

Thoughts please!
Happy Holidays to all!!

Adam

[particlewave]

You are severely over complicating things.
Just buy a good torque wrench and torque to spec.

[JFP in PA]

1. Buy the best torque wrench you can afford. Instead of looking at it as an expense, look at it as an investment in correctly maintaining your car.
2. Every fastener in the car has a torque range; they are published in the OEM service manuals, which again should be viewed as an investment, not an expense.

[bwdz]

It's really not that complicated. You will never achieve 100% accuracy when torqueing bolts or nuts. There are too many variables even in the production of the bolt or nut itself. The slightest bit of dirt in the threads will affect it, the slightest wear on the threads will affect it, the temperature will affect it, the stretch of your socket will affect it, how many times the bolt has been torqued and stretched and the threads stretched will affect it. The correct torque rating is a baseline and gives room for some tolerance, no need to get crazy complicated about it.
Heck, spend a $1000 on a wrench and buy new bolts every time and rechase and clean the threads every time and oil the threads before tightening every time and tell me how much more reliability you get? The answer is probably none over a common sense approach. I have never torqued an oil drain bolt, nice and snug is about it, all of us who wrench know not to crank on them and I have yet to strip one or have one leak.
For internal engine work I do follow all the specs and procedures as carefully as possible but knowing that it can never be 100%.
If you have not wrenched much and never developed that feel in your hands then certainly use a torque wrench but don't sweat being that exact. It is most important that you always clean and take care of the threads and use lubricants or anti seize where it is called for (especially in water passages or in different metallurgies )

[Pdwight]

Originally Posted by particlewave

You are severely over complicating things.
Just buy a good torque wrench and torque to spec.

Totally agree.....consistency is the answer

[Gelbster]

Thread locker of the correct grade is a wonderful aid if the concern is the bolts/nuts loosening from marginally inadequate torque. Use the Wicking grade after torquing to the minimum of the range.

[thstone]

In summary:

Threaded fasteners can clamp materials together only when they are holding with the proper amount of tension. For this to happen they must be properly tightened. To this day a simple, inexpensive, and effective way to consistently and accurately tighten a fastener does not exist. There are a number of tensioning methods that that do this better than a simple torque wrench but they are both complicated and expensive. In most situations, the less-than-perfect traditional method of a torque wrench is sufficient.

Why is a torque wrench with widely-varying known errors sufficient? Because engineers compensate for the inability to consistently and accurately determine bolt tension by massively over-designing joints. This accommodates inaccurate tightening and avoids catastrophic failure. Designers will specify more or larger bolts than needed in order to ensure that the joints are sufficiently clamped together. It is true that fewer or smaller fasteners could be used if bolt preload was more accurately and consistently controlled - but that is not practical nor realistic. And historically, the over-design of the fastener has been far cheaper than controlling the tightening process in a factory (and which is nearly uncontrolled in a DIY/backyard mechanic scenario).

Thus, the fairly large inaccuracies of measuring tension using a torque wrench are taken into account during the design phase such that large errors in applied torque will not result in failures.

What does this mean to you? Simply use the best tool that you have and torque to the specified value.

And now you know why manufacturers use 12 bolts to hold two parts together when it sure seems like 4-5 could have done the job!

[BoomerRoadie]

Originally Posted by thstone

In summary:

What does this mean to you? Simply use the best tool that you have and torque to the specified value.[/B]

And now you know why manufacturers use 12 bolts to hold two parts together when it sure seems like 4-5 could have done the job!

HAHA!! Yes...in summary...

Ok, I'm good now. I have been taking in a ton of information from people like you. This has been helpful. And I'm sure I was over complicating things, but it has more to do with understanding what goes into the actual torque, not the torque itself.

I also do not see tools as investments. Tool are assests. Investments have, by definition a quantifiable return, whereas assests do not. So the real qualifier for an assest is, will it fulfill its duty over its period of expected performance? There is much more to that definition, but for these purposes, your final summary says everything that's needed.

Thank you!

[JFP in PA]

Originally Posted by BoomerRoadie

I also do not see tools as investments. Tool are assests. Investments have, by definition a quantifiable return, whereas assests do not. So the real qualifier for an assest is, will it fulfill its duty over its period of expected performance? There is much more to that definition, but for these purposes, your final summary says everything that's needed.

Thank you!

An investment is something that gives you some type of return on what you paid for it. It could be an asset, or some other type of vehicle. Quality tools that give you the ability to properly maintain your vehicle without having to pay someone to do it because you lack the tools, those savings is where your quantifiable return comes from. And when you decide to stop working on the cars, you can sell them (now an asset) for whatever the market will bear, recover some, if not all, of your original capital investment.

[jcslocum]

Most bolting is a +/- 10% of indicated torque. So 50 Ft. lbs gives you a 45 to 55 window. This typically induces a load of 50% of the yield strength of the fastener.

When torquing, even with a really good torque wrench that has been calibrated, you will still get a 25% scatter of load on the bolts. This has to do with the friction in the fasteners which is impossible to predict. Engineers take all of this into account when they design the joint and fasteners to make it all work.

There is much written on the subject and you can go blind reading it all. Lots of it contradictory.

The best advise has been given; buy the best you can afford and use it properly. If all you can afford is a cheap one, then plan on upgrading as you go.

Archemedes invented the screw 2000 years ago and it's still going strong.

[Gelbster]

Which is why Wicking Grade Thread Locker is so useful if used in combo with correct torque .

[Pdwight]

I only work on my own stuff...but I have 3 relatively inexpensive electronic torque wrenches. I think I paid around $120 each for these

A bike wrench which maxes out I believe at 18 foot pounds, 1/4 inch drive

The do it all 3/8 : that goes to 90 foot pounds

A 1/2 inch one that goes to 150 foot pounds

When I was doing my rear wheel bearings I wished for a big one that went to 400 foot pounds so I could get the axle nuts correct....but had to rely on my 3 foot breaker bar and math and me standing on it.

[Xpit77]

Just set it somewhere between loose and stripped.

[jcslocum]

Originally Posted by Xpit77

Just set it somewhere between loose and stripped.

What I think you mean is tighten until you feel it get easier to turn and the loosen 1/4 turn back to "toight" :-)

[lkchris]

Originally Posted by thstone

. To this day a simple, inexpensive, and effective way to consistently and accurately tighten a fastener does not exist.

Actually, lock washers are indeed simple and inexpensive and effective. And you don't have to calculate or measure anything--just flatten them!

Not the whole story, obviously ...

PS: don't reuse crush washers

PPS: just remember ... threaded fasteners are springs.

[lkchris]

Originally Posted by thstone

Designers will specify more or larger bolts than needed in order to ensure that the joints are sufficiently clamped together. It is true that fewer or smaller fasteners could be used if bolt preload was more accurately and consistently controlled - but that is not practical nor realistic. And historically, the over-design of the fastener has been far cheaper than controlling the tightening process in a factory (and which is nearly uncontrolled in a DIY/backyard mechanic scenario).

Let's not undersell engineers or suggest sloppiness is ok for mechanics ...

There are plenty of instances I suggest where things are not overbuilt ... race cars, for example. (See Colin Chapman). Aircraft ... weight gets pretty critical.

Weight is pretty critical in a world where fuel economy is a big deal, too, as our politicians mandate. I owned several Mercedes before I purchased my Boxster and my first impression was the Boxster was pretty chintzy. Then I thought a little more about it having an aluminum jack.

You want the mechanics that maintain the airplanes you fly in to follow the rules. I worked in nuclear weapons and I know you want rules followed there, too. Let's encourage following the rules in Porsche maintenance, too. I've big doubts whether cheap tools are following the rules.

[plus3db]

You are all bunch of tool snobs

[Racer Boy]

Better to be a tool snob than just a tool!

[plus3db]

Not everyone can afford overpriced Snap-on or similar tools nor do they need them.

A professional , sure. You buy the best you can as they literally are the tools of your trade. You are using them day in and day out. You are generally doing a higher level of work and need the best to do your job properly. You need high accuracy for things line motor rebuilding .

For the bulk of us, DIYers, backyard mechanics, those doing basic maintenance and repairs, oil changes, brakes, an occasional water pump or the like, lower cost tools work just fine.

Is a $10 HF torque wrench the same as a $300 Snap-on, of course not. Is it fine for lug bolts and similar things, yes. BTW, my HF torque wrenches are within a lb or so when checked against my Pros recently calibrated ones. Again, close enough for lug bolts, rebuilding an engine, I would buy a better one.

Am I saying get the el-cheapo made in who knows where socket set for $10. No, but HF has reasonable good quality very suitable for most of us, especially when it is something I may use once in a blue moon

A good analogy is a software engineer, he need the best and fastest computer he can afford. The guys that sends email, browses the web, posts on forums, etc, is fine with a cheap netbook. It all comes down to what you are using the "tool" for and your level of need

Call me a tool if you like .

[jakeru]

Torque specs given in most "official sources", such as say, factory service manuals, are usually given as a range.

But if you only hear one figure being stated (in a venue such as an internet discussion board) it's usually the center of the range. Surely, there's some range. There may also be fastener lubrication conditions also stated or assumed in the manual (such as, oiled fasteners, or dry-assembled fasteners.)

Often times, the torque of reusable fasteners is set such that the fastener gets close to the yield point, but before it actually yields. If you torque a fastener by hand, you can feel right when the yield starts to happen, as the torque stops increasing, but rather, reaches a plateau. This is actually a little bit more than is ideal for fastener reusability, but if you catch it quickly enough, it's pretty close to ideal.