Specified Torque Specs on Automotive Fasteners -- Required or just suggestions?

My academic background is in engineering, but not mechanical. Going back to school for automotive technology. Need to get the perspective on this from someone with a mechanical engineering background.

Service information for automobiles specifies torque specifications for nearly every single fastener. I hear from the "mechanics to be" in my automotive classes, who work at automotive shops servicing cars and are in the field, that they almost never use a torque wrench. They seem to agree that there are critical fasteners such as head bolts, for which a torque wrench should be used. Very rarely it seems that professional mechanics use torque wrenches on nearly any other fastener though. Lug nut/studs are critical fasteners, in which a torque wrench should be used, but in reality at most shops, they just torque them down with an impact wrench. The same thing goes with other fasteners, just use an impact. If an impact won't fit a "calibrated torque arm" is used. Time is money, and using a torque wrench takes time, time is money. "Just tight" will do. So I understand the pressure to get things done quickly.

Anyways my question is, is it really OK to not use a torque wrench on the majority of fasteners on an automobile. I'm not sure how it works, but I would assume a mechanical engineer determined the clamping force required to clamp the two pieces together, looked at the grade of the fastener, it's length, and determined what the ideal torque should be.

I don't understand deviating from this specified value and not even using a torque wrench at all. Not being a mechanical engineer, or more importantly, not being the manufacturer or the exact engineer that determined the ideal torque for said fastener, leads me to think a torque wrench should be used. If the torque of the fastener did not matter, than an engineer wouldn't have gone through the trouble of specifying it. In fact I find a few fasteners that don't appear to be to important that have no specified torque in service information. In which case I just make them tight.

I don't know, I'm just thinking from an electrical engineering perspective, if I designed a circuit to do something, and I specify that a certain resistor should be 2 kOhm, and the technician assembling the circuit decides "oh what the heck, I don't have a 2 kOhm resistor, but I have this 2.1 kOhm resistor that should work just fine", could result in a lot of problems.

So I'm just looking for the perspective of a mechanical engineer, who knows mountains of more information than me, for their opinion on this matter. Is it really ok to just use an impact wrench and "just go by feel" or "use a calibrated torque arm"? Do the engineers who work for the manufacturer know that most mechanics rarely use a torque wrench? I just don't get it. Why specify a torque for a fastener, if the torque doesn't really matter, or nobody will follow it. Because I know nothing about it, I always use a torque wrench when working on cars when a torque is specified, on all fasteners, no exceptions. This is unless I don't have the clearance to do so, of course, which on my particular cars has only happened once.

There are some bolts where it is critical and some where the slop won't matter. Unless you know a priori which is which than you are obliged to use the torque wrench, seems to me. There are clear legal ramifications should you decide to not use ordinary care in doing your job, and this can be documented should a legal finding be required.

Torques are becoming more important, and more critical every year.

For small fasteners, the torque spec prevents you from shearing the bolt, and avoids the problem of pulling a thread, out of say, an aluminium block, or distorting a thin plate component against a gasket.

Bigger fasteners need to be torqued to spec, to ensure that they are tightened sufficiently to do their job, even after you have had a long day. There is no incentive to over-torque big fasteners, so the torque wrench actually saves you energy and time.

There are benefits in weight reductions gained by reducing the size and type of fasteners. As fasteners become smaller, there is less range between sufficient torque and a sheared bolt or stripped thread. That necessitates using some form of torque wrench when doing the job.

There are "torque wrench limiting extension bars" now available, that do not require setting or attention during operation. Those speed manufacture, but make maintenance and service slower, when a general purpose torque wrench must be used.
https://www.ebay.com.au/sch/i.html?...Torque+Wrench+Limiting+Extension+Bar&_sacat=0

There is a limit to possible weight reductions. It often takes more torque to remove a settled, thread locked, or corroded bolt, than it did to install it originally. For that reason, bolts need to be heavier than the critical operation requires. Cylinder head bolts now typically require the bolts to be stretched during installation, then if removed, replaced with new. That, and the use of plastic, is an indication of a mature design, that can be built cheaply, then disposed of rather than to be repaired.

After a couple of years in the trade, a skilled automotive mechanic knows what they can get away with, and what must be measured to the last click of the torque wrench. Until then, use a torque wrench, or you will waste time replacing the components you damage, and getting practice installing thread inserts.

YoshiMoshi said:

If the torque of the fastener did not matter, than an engineer wouldn't have gone through the trouble of specifying it.

YoshiMoshi said:

Why specify a torque for a fastener, if the torque doesn't really matter, or nobody will follow it.

For legal reasons. (Without implying that "It doesn't matter".) If something goes wrong, nobody can accuse the manufacturer or the engineers that they did not specify one. The only torque specs you have are for the suspension/steering components (in case of an accident) and costly items like engines and transmissions. As @Baluncore said, high-tech components are requiring more and more precise torque specs. But I don't expect to see this on fasteners regularly removed and reinstalled, especially where low-skilled people may encounter them.

YoshiMoshi said:

Do the engineers who work for the manufacturer know that most mechanics rarely use a torque wrench?

If they own a car and go to the garage like everybody does, they know for sure.

YoshiMoshi said:

Is it really ok to just use an impact wrench and "just go by feel" or "use a calibrated torque arm"?

Decades of practical experience, worldwide, tend to prove that it is OK.

I don't recall ever hearing about a story of someone losing a wheel on a passenger car because it wasn't tight enough (unless someone just forgot to tighten them completely). Overtighten? They usually break while tightening them. Wheels tearing apart while driving? Maybe, but IIRC the stories I've heard, they are always about trucks and it is more about an incorrectly installed wheel (On dualies) than an incorrect torque (which of course changes once the wheel moves).

And speaking of wheel nuts, they rust and get really dirty; this alters the required torque. Even if you clean them, the rust can pit the threads, which may also affect the required torque. Finally, there is the age-old question: Do you lube the thread or not? The info is rarely specified clearly. Most of the time, it is meant to be a dry torque and you can assume as much. But what if the mechanic before you already lubed the threads? What if the one who did it is the owner of the vehicle, wants it that way, and forbids the mechanic to clean it up? (Assuming a mechanic is ready to spend that much extra time on the job, which I never witnessed or heard of.) Are you supposed to use the same torque setting? In real life, by feel, may not be so bad.

When someone - even an unskilled person - has a flat and puts on the spare on the roadside, does this person absolutely need a torque wrench?

Really, decades of practical experience, worldwide.

YoshiMoshi said:

Because I know nothing about it, I always use a torque wrench when working on cars when a torque is specified, on all fasteners, no exceptions.

Keywords being "Because I know nothing about it". It really doesn't take that much experience to develop a feel for torquing a bolt. Just a little bit of confidence.

If I give you a bolt to tighten with a torque wrench, how many times would you have to do it such that you get a feel to repeat it with a simple ratchet or wrench with the same lever arm length? It really shouldn't be harder than to learn how much force you need to crack an egg without having it crush into your hand.

With power tools, especially impact tools, the feedback might be more difficult to interpret. Sight becomes more important where you notice when the nuts or bolt sets (rapid deceleration) and then you respect a certain angle displacement and/or angular velocity before stopping applying torque. With impact tools, some count the number of hits the tool makes, known as Ugga Dugga (which clearly depends on both the tool and fastener used):

Baluncore said:

After a couple of years in the trade, a skilled automotive mechanic knows what they can get away with,

If you need two years of practice to develop a feel for torquing a bolt - tight enough without breaking it - you probably are in the wrong field.

I had my wheels mechanic shop wheels changed at one time.
Taking them off and some of the lugs bolts broke, and had to change them also.
I would assume that they were overtightened and the thread had deformed and, which is as unsafe theoretically as under tightened.
Thankfully I had not a flat to be switched with the spare alongside the road - ie stranded in cold weather is not pleasant/

YoshiMoshi said:

TL;DR Summary: Is a torque wrench really needed for most fasteners on an automobile?

I specify that a certain resistor should be 2 kOhm,

Actually not a very good a comparison., since no tolerance was given.
The circuit should specify, or if not, picking one out of the grab box could be OK.

Do you know about the wide tolerances of resistances.
Critical circuits would specify.
and also wattage.

The most common way of specifying resistor tolerance is by percentage. When specified by percentage value, this percentage means the amount by which a resistor may vary from its nominal value. For example, a resistor which has a tolerance of 10% may vary 10% from its nominal value.
Typical resistor tolerances are 1 percent, 2 percent, 5 percent, 10 percent and 20 percent. The value can even be lower than 1 percent with high-precision resistors.
http://www.learningaboutelectronics...is the,value and, thus, the more stable it is.
So even there ( electrical circuits), acceptable repair should follow the circuit design criteria.
Note that as tolerance decreases, cost increases.
-

YoshiMoshi said:

Because I know nothing about it, I always use a torque wrench when working on cars when a torque is specified, on all fasteners, no exceptions.

That is a good practice, as well as developing a good hand feel.

Steel is a very noble and tolerant material; aluminum threads are more sensitive to over-torque.
You may find yourself in emergency situations in which a torque wrench is not available (like changing a tire on the side of a busy road in the rain or at night).

There are configurations that make using a big wrench impossible.
Those cases require creative use of several tools, couplings, extensions, etc. and precision torque is not achievable.

There are millions of mechanics in the world who never had access to a torque wrench.
Then, there are torque wrenches so poorly manufactured, or maintained, or stored, which never click, or give a huge error in torque input.

The post by @256bits is good stuff.
-
Concerning the example of using a 2.1 Kohm resistor instead of 2 Kohm: I would see nothing wrong with using a 2.2 in place of a 2 in the right circumstances. If there was a resistor needing to be replaced that is in series with an LED it would not matter one bit. The question arises of why something like this would need to be done in the first place. Only time I could see would be physical damage. Resistors in that circumstance just don't fail.
-
A good tech in mechanical or electrical repair is going to realize that there are places varying from specs just doesn't matter and other places it is out of the question.

Averagesupernova said:

Concerning the example of using a 2.1 Kohm resistor instead of 2 Kohm: ...

The tolerance band on the original component should be sufficient in deciding the replacement.
Without that;
2k0 is not an E12 value, as it would have been 1k8 or 2k2. That tells us indirectly that the value is more critical than the common range of E12, ±10%.
2k0 first appears as an E24 value, so should be ±5%. Unless tolerance was otherwise specified, the acceptable actual value required would be in the range from 1k9 to 2k1.
https://en.wikipedia.org/wiki/E_series_of_preferred_numbers

A torque wrench is usually adjusted to a linear scale graduated in steps of 5 ft⋅lb. That is surprising because specs typically range from 15 to 180, so a percentage tolerance would be expected. No doubt a linear scale is used because the deflection is linearly dependent on torque.

Torque limiting extension sets, come in fixed step values of 10 ft⋅lb, which would suggest that torque is not as critical as is often claimed.

A mechanic needs to be able to declare to the customer, or the coroner, that the wheel lug nuts were torqued to the specified value. For that reason, a mechanic develops a ritual, where the click of the torque wrench, plays an important part in the completion of the task. The task is not complete until the torque wrench has clicked. When some mechanics tighten a nut by hand, without a torque calibration, they say "click" when they are satisfied, before moving on to the next task.

In the novel, "Round the Bend", by Neville Shute, Muslim aircraft mechanics need to know when they must stop work to pray. The answer comes back, that they need not stop to pray during work, as each time they complete a task, they declare to Allah that they have done the job as required. Neville Shute was an aviation engineer in the 1930s, concerned with (the religion of) quality control.
https://en.wikipedia.org/wiki/Nevil_Shute
https://en.wikipedia.org/wiki/Round_the_Bend_(novel)

Baluncore said:

2k0 is not an E12 value, as it would have been 1k8 or 2k2. That tells us indirectly that the value is more critical than the common range of E12, ±10%.

Not necessarily. If another part of the circuit uses 2K resistors, using a 2K where the tolerance could be wider is not uncommon in order to keep the number of different parts on a bill of material low. Keeps from having to load yet another reel of parts on a pick and place or auto insert machine.
-
Similarly, if I have a machine loaded with grade 8 bolts where they are legitimately needed it seems silly to use grade 5 or 2 bolts in a few places where 8s are overkill. Eat the extra penny and don't take a chance on getting the grade 8s mixed up with inferior bolts out on the production floor.

Averagesupernova said:

Not necessarily.

There can be exceptions to every guideline, but without significant reverse engineering, and a thorough understanding of the design requirements, we must replace each faulty component with one of the same specification, or risk failure.

I do not trust a reverse engineer to understand the subtlety of my mechanical, or circuit design, nor my particular choice of a material or component value. Indeed, I go out of my way to make it difficult, so they will make mistakes, to destroy their reputation in the market. I do enjoy their failure. They must pay for the IP, or produce an inferior product.

"Chinese junk" is manufactured by copyists, who make it look OK in the listing, so it will sell. When their customers complain that the "visual likeness" is functionally unreliable, they dump it on the mail order market, and fly-by-night.