Table of contents
Torque wrench introduction
Basic notes about fasteners
Using a torque wrench
Torque wrench types
Torque wrench brands
A torque wrench measures the torque on a fastener. Torque is a measurement of twisting force. There are other methods of tightening fasteners, such as through tension or bolt stretch but this is beyond the scope of this article and not used in your VW manual. A torque wrench is similar to a normal wrench but you should only use it for tightening. Do not use a torque wrench for loosening. When looking at torque specs, always check to see if it's in ft-lbs, in-lbs, or Nm (Newton meter)! Also, the unit of measurement is "ft-lbs", not ft/lbs, that's foot divided by pounds. Lb-ft is the unit when referring to the amount of torque an engine produces.
All critical components must be tightened with a torque wrench because improper torque leads to damaged parts. Some parts call for "x" torque, some call for torquing in stages of increasing force, and some should be torqued in stages even though it's not specified. For example, pressure plate bolts, head bolts, or camshaft retainer cap bolts should be always torqued in stages to avoid warping. Threads should be clean and fasteners should be dry except when specified. Using lubricant will change the torque values, so check to see if the VW specs call for oiling the bolt. Antiseize or threadlocker changes the applied torque so check their product information sheets to see if there's a recommendation on how much to change the torque. I often use blue or red Permatex threadlocker and their technical data sheets say that no torque compensation is required. Your product may be different.
Disclaimer: I'm not an engineer so please suggest additions and corrections to the information below. This is a basic article about using a torque wrench on a car. Have a wrench brand recommendation? Post it in the myturbodiesel forums.
When any fastener is torqued, it stretches a little bit. For simplicity, this article uses the term bolt for all fasteners and assumes metal bolts. Think of a bolt like a coil spring. If you use too little force it's floppy and doesn't hold. If you use too much force the bolt stretches too much and fails to return to its original shape, like an overstretched slinky.
Some bolts used on your VW are torque to yield bolts (TTY) aka. stretch bolts that should only be torqued to their final value once. In your VW service manual, if a bolt says to tighten to "x" torque and then turn another 1/4 turn or 90o, it's probably a stretch bolt. Adding to the problem is that someone might overtorque or reuse them, further weakening them and bringing them close to their failure point.
The cylinder head bolts, crankshaft bolts, and certain engine mount bolts are examples of stretch bolts. Mechanics sometimes reuse them and it holds, but it's also possible (and it has happened many times) to have the bolts break and have the passenger side of the engine fall down, which could result in thousands of dollars in damage and possible injury. The engineers who designed your car, the service manual, and good practices all call for replacement of these single use bolts when removed, so I suggest always replacing them as called for in the manual.
If using aftermarket or replacement bolts of a harder grade, you may have to change the torque spec and not add the additional 1/4 turn. The torque specs for TTY bolts assume that the bolt is of a certain grade and will yield on the last turn. If you use harder bolts it won't yield and could damage the threads on a softer material like an aluminum.
Why use stretch bolts? This method of torquing, the torque turn method, can be more consistent for critical fasteners than just going to a certain torque. "The angle of rotation of a threaded fastener is more directly proportional to the clamp load than the measurement of torque...Because of the additional expense of angle measurement, the torque-turn methods are usually reserved for only the most crucial fasteners in an assembled product" - 2007 Quality Magazine. As a bolt is being tightened, it enters the elastic phase where it beings to stretch but can return to its original shape. Most bolts are tightened to this point. If you tighten it further it passes the yield point and enters the plastic phase where it becomes permanently stretched. You want to tighten stretch bolts to this point. At that point it sometimes feels like the bolt isn't getting any tighter and othertimes you don't feel anything unusual. But if you continue to tighten it, you'll notice a drop off in resistance. Tightening it further to the shear point will snap the bolt.
The majority of clamping force is at the head/shoulder of the bolt and the first 1/3 of the threads. This is why when you apply PB Blaster, liquid wrench, or another penetrating lubricant, the best place to put it is at the thread tip (if you can get access) and under the head. Unless the bolt is holding a clamp, those are the only parts that are exposed anyways. As a rough estimate, about 40% goes into the head, 40% goes into the threads, and 20% goes into stretching the bolt. As an example, the failed bolt shown right was pulled between the first 1/3 of the threads and shoulder. The stretched part has also reduced in diameter. Not even stretch bolts should look like that!
As a total guess, I think one reason why stretch bolts are used for the motor mounts on mk4 VW is so that they break away more easily. The bolts are strong enough to securely hold the engine but in the event of a frontal crash, further stress on the bolts (they are already in the plastic phase) + the design of the motor mounts will drop the engine so that it's more likely to go underneath the passenger compartment. Other cars are designed to do the same thing and they don't use stretch bolts on the motor mount so again, this is just my guess.
Unless otherwise noted, make sure that the threads are clean and dry. You should not apply lubricants such as WD-40 or oil to a bolt because these decrease the resistance to the fastener. You should also clean any damaged threads and in the case of a machined tapered fit, like the camshaft sprocket, the taper should be free of burrs or dirt/oil. "Threaded fastener analysis has shown that most of the energy applied goes into overcoming the friction that exists under the head of the fastener and in the threads...Together these two sources can typically be as much as 80% to 90% of the applied torque energy, leaving as little as 10% of the energy to transfer into clamp load...the presence of dirt on the threads can increase the amount of friction....a drop of oil under the head of a fastener can reduce the friction and allow more energy to flow into the clamp load, which can overtighten..." - 2007 Quality Magazine. In other words, rust and corrosion, or WD-40 and oil on the threads can throw off your torque readings. In other cases, such as the VW head bolts or front crankshaft bolt, a very thin coat of oil is called for. In this case, the oil is being used as a friction stabilizer. Oiling the bolts can even double the loading force! If you use antiseize or threadlocker, see its product specification sheet to calculate the change in torque value.
Another example of lubricating bolts is the VW lug bolts. When tightening the lug bolts, I suggest a thin dab of anti-seize for all bolts on the brakes or suspension. This will prevent seizing of the lug bolts because they would otherwise rust solid and become impossible to remove in an emergency. However, the torque spec that VW gives is for dry and clean threads. Antiseize acts as lubrication and you should lower the torque spec as a result. I've heard of between 10% and 30% lower, depending on the car and the suggestions in the anti-seize product data sheet. I personally don't go too low because many people don't lower the torque spec when using anti-seize on lug bolts and don't report failure. The greater evil is loose or seized lugs when using hand tools and seized bolts/nuts when using impact wrenches. Always retighten the lug bolts after a day or two of driving since they can loosen.
When tightening a fastener to the final value with a torque wrench, you should make a smooth, controlled motion until the final value is reached. The reason why is because of the difference between static friction and kinetic friction. In most cases, static friction is higher than kinetic friction but the relationship changes as torque value increases. Stopping and re-tightening will change this relationship and could change the final value. Jerking the wrench can also throw off the perceived torque spec.
Unless otherwise stated, use the wrench on the part that is being rotated. For example, if you are tightening a nut and bolt, counterhold the bolt and tighten the nut. If you are tightening wheel lug bolts, tighten the bolts instead of turning the wheel and counterholding the wrench!
When tightening (or loosening) head bolts, camshaft cap bolts, clutch pressure plates, etc., do it in stages. For example, tighten it hand tight, then 25% on each bolt, then 50% on each bolt, then 75%, etc. This keeps the object being tightened level and avoids warping. The exceptions are when something has a specified order of loosening.
As a rule of thumb, torque wrenches are most accurate in the middle of their acceptable range and least accurate at the top and lowest 10%. For example, if a wrench is rated for 10 ft-lbs to 100 ft-lbs, it's least likely to be accurate below 19 ft-lbs and above 91 ft-lbs. Of course a well calibrated wrench should be accurate within a certain percentage throughout the whole rated range but you never know! Because of this, an assortment of torque wrenches is a good idea for a well stocked garage. If you were to buy some wrenches, I would suggest a low range one that includes the in-lbs scale up to 20 or 25 ft-lbs and one for about 10-100 or 150 ft-lbs. This covers most auto use fasteners. A third wrench for higher values can be useful for some fasteners.
Universal joints - don't use them with a torque wrench but you can use a socket extension. When using extensions, take care to keep the wrench perpendicular to the fastener and avoid tilting or twisting the wrench. Crow's foot extensions change the lever arm length and add leverage, causing you to underestimate actual torque. If you must use an extension that changes the angle of the wrench, use it at a 90o angle to get an approximate torque and use your best judgment because it's still better than nothing.
If you are using a micrometer type wrench, reset the torque wrench to the bottom of the range after you are done using it. If you don't set it to the lowest range (not necessarily 0), it will wear out the spring inside by taking a set and losing its calibration. After letting a micrometer type torque wrench idle for an extended period, you should set it to the middle of the rated range and exercise it a few times. It can also be damaged if you use it to loosen bolts. Most torque wrenches are for tightening, not for loosening. An exception is the dial/beam type wrench since it's just a bar but they should not be maxed out to either end of the scale to prevent any possible damage.
If find that your micrometer type wrench has been stored improperly and it's urgent that you use it, exercise it a few times to try to reset the spring as much as possible. Tighten a spare fastener to 25%, 50%, 75%, 100% of the wrench's rated range. Let it rest for 10 seconds, exercise it back down, and let it rest for a minute or so. Obviously you should store your tools properly but in an emergency, it's better than nothing.
The most common types of torque wrenches are the beam, click micrometer, click split beam, dial, electronic, and digital. Here are some of their pros and cons. I prefer the split beam, then the micrometer type for auto use. Regardless of which type you get, it should be calibrated to check its accuracy.
The beam torque wrench (below left) is the most basic, reliable, and least
expensive type of torque wrench. Another big advantage is that it can
measure how much torque is needed to loosen a bolt. (You shouldn't use
that torque to tighten the same bolt). If
you don't exceed the metal fatigue strength, it will also be accurate for a long long
time. It has almost no moving parts and if the needle gets bent you can
just bend it back to 0. The
problem is that you have to view the pointer needle at a 90o angle or
else you get parallax, or viewing
error. It's also the hardest to use in
tight spaces where you might not be able to see the pointer needle at all.
Because of this problem in auto use, I avoid this type of wrench. There is
also no way to prevent slight overtorquing since it relies on user precision.
The dial wrench (above right) is also a good design. Like the beam wrench, it's simple and reliable but due to the size, you can't use it everywhere. It also uses a torsion bar but it reads on a round dial instead of a scale. It can be hard to see the dial when used on cars.
The click micrometer style torque wrench is the most common type of wrench
for auto use. To adjust the torque setting, you unlock the handle and turn it to
set the scale on the handle with the scale on the shaft. There
are a few variations on internals, but they are similar in function. When
you tighten the handle, you are turning the handle on an internal screw shaft.
This shaft adjusts the main spring preload which is holding the ratchet head
(with a ball detent between the spring and the head). When the force on the ratchet head overcomes the spring force the
ball detent breaks away and releases the head, hitting the inside of the housing and making the click noise. This
click noise lets you hear and feel when the torque setting has been reached and lets you
use the wrench where you can't see the numbers. Because of the
strength of the click depends on how much force is being overcome, a low torque
setting will generally make a smaller and softer click than a higher
setting. This clicking breakaway also prevents overtorqueing (as long as
you don't continue to turn the wrench) since it releases the force. Below is
a simplified illustration
that helps show how it works. Cordless drills use a similar spring
loaded ball/detent clutch just behind the chuck to limit torque at the
bit. When you exceed a set value, the balls release the torque and it makes
a rapid clacking noise.
Below is a click type
micrometer wrench (lower), with a split beam wrench (upper). These are the
two types I recommend for auto use. Note the
pivot just below the head on the micrometer wrench. If you wish to see the
internals and disassembly of a micrometer style craftsman wrench, check out
The main disadvantage of a click micrometer wrench is that it can be sensitive to being dropped and you must reset the spring to the lowest torque setting after each use or after about 1/2 hour, whichever is longer. If you haven't used the wrench in a long time, it's also suggested to set it to 50% of the wrench range and exercising the spring a few times before use. This is because of spring fatigue or the non-linear properties of the main spring. Another disadvantage is that it can take extra time and effort to turn the adjustment handle since you are working against the pressure of the main spring. It's tiring if you're setting it to a high torque because you're working against a heavy spring. They are also less accurate in the counterclockwise direction vs. counterclockwise.
The split beam click torque wrench is also a common
wrench. I think it's the best for auto use because of durability and ease of use. It has fewer moving parts and you don't have to reset it to the
lowest setting after each use or exercise it before using. It's slightly
faster to use because you're not turning the handle against the force of the main spring when setting the
torque spec. To set the torque spec in a split beam, you turn a small knob on
the side which only needs a light finger touch for adjustment. Its position is
locked under the guard/lock. Due to its design, they
only measure in either counterclockwise or clockwise but not both. Here is
a picture of the scales on a split beam type on left, micrometer style wrench on right.
An electronic torque wrench is like one of the above but with an electronic readout or torque selection. It is not to be confused with a digital strain gauge torque wrench.
The digital strain gauge torque wrench measures torque via a strain gauge. A strain gauge uses electrical current variation when deflected to measure torque. I don't have any due to cost so that's all I really know. Instead of a click, they use a light or beep to indicate that the torque has been reached.
My personal opinion is to spend a little more and get a quality tool for as little as a tank of gas. This article was written in 2008, if the picture to the right shows current gas prices, then tools are cheap by comparison! Remember that a failed fastener on a critical engine component can result in $1000's of dollars in damage!
Cheap torque wrenches from H---- F---- work but don't rely on them for accuracy. If you just need something to torque the lug nuts and don't want to spend much, it's better than the "loosey goosey, whoops, stripped" method. With lug nuts, torque consistency between the nuts is important and a cheap wrench shouldn't be too far off. I would not use them on any critical component like for timing belts or head bolts. Any wrench that doesn't come with a custom certificate of calibration with the tool serial number is not something you should use on the most critical components. You may also have trouble having it calibrated because some torque wrench calibration services won't calibrate these cheap wrenches. It also doesn't make sense to spend $60 to calibrate a $25 wrench but a cheap torque wrench is still much better than nothing. Some report their cheap wrenches are accurate but I wouldn't rely on it.
The Craftsman torque wrenches (pictured below) used to be higher quality but
they have some issues with the current designs. I don't like the plastic handle with the scale painted on them. Some report that
after exposure to auto type chemicals, the painted numbers can come off. The
locking rings on the handle have also had some poor reviews due to
failing. (Mine is still working). The calibration is supposed to be acceptable in most cases but
some say they got a really bad one.
The first one I got had a scale that wasn't perfectly lined up with the shaft's scale
so I had to unscrew it and turn it back to the lowest setting to be really sure
where the marks
were lining up. The second one I got didn't have this problem.
Although craftsman sockets and most tools have a
lifetime warranty, the basic craftsman wrenches come with only a 90 day warranty and no warranty on the
calibration. I own mostly Craftsman tools due to their value and lifetime warranty
but I only bought their torque wrench because I needed one fast and I had to
return the first one because it was unsatisfactory. I also don't like
their hydraulic floor jacks because of leaking and reliability issues. I
don't mind breaking a socket or ratchet but I won't use a leaky floor jack. For the price of 1 tank of fuel, you can buy a quality
torque wrench that comes with a custom calibration certificate and which I found
to be much pleasant to use.
The Snap On tools are high quality but expensive. You can expect a custom certificate of calibration and the tool will last a long time with care. One way to get a good tool at a cheap price is to buy a used wrench off ebay and calibrate it. One seller even sells wrenches that are recently calibrated and he guarantees them too. I would be wary of a micrometer type wrench that has been drop kicked by United Package Smashers but if the tool is packed well, it should be fine.
CDI wrenches are now known as "CDI by Snap On". CDI used to make the wrenches for Snap On until they were purchased by Snap On. It's basically the same exact torque wrench with a different ratchet head. This is a good deal if you want a new wrench. They should come with a custom certificate of calibration.
Precision instruments used to only make wrenches for Snap On and have been making torque wrenches since 1938. Their split beam wrenches are the "old style" Snap On torque wrenches. They should come with a custom certificate of calibration. They make both micrometer and split beam wrenches and I feel they were the best value for what I wanted. The split beam wrench in the pictures above is a Precision instruments wrench. I own this brand and type because I think a split beam wrench is easier and faster to use than a micrometer style wrench.
SK and Matco torque wrenches have good reputations but I don't have experience with them.
The ugly truth is that the techs at the local garage or dealer often don't use torque wrenches because time is money. While there is some margin of error built into fasteners, there are also lots of stripped or loose fasteners. Remember, a torque wrench is measuring only torque applied on the wrench: bolt hardness, size, dirty threads, lubrication, measuring torque instead of bolt tension or stretch, variations between wrenches, etc., all result in variations in force. You've probably cursed over tightened oil filters before - this problem is only the tip of the iceberg. Sometimes they can get away with it but oftentimes, a timing belt fails, a bolt gets stripped, and various components fail, resulting in thousands of dollars of damage when it could have been avoided. Hearing about TDI engines that fell out of the engine bay due to reused and overtightened motor mount bolts or failed timing belts is a tragedy since it's avoidable and many times, due to cover ups or lack of consumer knowledge, cause damages to the consumer. At least in those cases there can be financial remedy but if you're reading this, you're probably working on your own car and you are your own warranty. On my own car, I always use a torque wrench on anything that is important. The responsibility is in your hands, so make sure you've stuck a torque wrench in it!
calibrating your torque wrench: http://home.jtan.com/~joe/KIAT/kiat_3.htm
Have a question or torque wrench review? Share your suggestion in the myturbodiesel forums.