Cool Tool Tuesday #11: Talking Torque wrenches, part two

A guide on how to correctly use a torque wrench.

Don’t miss out on the latest CyclingTips updates.

Jump To Comments

Welcome back to part two of Torque Talk! OK, so you’ve now got a good idea of why you need a torque wrench and what type may be right for your bicycle-fixing needs. Next is to understand how (and how not) to use such an instrument. You’ll be surprised how easy it is to mess up using a torque wrench. 

This topic could quickly become an e-book that only I would buy. But today I’ll stick to the most important things to know and the most common issues to avoid. Alright, it’s time to talk torque, again. 

A quick recap 

Part one of this series covered why going by feel isn’t recommended. Sure some people genuinely do have an experienced feel for torque, but humans are really only as good as their worst day, so a torque wrench is a good idea at least where more delicate or safety-critical components are concerned. 

I also explained that torque wrenches are most accurate in the centre of their intended torque range. “In the aerospace world, It’s common to set some sort of limits. With NASA, you have to use within 20-80% of the torque range of a given tool,” explained Alex Boone, a hardware quality engineer for Motiv Space Systems (a contractor to NASA) of the widely adopted 20-80% rule. Due to this, using the 10 Nm setting on a 10-60 Nm torque wrench isn’t likely to yield the accuracy you should be seeking, and equally, the 2 Nm setting on your 2-24 Nm torque wrench is probably as trustworthy as someone trying to sell you a ‘real’ Rolex out of their coat. 

The largest part of the previous article explained that each torque wrench style has unique positives and negatives. Equally, each style has unique quirks in how you should use it. Unless I specifically mention a style of torque tool below, then just assume it applies equally to all. 

It’s a precision instrument. Treat it that way

A Park Tool beam wrench may be more capable of taking a knock than a jewellery-like Precision Instruments dial wrench, but all torque wrenches should still be treated like the precision instruments that they are. Store your torque wrench somewhere secure where it won’t rattle around or be knocked about by other tools. Don’t drop it. Keep it clean but away from harsh chemicals. And don’t feed it after midnight. 

If you show your torque wrench the same care as what you would a digital caliper or a fine bottle of Japanese whisky, then it’ll reward you by staying within calibration for longer and just generally last longer, too. 

Most torque wrenches are provided with a protective case of some form. This isn’t a tool to chuck into the bucket of rusty screwdrivers.

For those with adjustable clicker-style torque tools, a big part of that safekeeping is to store the tool in its lowest marked setting (never below!) – doing so releases tension from the internal spring and prevents premature calibration drift. I tested this by leaving a Wera A5 torque wrench at high tension for nine months, a test that saw the mean measured torque drop off by 1 Nm at an 8 Nm setting – not a statistically significant test but still one that aligns with why most manufacturers of click-type torque wrenches ask you to back off the tension when the tool is not in use. 

Speaking of storage, if your torque wrench is of the clicker variety and sits unused for weeks or months on end, then it’s a good idea to break it in before each use. “Set the wrench to 50% of the maximum torque and operate the click mechanism five to 10 times before using the wrench on a fastener,” states Jay Seiter from product management and engineering at Pedro’s. “This is also suggested before the first use of a new torque wrench.”

A dry bolt is an under-tightened bolt 

A torque wrench measures the friction of a fastener, and the recommended torque figure makes some assumptions about what that means for the clamping load. And so, to correctly use a torque wrench, you first need to correctly prep the fastener in question. 

“The difference between a dry and lubricated fastener will affect the amount of clamping load you can apply to that fastener for a given torque,” explained Boone. “A lubricated fastener allows you to apply more clamping force for the same twisting force.”

Many brands in the cycling industry have a bad track record for recommending when and how you should prep fasteners for specific products. Speaking with several industry experts, the answer was always the same – unless specifically stated, assume you should prep a threaded fastener with grease, anti-seize, or thread-locker (wet thread-lockers act as a lubricant upon assembly). If a bike or component manufacturer goes to the effort of requesting that a bolt head be kept dry or be greased, then follow that advice. 

Different greases/lubricants/thread-preps do provide differing coefficients of friction for fasteners. However, these differences are minor compared to what is experienced with a dry fastener.

“When the fastener is dry, you have unknown and often variable friction between the thread surfaces, and are also much more likely to experience high and/or inconsistent forces or binding under the bolt head,” added Josh Poertner of Silca. “This is particularly true with anodized aluminium surfaces, which can be quite high friction under load. Things as simple as the diameter of the bolt head may lead to lower than desired bolt tensions even at recommended torque.

“The average rule of thumb is that greased fasteners will require 30% less torque to achieve a given tension, but the more important aspect from my point of view is that bolt tension for a given torque will be significantly more consistent with greased fasteners,” added Poertner. 

This is, of course, a topic of great nuance, and the rabbit hole runs deep once you start to think about things such as specific material differences, wear of threaded surfaces, quality control, how and where the grease is applied, what that grease is, how the threads were manufactured, etc. Truthfully, most brands in the cycling world don’t have answers to such puzzling thoughts, and often it’s a case of doing the basics – use a “wet” fastener and combine it with a torque wrench when suggested. Doing so will get you into the tolerance window the part was designed to handle. 

“But Dave, what about that dried thread-locker compound already on my bolts? Isn’t that dry?”

Ah yes, I hear you. You have some brand new brake caliper/stem/rotor/derailleur bolt in your hand that has a patch of blue or white thread locker. According to Boone, in aerospace the best practice would be to use a digital or beam torque wrench to measure the “running torque” of the fastener: that is, the torque figure required to move the threaded fastener through any tolerance resistance. This measured running torque is added to the final torque figure to meet the intended clamp load. 

However, speaking with brands such as Ritchey and SRAM on this topic, it appears this isn’t a common concern in cycling applications. That hot patch of thread locker is intended to be the lubricant for installation (and a corrosion inhibitor), and you should follow the recommended torque figure and not get fancy with adding a running torque figure on top. It turns out bikes aren’t rocket science. 

The dried thread-locker found on many fasteners is claimed to act as a lubricant and not impact final torque. I know enough to think this is a topic that requires further investigation.

Your grip matters 

How you hold the torque wrench is hugely important to getting accurate results.

Firstly, the tool has a handle for a reason. “It’s fundamental to how a torque wrench operates,” said Boone. “Every wrench is a lever – the longer that lever is, the more force you can apply for a given amount of input. And that’s why torque wrenches have a handle that’s knurled or coloured, and the reason is that you get a consistent moment arm on the force. Don’t choke up on your torque wrench. Don’t put a cheater pipe on it.” 

Every torque wrench has a handle or gripping area of some form. This is very much by design and is critical for correct usage.

Similarly, you should always aim to pull a torque wrench, not push down on it. “It comes down to wrench safety and not directly related to the torque tool,” said Benjamin Glenn, program director at the Bicycle Technician Program of Northwest Arkansas Community College. “It’s basics principles of control and balance. Pulling improves lever control and operator balance.”

Following the pull advice will allow you to achieve your torque slowly and in a controlled way. Beam and digital wrenches force this good technique, but click-type wrenches will provide a more accurate torque setting if you don’t attack into the click. 

It’s not a breaker bar nor a ratchet 

Repeat after me: a torque wrench is a precision instrument. A torque wrench is a precision instrument.

With that in mind, you should only use your torque wrench when wanting to torque a fastener correctly. Don’t use it to undo fasteners because the breakaway torque (which can be higher than the initial torque) can overload and damage the tool. And please, I’m begging you, don’t set your torque wrench to its highest setting and then use it as an everyday ratchet – that’s the tool equivalent of kicking a puppy. 

Many torque wrenches do have a ratchet head, but that doesn’t mean you should use the tool as you would with an actual ratchet.

There are, of course, exceptions. Any pre-set torque wrench with “TorqKey” stamped on its centre shaft (such as from PrestaCycle, Ritchey, Giant, Bontrager, etc.) features a one-way patented mechanism that allows it to be used for undoing bolts. Similarly, the small beam or dial wrenches sold by PrestaCycle and Feedback Sports can be used as a regular bit ratchet if you keep your hands off the torque beam at the base of it. And then there are a few ultra-clever torque tools, such as the all-new Wera Safety Torque A1, which has a sliding collar that allows you to lock out the torque mechanism for it to be used as a ratchet. 

Many Preset torque wrenches of this style can be used to undo fasteners. Look for the “Torqkey” logo on the centre shaft to confirm whether the tool can safely be used for this purpose. If the tool lacks this marking (such as Pedro’s and CDI tools shown in this photo), it is not designed to undo things.
The new Wera Safety Torque features a sliding collar that locks out the torque mechanism. Smart, but also one you need to double-check is not engaged when you’re actually waiting for the tell-tale torque click.

One click is the right amount

“My big pet peeve, seen even at the WorldTour level, is multi-clicking the torque wrench,” said Poertner on the subject of clicker-style torque wrenches. “It’s common to see mechanics re-click the tool two-three times. I have worked with one guy for years who has a ‘rule of three’ where he hits every fastener three times!

“The issue here is that with click-type tools, you can get a sort of hammer effect where you get a torque spike when the tool bottoms out after the click, especially if you are working quickly, so you often see the fastener rotate during these additional clicks which tells you that something is happening. If the first click was 5 Nm, then why did each of the subsequent 5 Nm clicks manage to tighten it further?  Not good.”

This point from Poertner is indeed only relevant to mechanical clicker-style tools. Meanwhile, this isn’t an issue with beam, dial, digital, break-over and newer “cam-over” clicker torque wrenches. Instead, many of these alternative wrench styles encourage you to slowly build toward your final torque and provide you with obvious cues in doing so. 

Only torque a moving fastener 

And what is Boone’s biggest pet peeve regarding incorrect torque wrench usage? “You have to achieve your torque while the fastener is moving,” said the engineer whose technical knowledge is literally out of this world. “I’ve seen mechanics put a torque wrench on a tightened fastener, click it a few times, and think it’s tight. The initial force it takes to get a fastener moving is much higher than your running force. If you’re in doubt, back it off a quarter or half a turn and then get the measurement while the fastener is moving.

“A torque wrench should not be used as a go-no-go gauge,” concluded Boone. To add to this point, the “impact” effect of some torque wrenches, as described above, could also contribute to over-tightening.

Crowsfoot usage 

A crowfoot wrench or similar torque adapter that extends outside the torque head direction may be needed to torque something with wrench flats or a spline. This may be a disc brake hose nut, a powermeter pedal, a direct-mount chainring, a bottom bracket, etc. The challenge is that these crowfoot wrenches or similar can change the effective length of your torque wrench and therefore change the measured torque. 

Simply put, all you need to do is to keep the crowfoot wrench at 90° (perpendicular) to the handle of your torque tool. Doing so won’t change the effective length between the fastener and the torque wrench drive head. This is an often (and strangely) contested argument amongst a few engineers. Still, those who work for industrial torque wrench companies or who have access to calibration machines know it to be fact

The bottom bracket tool adds leverage to the torque wrench in this position. This will add torque to your intended figure. Don’t do this.
Instead, place the tool at 90° and all will be right.

If you cannot keep the adapter at a 90º angle, you’ll need to dust off your calculator. 

Image courtesy of Pedro’s.

Tightening patterns, balancing torque, and compression settling

Tightening a part with a single fastener is easy. However, there is some nuance to parts with multiple fasteners. 

For rotor bolts and four-bolt stem face plates you want to follow a star-like pattern and slowly build up the torque. “It’s a similar idea to how to change car tyres – it’s important to distribute that load across the faceplate evenly,” said Boone. 

An increasing number of products come with specific instructions for correctly adding torque. However, some are still rather vague.

When you’ve got two pinch bolts, such as a stem clamping onto a steerer or Shimano crank bolts, you want to slowly alternate between the bolts, again slowly building up torque. Building up torque slowly will prevent the desire to ‘double click’ parts. 

Many of these multi-fastener parts are likely subject to compression, stretch, or general settling. Because of this, it can be a good idea to re-check the torque on some of these parts after a day or two. Poorly supported steerer tubes can compress, and likewise so can some lightweight seatpost designs. 

Interpreting provided torque figures 

I feel like this article has just cracked open multiple cans of worn, and the following is certainly a large one. 

Firstly, if a part calls for a “Max XX Nm”, then that figure is likely the maximum the part was designed for, and you should torque it to less. I’ll commonly aim for approximately 10% below this quoted figure if I know the part is new and doesn’t have a reputation for slipping. 

Then there’s the fun situation when a part manufacturer gives you a torque range. The 12-14 Nm guideline that sits on almost Shimano Hollowtech crank is a good example of this. Certainly, you should read this as a minimum and maximum figure. But why offer a range and not a single figure? Despite trying, I couldn’t get an official answer on why this is, but others were willing to speculate. 

“Companies doing it right are testing parts in various situations to understand the functional range,” explained Poertner. “We used to do this with stems and other parts where you might test your stem with six different handlebars over a range of torques and find that none of the bars slipped at torques above 6 Nm and that none of them broke below torques of 10 Nm, so you construct a range for people to try and hit. This is especially helpful for people with digital torque wrenches as many of them read a few decimal places of precision, so people will call and ask things like ‘I was trying for 6 and it went to 6.2 before I stopped, is that OK?’

Some components offer a window for recommended torque.

Generally speaking, I’ll torque parts to the lower provided number for new parts and slowly creep up as that part ages. This isn’t a hard rule, and picking a number in the middle of the range is a good approach, too. 

Calibration woes and knows

A torque wrench out of its intended calibration is about as useful as a phone with a dead battery. Buying a torque wrench of at least decent quality from a reputable brand should mean that, at least when new, the torque wrench falls within the requirements of a calibration standard. 

For many torque tools destined for industrial use this would fall into the “ISO 6789” standard (for which there are numerous sub-categories for various torque wrench types). This requires a maximum 4% or 6% +/- deviation from the desired torque figure. The standard also calls for torque wrenches to remain calibrated for a minimum of 5,000 cycles – some professional brands guarantee to exceed the standard greatly. 

Is your torque wrench working as you expect it to?

In many regulated industries (such as aerospace), these standards are taken as law, and torque tools are required to go through continuous calibration to be suitable for use. Other less regulated industries – such as our beloved cycling world – often have far less red tape and legal requirements around such things, which means some brands are free to sell crummy tools that fall outside the international standard requirements. 

Still, while there may not be a governing body watching over your shoulder and ensuring you get your torque wrenches calibrated every year, your torque wrench should work as intended. Beam-style, dial, and digital torque wrenches aren’t subject to traditional torque wrench wear, and at least beam-style tools are always calibrated if their home position falls on zero. However, mechanical-based clicker-type torque wrenches are subject to wear and will most commonly begin to measure low with repeated use.

My first advice would be to follow the manufacturer’s directions related to cycle count and calibration requirements and work out an approximate service lifespan based on your typical usage. 

For example, 10,000 clicks between calibration services sounds like a lifetime of happiness, but it can come up quickly for those working on bikes for a living. Those 10,000 clicks will let you install about 1,666 stems, assuming you get the stem and bar straight the first time. It’s also assuming the tool is never dropped, isn’t exposed to rust-encouraging moisture, and the like. Thankfully there are things you can do at home to approximate the calibration of a torque tool. 

The easiest approximation approach is to test one torque wrench against another which is known to be accurate and/or recently calibrated (easier said than done). This can be done by getting creative with connecting sockets between any two wrenches. Similarly, you can use a digital inline torque adapter for this purpose (but again, you’re likely blindly assuming its accuracy).

And, of course, there’s always physics. This involves working out the lever arm length and using a specific mass. However, such a method is tough to do with certain handle styles of torque wrench (such as the T-handles commonly used in pre-set torque wrenches or screwdriver handles of precision torque wrenches).

Inline torque adapters are a popular method for checking torque calibration at home. Unfortunately, my personal experience points to these tools not always being as accurate as you may expect.

Alternatively, there is the professional calibration route. These services are easily accessible to anyone working in machinery, automotive, manufacturing, or similar industries, and most companies are happy to work with consumers, too. Alex Boone recently got a quote for his home-use clicker torque wrenches: measurement and calibration was approximately US$50 per tool. Whether that’s a justifiable expense will surely depend on the original cost of the tool in question. Certainly, it would be foolish to pay that to re-calibrate a well-used PrestaCycle TorqKey. However, it’s a fair price to pay if you’ve made the initial investment in an industrial-quality tool and/or are working with high-value bikes. 

What did I miss?

A torque wrench is merely a tool; it takes experience to know how best to use it.

The bicycle industry often assumes a broad range of technical knowledge and know-how. Some brands are engineering-led and offer precise installation instructions. While others are pseudo-marketing companies that outsource every part of their product’s creation and leave it up to the user to figure it out. If a brand has gone to the effort of giving exact torque and fastener preparation instructions, then you should assume they’ve done so for a reason. Meanwhile, other components will call for a little knowledge, care, and sensibility.


Alright, I hope all this torque hasn’t made you wrench (sorry). Cool Tool Tuesday isn’t an every-week thing, but it will return next week! I’ve been trying out many new tools that I’m keen to discuss …

Note: A number of the tools mentioned in Cool Tool Tuesday are not sold through traditional cycling channels and can be hard to find, which is also kind of the point of the series. Access to the tools covered will be easy for those in Europe and the United States. Use a search engine to find the products mentioned.

This content is produced independently and purposefully without related advertising, affiliate links, or other commercial interests. Like much of our content, it is funded by VeloClub members. If you found this content valuable then please consider joining.

Editors' Picks