How to do a sanity check on ur torque wrench (Vid)
 

Honda CBR manuals list the torque setting on the critical components on your bike. Too tight and the bolt can snap, too loose and it'll shake loose. Mistakes can be nasty. So just use a torque wrench right?

But let's say you haven't used your torque wrench in a while. Or did u buy a 'cheap' torque wrench and you're not sure if it's accurate? If you put the engine back together with the wrong torque settings, it's not fun taking it all apart to retest the settings. You could send your torque wrench away to be tested - but that could take weeks and cost you hundreds. Ideally you should get your wrench calibrated once a year and that will cost you ~$100. I just bought a Alltrade Digital Torque Adaptor on Amazon for <$40 so when I get to try that out I'll update this How-to.

Here's a sanity check that will take you less than 30 mins and you can do it with ordinary stuff around your house. It's also a good idea to do this ASAP to test your torque wrench - this will provide a benchmark for you. Then next year if the results change U'll know UR torque wrench is off.

To do it yourself all you need is:

1. Torque wrench
2. Bolt to anchor it (using a bolt on the bike frame)
3. 2 gallon jugs
4. hook to attach weight (I used a coat hanger and an S hook.)
5. professionally calibrated accurate scale (if you don't mix chemicals in your house you probably won't have this :) so run over to the self serve kiosk at the post office like I do - others recommend the grocer) *

Whenever you use your torque wrench prep it by giving it a 'warm up' so that it will click consistency. Set the toque in the middle of the range. Attach it to a nice big bolt on your frame, and provide constant pressure on the handle until it clicks 6 times.

Simple formula:

T = L * W + Tw
T is the torque*
L is the length from the center of the socket head to the point where the weight is applied (measure in inches for inch-pound wrenches, in feet for foot-pound wrenches)
W is the weight of everything that is attached to the end of the wrench.*
Tw is the torque associated with the weight of the wrench. Ask your manufacturer to supply this! They should do this for each model they produce. If we all ask them maybe they will start doing this!*

Tw an be approximated*:
Tw = 0.5*L*W

In the video below the torque wrench should have been set a little higher as it didn't account for the weight of the torque wrench itself:

T = L * W + Tw
174=9.5" * 17.5# + 8

(Tw=.5*9.5*1.7= 8)

Check out the sanity check in action at:
Simple DIY Torque Wrench Test - YouTube

* NOTES
1. When weighing: Include everything: the water jugs, carabiner etc. Get everything weighed at the grocery and/or post office - as they use pro calibrated scales. If you're going to the grocer, try to buy something with a convenient hanger on the top. I used water jugs since it's so obvious if there's something wrong since 1 Gallon of water = ~ 8.35 lb.

2. Try to pick weights that will produce a torque matched to the job at hand. In my case I was trying to produce 180"# torque. Remember that your wrench will have difference variances at different settings- although as a rule of thumb, wrenches have smaller variances at the higher ends of their scales. For higher torques, pick something bigger like liquid laundry detergent. If you are testing bigger weights and go with dumbells, don't forgo having it professionally weighed along with whatever you're using to suspend it to ur wrench.

3. I called Tekton who sells my wrenches. They put me thru right way to a really knowable technician who turned me on to the digital tester, and they mentioned in the future they plan on providing Tw! How kewl is that? Either way when I get my digital tester i'll measure Tw. This formula is only accurate if the center of gravity is in the center of the wrench. If your torque wrench is fairly well balanced the center of gravity will be in the center of the shaft and this will be a good approximation. If you want something more exact, use @DonnyBrago's "method B" further on in this thread which measures the torque applied TO the socket so the torque OF the socket wrench is not important.

4. Always store your wrench on the LOWEST setting. This will prolong its useful life.

5. If the torque setting are off, remember its often faster & cheaper to pick up a new one then to get ur old one recalibrated. Some torque wrenches don't even allow for adjustment.

In case you're interested I use three clicking torque wrenches. I don't like the ones with the scale on them 'cause they have smaller ranges and aren't as accurate. In addition to the Tekton 2432 1/4" 20 - 200" lb shown here I also use a 3/8" # pound Tekton 2435, 120-960-"# and a big 1/2" foot-pound wrench.

I have two criticisms:

1: You are not taking into account The mass of the plastic jugs. A better thing to do is go to a supermarket and buy a big bottle of water, one of the bigger bottles with the handle around the cap is perfect. Take that bottle over to the butcher/deli counter and ask them to weigh it and put a sticker on the side, i.e. weigh it as steak and put the sticker with the weight and price on the side of the bottle. The balances in these places need to be accurate and routinely calibrated, as otherwise they would be stealing from customers or giving away free stock.


Use that as your weight and it will be spot on accurate.


2: in your case you are not taking into account the weight of the torque wrench itself. The weight of the wrench produces a moment around the bolt, i.e. it is producing torque on the bolt, your calibration will be wrong as you are not accounting for this additional torque. That coupled with your magically massless plastic jugs means you will be producing considerably more "torque" on the wrench than you think you.


No offence, but your way is really quite crap.

@Donny thanks so much for your helpful suggestions! I will update my post above - This is yet another reason why this forum is so helpful!

You are absolutely right that it could be that my scale isn't calibrated correctly. It is a good suggestion to get the weight doubled check on a grocery scale or postal scale. I will weigh again and add ur commented to the How-to.

I used an electric scale to weighed everything I stuck on the weighing end. I didn't weigh the water separately from the jug :) Also using the known quantity of gallon containers was as a good sanity check. This way if my scale was telling me my gallon weighed less we would have a problem with the scale. And maybe the grocer's could have an incentive to overstate each item they weigh :)

Yup again ur absolutely right. Not taking into account the weight of the wrench produces systematic error in favor of wrenches being under calibrated (ie. needing to turn too much before a click) SO in my case if the wrench were suspended as a weight on the end, this would produce a torque of ~16"#. However since it's acting as 1/2 full moment it produces 8"#s. The formula for the torque should be:

T = L * W + Tw

Tw = 0.5*L*W
Where Tw is the torque of the wrench.

in my case Tw=.5*9.5*1.7= 8

Um... No offence taken? Now that I've updated with your suggestions hopefully my NEW and IMPROVED sanity check is less crap. :)

I'm still not sold on the torque wrench torque estimation, you would need to know the centre of mass of the torque wrench and calculate the torque produced as a product of the distance to the centre of mass from the pivot. Even then it would only be an estimate. To illustrate what I mean consider this example: try calculating the torque produced by an object like a sledgehammer - hold it at the head and the torque produced is minimal, hold it at the base of the handle and the torque produced would be great - unless you know the centre of mass of the object and the distance of this centre to the the pivot point you cannot calculate the torque produced by it accurately.

To my mind in order to do it simply and accurately, the easiest way is to avoid the torque produced by the weight of the wrench altogether, this is achieved by having the wrench sitting vertical or by using something else to attach your weights to making the weight of the wrench irrelevant. Two ways I can think this would work, detailed below as A and B. Method A has the wrench vertical, no torque produced on the bolt, method B separates the torque wrench from the torque calculation system entirely.

A - essentially have the wrench vertical and transfer the force to it via a pulley, essentially your method rotated 90deg and using the pulley to transfer force to your wrench - the calculation would be simply the first formula you posted.


B - this uses a rod (orange) that pivots on the yellow triangle in exactly the centre i.e. if you were to not support the orange rod it would be balanced perfectly. Attached to the orange rod is your nut for the torque wrench, and hanging on the end of the orange rod is your weight. The second yellow triangle keeps it horizontal. Calculate the torque produces by length x weight - set your torque wrench, a calibrated one will click at exactly the point that the orange bar starts to move.


http://i1015.photobucket.com/albums/...rquewrench.jpg


Hope this helps.


I have used method A as a rough estimate in the past, the downfalls are that we must assume the pulley has 0 friction losses and it is quite difficult to set up and operate accurately (i.e. the pulley height needs to change every time you adjust the torque wrench so that the string is exactly parallel to the ground).

Again Donny you are right- there will be systematic error introduced in any test where Tw is unknown. In the how-to I'll point ppl to ur method down here for more accuracy.

We need to know the center of gravity and precise weight (with relation to the turning axis of the wrench). Wouldn't it be nice if each manufacturer just measured and provided Tw? I called Tekton and they're going to do it! In addition, they turned me on to an Alltrade digital tester on amazon for <$40 so I'll put those Tw measures here for my torque wrenches.

So using the 1/2 method I calculated Tw as 8. even if it's off by 2 that only introduces a 1% error in the test. In any event in a few short day's I'll have the tester and I'll know for sure! If we do the sanity check when the tool is new (as I'm doing) and then check year over year with the same methodology and the same results we'll know the tool hasn't degraded over time. So once it changes from the baseline it's time to chuck the tool.

If autozone or Halfords or advanced is listenning - wouldn't a torque tester be a great service to provide IN STORE. - Certainly would get me in their stores as often as their fancy $400 OBD II reader :) [Donny I added Halfords so you would know what I'm talking about]

I really like your method B. In fact I have a really good way to simplify it and it could be commercialized- for like $15! Call it a torque bar.

It's just a bar with a bolt head in one end and a hook one foot away measure the torque the bar produces and your good to go:

T = 12W + 200

Where 200 represents whatever Tt was measured to be.


You could also make a breaker-t-tester and it doubles as a breaker bar.

Measure (or Calculate) the Tt (torque factor associated with t-tester)
1. Take a breaker bar,
2. Socket head to socket head socket duplexer (something to connect torque wrenchs -1/4" 3/8" 1/2" to the breaker bar)
3. provide a hook at exactly 12 inches from the head

T = 12W + 200

Where 200 represents whatever Tt was measured to be.

And provide the formulas for foot-pounds and maybe even metric for you guys across the pond. :)

As long as the weight of the breaker bar stays as shipped it's accurate. No recalibration necessary like other testers. And every wrench head keeps one in their tool box.

Glad I could help :icon_wink:.