I bought a torque wrench in May, I sent off the warranty registration, I have now received a letter dated August from the manufacturer offering to check the calibration ... very kind of them ... at a cost of £15.00 plus postage ... but it's only three months old ... It should not need recalibrating so soon, should it? Perhaps it is not really neccesary and this is a nice little revenue generation exercise, the weight of the thing means the Post Office could do quite well out of it as well!
Surely it should either be guaranteed to be accurate for the one year warranty period or if recalibration is required it should be at no additional cost?
Any thoughts?
|
Many years ago when working for British Rail Engineering we started using torque wrenches - and started snapping bolts!
Was it the torque wrench calibration - no (we had them re-calibrated!!!) Was it the plating (we tried ECp, EZP, Black) - no! Was it the oil on the bolts - no! Was it .......Caused many days of thought - and we were snapping bolts up to M24 and in grades up to 14.8 (8.8 is the norm).
The cause was the calibration body who had calibrated our torque wrenches had not got their calibration machine correctly calibrated to a national standard)! When that was sorted everything was fine.
1. Did it come calibrated?
2. Is their calibration system accurate?
3. How much is it used?
These are all factors in calibration.
|
Like Ian, I have had to deal with problems with bolted joints in rail applications - although in the problem I was looking at, the bolts were failing under fatigue loading in service - which caused me the odd sleepless night!!
An intermediate solution, which may be appropriate if the bolts are critical, but not aerospace critical is to install a torque meter, and set the wrench to the torque meter, rather than the scale on the shank of the instrument. See the link below.
>>www.norbar.com/products_range.php&category_multid=...7
This enables many mechanics to set the torque wrenches to a traceable standard, without everyone having to send their torque wrenches away every so often. This type of solution is typically used in larger workshops, such as railway depots.
Despite looking simple, the torque bars which rely on reading the bending of a bar against a scale are not bad devices (as long as they have never been stressed beyond their elastic limit). As the torque / displacement relationship for this type of bar is
Torque/Displacement = 3 * E * I / (L^2)
and E, the Young's modulus of the bar is well known for steels
I, the second moment of area is very closely known for a manufactured bar
The only thing which is at all sensitive is the length which is subject to bending - that is controlled by alway measuring at the same point along the bar, using the scale.
Number_Cruncher
|
If it was just being used occasionally (like mine) I wouldn't worry about calibration. Just test it against something else every so often.
the bolts were failing under fatigue loading in service - which caused me the odd sleepless night!!
Surely you mean sleeperless nights?
...
I'll get my coat!
--------------
Mike Farrow
|
|
|
|
It sounds like it's just a quirk of the manufacturer's system. If you know someone else who has got one, or someone who uses one at work, then you could check one directly against the other. It's perhaps not a method that a pedant/purist would agree with but it should be good enough for most purposes.
--
L\'escargot.
|
|
Sounds to me like it would only be useful for a professional using it on Aircraft or some other critical area, where if there was a problem, they could cover their backs and show they had their tool regularly calibrated.
|
1. Did it come calibrated?
Yes, with a certificate to prove it.
2. Is their calibration system accurate?
I guess I can only assume so.
3. How much is it used?
Really not that much, not every day anyway.
|
|
|
using it on Aircraft .......... critical area, ............could cover their backs
That's a could point Ashok.
|
That's a could point Ashok.
Doh, a GOOD point !
|
|
|
|
|
then you could check one directly against the other. It's perhaps not a method that a pedant/purist would agree with but it should be good enough for most purposes.
>>
I have checked it against a simple torque bar that I have had for a while and they are very similar, if anything it only served to make me think whether I need the wrench in the first place.
|
<< ............. it only served to make me think whether I needthe wrench in the first place.
Using a torque wrench can only do good. The torque for my alloy wheel bolts is 65 lb.ft, or whatever that is in Nm, and I stick to that rigidly. Without the torque wrench I could easily exceed the specifed figure and cause damage to either the threads or the wheel. Incidentally, you'll notice that with most open-ended and ring spanners the smaller the thread the shorter the spanner ~ this is in an attempt to stop you over-doing it. Now you've got a torque wrench, use it whenever possible or appropriate.
--
L\'escargot.
|
Using a torque wrench can only do good. ........ Now you've got a torque wrench, use it whenever possible or appropriate.
Yes I agree, TBH I am not so warried about wheel nuts at over 50lb, a torque bar does the job, however there are some fixings into alloy on motorcycles that are in the 10 to 20 lb/ft range, a torque wrench is vital.
|
|
|
|
|
|
We have our torque wrenches calibrated once a year.They are all snap on and are returned to them to service and calibrate.Have never had a problem with any snap on torque wrenches despite the use and abuse they get!.
|
|
What are the possibilities of a DIY check? Like gripping the square end in a vice and hanging a weight on the (horizontal) lever at a measured distance. Problem is to get hold of an accurate weight. If bathroom scales are used to fill a sack or box with something, you then need to know if they are accurate. I suppose you could pop in to see the nurse with your sack and try it on her scales that use moving weights. Ah well, just a thought.
|
It had never occured to me that an ordinary DIY torque wrench would need calibrating. I do unwind mine to zero after use however, as advised in the instructiuons, so as to avoid permanently stretching the spring.
It would be useful to know which way they tend to drift.
When they claim to recalibrate it, do they mean simply giving you a correction chart, or resetting it so that the scale reads correctly?
I've often noticed using a torque wrench that the first reading is not as tight as the second. ie I do up the nut until it clicks, but on relaxing the torque it will then click again but a bit tighter. I'd have thought that torque was an absolute quantity, not dependent on how the wrench was tightened. Or is that just a feature of a cheap wrench?
|
|
I have two fairly cheap Draper ones, a short low range one and a longer higher range one. They both have adjustment screws for changing the calibration, complete with sticker saying do not twiddle. I have not had them recalibrated though. The instructions are to reset to zero after use. If yours does not have any left over screws (the ones on mine are slotted brass screws flush with the surface) then it presumably isn't adjustable as to the calibration.
|
I have two torque wrenches. One is American made about 25 years old and starts from 30lbs upwards.
I needed another for the Vectra which needs very low torque readings for the sump plug and the oil filter and bought a far Eastern one off a market stall which although looks very similar and is the same length, starts at a very low reading and goes up to high.
A few weeks ago I was replacing the wheel nuts after checking the brakes and was using the latter.
The torque for this is 81ft lbs and to get the click it really needed some welly.
Tried the Yank job and it was much less effort.
Then did some testing them against each other in a really technical manner.
Putting them together in a vice and setting them to equal torque and using both arms straight and using body weight only to exert the same pressure.
To ensure I was doing it in an absolutely scientific manner I changed arms regularly.
There was substantial difference between the two which increased with a higher torque.
Doubt if this test would conform to Engineering standards (or any other) but as a rough guide it told me that one of them was out.
Which do I trust?..The American one as in my work experience they do make fine tools (Rigid tools come to mind) and 81ft lbs felt right.
I think the mistake I made was buying a torque spanner which does both low and high readings..
|
|
|
|
>>I'd have thought that torque was an absolute quantity...
If you connected your torque wrench to a torque meter, you would find that it is an absolute quantity - force multiplied by distance - there's nothing more to it.
But in your example, Cliff, you have to consider the whole system, i.e., bolt head friction washer friction, bolt shank torsion, nut thread friction, the potential for gasket settling or any foreign material in the joint to embed - in short, there are lots of variables in a bolted joint, lots of possible ways for the joint to relax a little after initial tightening.
If you tighten a bolt quickly, the friction behaviour of the bolt is different to when you do it slowly. Although the bolted joint relies on friction to remain tight, this friction also reduces how effectively you convert torque into bolt stretch.
Measuring torque is actually a poor way to tell if a bolted joint is well assembled, because the torque is an intermediate quantity - a by-product - you really want to ensure that the bolt is properly stretched. Torque plus angle is a better way to achieve a well defined bolt pre-load or stretch. Even better is to measure the extension of the bolt directly using a micrometer, or, if the faces are polished, via ultrasonic measurement - these methods are used in more critical applications. For large bolts on safety critical structures like reactor pressure vessels, an even better technique, is to preload the bolt via a seperate hydraulic ram, and then run the nut up to a small snug torque.
In the railway application I mentioned above, during tightening, the bolt was only stretched by 25 microns. So, 25 microns of dirt in the joint, or settling, or embedding, and the joint lost all of the pre-load. This was happening in a number of cases, which was exposing the bolt to a higher stress range.
Oddly enough, it wasn't until I got to university that I finally learnt how bolts work. As a subject, it isn't properly covered on the car mechanics courses, or on the BTEC engineering course I did prior to uni. All mechanics are taught is how to use a torque wrench, the mechanics of bolted joints isn't taught - and despite their ubiquity, the way bolted joints work isn't immediately obvious either.
Here's the crux of the theory (which is borne out in practice) that isnt obvious about bolted joint design - in a well made joint, the bolt may only bear 15 - 20% of the applied load across the joint - i.e., a well made bolted joint protects the bolt from the full load of the joint.
Number_Cruncher
|
I've used an old spring balance to check mine out. e.g. 50 ibs/ft 25lbs on the spring balance 2 foot along the wrench handle with the square drive in a vice. It's supposed to be more accurate when the force is applied at the end of the handle. Close enough for my purposes.
Martin
|
|
|
|
|
|
