Calibrating a hydraulic torque wrench for tension in threaded anchor rods
3

Question

Calibrating a hydraulic torque wrench for tension in threaded anchor rods

Ingenuity (Structural)

(OP)

14 Jan 21 05:55

I have a project with a 50 ton crusher where the anchor rods have fractured and they are being replaced. With the new anchor rods we are providing approx 18" of free length such that we can preload the rods and reduce fatigue effects.

We would usually install the new rods and stress axially using a calibrated center-hole hydraulic ram/rump apparatus to achieve the required magnitude of preload. Unfortunately for these specific anchor rods there are obstructions that do not permit the ram and stressing chair to have enough space, so we are contemplating using a hydraulic torque-wrench which requires less space above the nut.

We need to pre-calibrate the torque wrench for a given tension, similar to what is used for structural steel bolting i.e. Skidmore-Wilhelm (which we do not have and would not work in our situation).

Any opinions of the following setup:

We would have to restrain the pipe assembly and the right-hand end of the anchor rod to prevent it from spinning.

We propose to stoke the ram by an inch or so, then apply a prescribed torque through the hyd wrench, stop the wrench and record the hyd ram gauge pressure (and hence tension), increment to another torque, stop and record the ram gauge pressure, and repeat until we are close/exceed to the preload level, and repeat three (3) times to verify repeatability.

We are not looking for Swiss-watch accuracy, so within 10% of our preload would suffice.

Any comments?

3DDave (Aerospace) · Answer 1 · 2021-01-14 08:56:27

I would suggest using a high-pressure lubricant on the nut face and threads to lower the friction losses and provide a known friction state. Also, is it long enough that simply measuring the stretch would tell the tension load? Since that strain is what is responsible for the stress it should give a direct reading.

SwinnyGG (Mechanical) · Answer 2 · 2021-01-14 13:42:16

Quote (3DDave)
Since that strain is what is responsible for the stress it should give a direct reading.

I agree 100% with Dave.

He is right that the ideal way to do this would be to calibrate with a known friction condition. But, unless you apply the same condition at installation (which you may not want to do for a number of reasons) you will ultimately have to calculate the correct torque value; the change in torque required for preload X, between lubricated and not lubricated faying surfaces, will be large (much more than 10%).

If you measure the stretch of your anchors and set preload that way, your final installed preload will be both very accurate and very precise (much better than 10% is possible if you're careful).

Ingenuity (Structural) · Answer 3 · 2021-01-15 05:22:07

3DDave and SwinnyGG:

Thank you.

I do intend to lubricate the threads and bearing face for both the calibration setup and the in-service condition.

I shall use a dial indicator at the rod end to measure displacement of the 18" free length (effectively strain).

Thanks again.

kingnero (Mechanical) · Answer 4 · 2021-01-15 11:24:22

This only works if you can restrain the "bottom end of the free 18". This means a large washer or a plate or something. If you only wrap the threaded part and cast it in concrete, some of cast-in length will also deform/stretch, and you'll measure a larger elongation than accounted for (resulting in a lower effective prestress)...

3DDave (Aerospace) · Answer 5 · 2021-01-15 11:43:26

kingnero,

That's a good point. There should be a way to estimate the stress falloff into the concrete for the embedded length unless the rod is not cast in. If it's not cast in then it should scale with length.

Found this thread: https://www.eng-tips.com/viewthread.cfm?qid=221027

SwinnyGG (Mechanical) · Answer 6 · 2021-01-15 12:59:46

Alternatively, if you know your embedment length and have enough time for concrete to cure (I'd want at least 28 days for a standard mix, or perhaps 14 days for a high early mix- you'd want to use the exact mix that will be installed) it would be relatively easy to construct a test. Case your rod in place in a block of concrete, then slip over your XXX pipe with your measurement cylinder on top, then your torque wrench of top of that.

MintJulep (Mechanical) · Answer 7 · 2021-01-15 13:15:23

Why not use direct tension indicating washers?

https://en.m.wikipedia.org/wiki/Direct_tension_ind...

You could do your calibration to dial things in, and use one in the real installation to confirm.

Ron (Structural) · Answer 8 · 2021-01-17 19:47:24

What about a Skidmore-Wilhelm tension calibrator? Several varieties available as well as other similar products. Don't reinvent the wheel!

Stick. (Mechanical) · Answer 9 · 2021-01-18 16:49:22

I don't know if they'll work in your application, but have you looked at Superbolt style nuts (or similar multi-jackbolt tensioners)? Nord-lock claims accuracy within 10% of specified tension and they have the added bonus of only needing hand tools to install.

DAVIDSTECKER (Mechanical) · Answer 10 · 2021-01-18 23:05:20

I'm a big fan of Superbolt myself.
If you know what stress you want, calculate the stretch required, then put a dial indicator on the end of the bolt and tighten the nut until you get the reading you want.

Ingenuity (Structural) · Answer 11 · 2021-01-22 19:20:16

==> UPDATE 01-22-2021

I thought I would circle back and give you all an update.

We did 3 runs of torque/tension setup in the client's welder's workshop.

Each run was incremented in 5 equal steps. The results were not very repeatable, nor particularly linear. After the 3rd run the threads were galled.

I revisited my field clearances issues and made it work by applying direct tension to the anchor rods using a calibrated 50 ton center hole ram and a loading stool. Preloaded to 45 kips, and we shall return in a few days to re-check pre-load.

I did look at the other alternatives suggested (Skidmore-Wilhelm, Superbolts etc), but given the time available the above resulted in a satisfactory solution. The above setup is especially nice in that we can go back and adjust the preload, if/as required - an option that was not available with the hydraulic torque wrench due to the limited projection of threads above the nut.

I appreciate all of your input. Thank you.

DAVIDSTECKER (Mechanical) · Answer 12 · 2021-02-12 14:44:33

I've had this rat holed for years and just found it.
This is for using turns of the nut to get pre-load.

https://res.cloudinary.com/engineering-com/image/upload/v1613140985/tips/Torque-Turn_Tactics_qbhok2.pdf

Come Join Us!

Posting Guidelines

Calibrating a hydraulic torque wrench for tension in threaded anchor rods
3