Minimum Motor Flange Tightening Torque

Hi,

There are a few calculation packages out there for calculating bolt pre-load and a few companies which can assist you. The calculation packages however are usually linked to ASME VIII, Division 2, Appendix 3 - Mandatory Rules for Bolted Flange Connections.

You might find that whe you do your hand calculations that the load is low. This might have something to do with the size or material of bolts you are using.

Lubrication is an important factor whenever tightening a joint. The inclusion of lubrication (and a known CoF value) means that you can get a more precise bolt load and ultimately have more confidence in the joint tightening correctly.

I have done tests on standard carbon steel bolts and galvanized bolts without the inclusion of any lubricant and the variation in bolt loads is quite vast at times using the same torque.

I know of a lot of cases on platforms where untrained bolting techs use the 'hammer and spanner' approach to tightening flanges and take it upon themselves to decide whether or not their arm muscles are calibrated. A lot of the time a leak can appear from an over-torqued joint as opposed to an under-torqued one. Therefore dont be shocked if you have a low value for your pre-load and torque but DO get it looked over by another person or evaluated by an independent company first before proceeding.

Kind Regards

Kev Brown

I don't think that you'll have to worry about leaks on this particular joint ;-)

However, Kev's underlying point is very valid: Please don't assume that once you've correctly calculated the proper preload and have then worked some magical voo-doo (to try to conjure just what the actual K factor might be in real life) the end result will be fine. It just may result in a bit of soft foot, excessive vibration, bearing blow out and eventually, process shut-down.

"Torquing" without verifying the end result is a dangerous practice. Verification is done by measuring the actual stretch of the fastener after load has been applied. Thus, regardless of the as-is K values, you can then "tune" the fasteners accordingly.

Granted, this might be a bit excessive for this application. A nifty way would be to use load-indicating fasteners as simple go, no-go devices. These bolts change colour upon reaching the specified preload. Thus, it's easy for someone to quickly see (for example) that although the bolt has been "properly torqued", it's still loose (!). This often happens on base bolts that have bottomed-out in their respective holes before they're "tightened".

Thanks fro the comments BoltIntegrity and Kev Brown. This is the first time I have used this source and I quite like it.

My problem is not so much related with the mechanincs of the bolt tightening. It's a bit more basic than that. What I am looking for is the determining the relationship between torque, normal force, and coefficient of friction. Simplified, suppose I have a hollow tube of given OD and ID standing on a metal surface. Given a torque applied axially to the tube, how much normal force would be required to prevent relative rotation between the surface and the tube. Once I confirm this I have an understanding of how to relate bolt tightening torque to the required downward force or preload.

Hmm... interesting exercise. I assume that the intial movement of the tube (eg, when it starts to topple over) would depend on the area of the tube's end, the friction between the tube and metal surface, the degree of applied force, the length of the moment arm, the point on the other end of the tube upon where the force is applied and, the compressibility of the tube. I'll have to chew on this one. No promises though... :-(

The motion I am interested in is rotation around the tube's axis; I would not be applying a force to topple the tube. Rather I will be appling a torque to it as if it were a screw.

Codey here, not logged on.

Assuming the wall thickness of the tube is small compared with the radius, if the coeff of friction = 1, force required = torque/radius. In practice coeff of friction likely to be ~ 0.5 so force ~ 2 x above,

For your problem of motor flange bolting, I doubt it's much of an issue. Even if the bolts were loose enough for the flanges to slip axially, they would stop when the bolt shanks meet the sides of the holes. For a typical motor, the total bolt cross-sectional area is of the same order as that of the motor shaft. Since the bolt circle radius is much greater than the shaft radius (or the shaft's torsional radius of gyration, to be more precise), the bolt shear stress is much smaller than shaft stress. If you tighten the bolts to recommended torque you won't go far wrong. Use a torque wrench if you have one, but I don't think it's that critical.

Cheers.........Codey

I think your calculations are probably correct: the clamping force is likely to be high enough to restrain motion far before the bolt is torqued to its ordinary value. However, in practice, bolts are torqued to the values you can find in a bolt torque table, to ensure adequate friction to keep them from loosening as a result of vibration and shock loadings.

That's what I had thought. Your description suggested otherwise ;-)