How does water in threads affect tightening torque?

[mvarnell]

I know that thread lubricants and contaminants cause the fastener torque requirement to be lower. How does water affect torque though? We have implemented torque equipment on a hydraulic motor assembly where we have 21 3/4-10 bolts to mate the plates together. The drawing is calling for 56 kg-m on the fasteners. We have the equipment set for 56 kg-m and we keep breaking fasteners. The plates are shipped with a light coating of oil on them to avoid rust, and we are pressure-washing them off before assembly. There is definately water and possibly light oil in the threads during assembly. I am assuming this is the problem, so I am about to start testing this theory. Just wondering if anyone has run into this before, and what was done. The bolts have been checked and are okay.

 

[NickE]

I think that you are going to have to remove all of the contaminates from the threads and holes. You might try a water displacing solvent that evaporates. Then you might want to re-evaluate your torque specs and modify them to include an anti-sieze to stabilize the torque requirement.

nick

 

[Screwman]

Talk about mixing units! Inch fasteners with a KgM "torque" callout. That works out to 4860 inch pounds if we want to keep our units Imperial, so you shouldn't be breaking bolts if they are grade 8.
I would question how the installers are measuring that torque...there are not many wrenches calibrating in KgM.

If the failures are a result of excessive lubricity, it is most likely coming from residual oil. At the contact stresses that you achieve on the pressure flanks at yield, water will not provide much lubrication.

Dick

 

[mvarnell]

We know the torque measurements are correct, we are using a transducerized Aimco dynatorque nutrunner that has been checked numerous times and calibrated. It gives a torque readout and elapsed time of torquing, and it will also do angle if we set it up to do so. Reason for the mixing of units is American company, but Japanese parent company that does many of the drawings. Units get confusing at times. Thanks for the help.

 

[btrueblood]

Are you running the fasteners into blind holes?

 

[mvarnell]

I guess you would call it a blind hole. All the bolts are placed into the front plate (hole), then the motor assembly with the threaded holes is lifted into place and all of the 21 bolts are started by hand, tightened with a small impact, and then brought to final torque with the transducerized nutrunner.

 

[Screwman]

Do you know what the target preload is on the fasteners?

Interesting that you can get angle on the drive guns, that may give us another way to get to your target preload. How thick is the clamped component and is there a gasket in the joint? It may be possible to come up with a rough Torque / Angle stratgey via calculation.

I questioned the gun calibration because it is very often the source of the problem and we spend huge amounts of time chasing problems that cease to exist once the units are properly calibrated.

Dick

 

[MintJulep]

Presure washing how, and with what?

Plain water, soap, detergant, solvant?

Are you blasting each fastener individually, or tossing a bunch in a bucket and blasting away?

It seems quite possible that you are creating some sort of emulsion of soap and oil.

What is your tightening sequence?

 

[arto]

If your bolts get rusty, it's gonna take a lot more torque! ;-)
Perhaps it would be a good idea to wash & dry the parts before assembly? then paint or preserve afterwards.

Either that, or RTV seal the thread/blind hole leak paths be for blasting them full of water

 

[mvarnell]

I would have to look up preload, not sure at the moment. The plate thickness is around 25mm for the thru hole and the hydraulic motor casing that is threaded is around 150mm. There is no gasket in the joint.

The fasteners are not pressure washed, only the threaded motor housing is washed. It is straight water that they are using just to blast some of the oil coating off. The tightening sequence is the 4 corners, and then all the rest in a line from one end to the other. The bolts are arrayed in a c-shape.

The weird thing is that we have two different motors and casings that we use. This is the larger set of the two. The other uses slightly smaller bolts and is torqued to 28kgm and we never have any problems.

 

[Screwman]

The tightening pattern might be causing you a problem. You may be getting cross-talk between the first bolts tightened and the remaining ones. This can cause the clamp load to go much higher than you would expect because of the joint interactions. You might want to try snugging all of the bolts and then working from the center of the "C" out to the tips, alternating sides.
If these are blind holes, are you sure that you have enough thread depth? If the tapped threads are running out before the part is fully seated you will be putting all ofthe torque onto the shank of the bolt and you may be getting a torsional failure before the parts see a significant amount of clamp load.
Is there any pattern to what location the broken bolts occur? If you are seeing them in just some of the locations that will tell us a lot about possible causes.

 

[mvarnell]

All of the bolt breaks have been in random places. Some at the beginning, some at the end. Some of them they don't let break, they back them out and they definately had already entered plastic deformation. All the ones that break are in the same place, right where the threads start in the middle of the shank, which is normal for fractured bolts. I am still investigating residual oil left in the threads as a big possibility. Thanks for all the help.

 

[BillPSU]

I have seen localized loading of fasteners when support members do not evenly spread loads across the surface being mounted. I saw this situation in the mounting of a rotation bearing on a mobile crane. The structure of the turntable did not transmit the loads equally to all bolts when the crane was lifing. Is there some sort of structure below the 25mm plate which could cause localized loading? You mention that the breakage is random across the bolt pattern. Are the positions farther from the centerline of the motor than the rest.

Solutions are varied. Increase thickness of 25mm plate to increase stiffness. Change fastener to higher strength such as Bowmalloy bolts. Add more fasteners at the high stress points. Go to a larger fastener. Redesign mounting structure to eliminate localized loading if enough strength is supplied by the fastener.

I would also recommend using the correct units when torquing either Newton meters or foot pounds. Kilograms are a unit of mass not force.

 

[mvarnell]

kilograms are a unit of force (kgf) and a unit of mass(kg). kgm used as a torque value is extremely common in a Japanese company. A redesign of the motor and mounting plate is not an option. Changing fasteners is also most likely not an option. I will either adjust the torque spec based on the conditions, adjust the torque sequence of the bolts, or make sure the threads are clean and dry before assembly. I am in manufacturing engineering, not design engineering, so making a major change is something out of my control in this situation.

 

[Screwman]

I agree about the units. When I first saw the post I figured that we were dealing with an Asian company. Drives me crazy,I wish they would go along with the rest of the world. Oh well, we can live with it.
Is there any caoting on the bearing surface that the bolts are seating on? We just solved a sporadic breakage issue (<1%) that turned out to be caused by contamination of bearing surface with an aluminum bearing undercoat. It increased the lubricity enough to cause breakage during installation.

 

[btrueblood]

If you run fasteners into blind holes, and aren't sure if all the water is blown out, the water will be pushed to the hole bottom. Pressure will build when the bolt tries to compress the residual fluid in the hole bottom. Hand torquing is slow enough that the water will slowly squeeze out, and torque/preload will be satisfactory. But running the bolts/screws in with an impact driver or similar could result in screws not driving in fully, or breaking if torque is increased to get the screws to fully engage. Try giving your crew a needle end for their air blow-off gun (like the needle used for filling basketballs/socker/ footballs), to ensure the hole is completely clear of water before installing screws.

 

[BillPSU]

mvarnell
I also am a manufacturing engineer but I do not have blind faith in a design until I know it works. Check what grade of bolt you are supposed to be using. Verify the bolts you are using meet that specification. There has been many cases where inadequate bolts (usually foreign) are substituted. Your purchasing department may not even know to buy the correct bolts. Your Quality department should already have that answer but if not have the bolt supplier certify the bolts. Changing a torque spec is a design engineering change as you are changing the clamp load on the motor.

A blind tapped hole has a lot of room for the chamfer for the tap and unless the hole is completely full of water I don't think the water will matter. The thread has clearances between the pitch diameter and also clearance in the crest and root of the threads. You could be bottoming out the fastener in the hole but even then the torque should not be enough to twist the bolt off. I think you have the wrong grade of bolt.