BOLT SHEAR STRESS AND TORQUE 2

[nschiermeyer]

Hi, i have a question.

We are mounting a hitch which has a high force (40,000+lb) applied. what we have and are trying to do again, is mount the hitch by only using bolts. (10x, 5/8" grade 8)

I am trying to find the appropriate torque to apply to the 10 bolts to hold the force of the hitch. if the friction force is not accounted for, the bolts being loaded in shear only, will fail. So, the friction force between the 2 steel contacts is important.

I need a good way to figure what the needed torque should be, or at least the force required betwen the 2 steel contacts. Any thoughts?

Thanks,
nschiermeyer@maintainer.com

 

[jetmaker]

First you need to figure out how important the consequences of failure are. If they are high... then ignore friction and change your design. If they are low, assume a percentage of the total load you wish to carry by friction. Once that is determined, determine a friction coefficient that you can achieve. From that, you get the normal force required to achieve the desired friction. Calculate the tension load on the bolts due to loading conditions, then add that to the normal force required. That's your preload. Calculate torque using an approved/reliable fomula, and voila.

What I have suggested to you probably flies in the face of most building codes in the civilized world, so check your local/state/fedral regulations. Also, if you are truely going to rely on friction for contribution to an ultimate loading condition, you need accurate data and formulas that are true to your situation. Probably you will need to do some testing. Finally, you also need to ensure that vibration effects and cyclic loading do not cause the torque to relieve. So look at dual retention methods, etc.

What I'd suggest is that you design the shear loads to be carried by dedicated shear pins, and use bolts in loose fits to handle tension loads only. If you are limited in the number of fasteners, look at using high quality shoulder bolts. Relying upon friction is a risky business... and needs to be approached with due diligence and respect.

Regards,

jetmaker

 

[Tmoose]

Various steel construction codes get very specific about calculating the numbers of bolts or rivets and spacing when designing a structure to support a load.

However..................

Chrysler chose to their 426 Hemi flywheel and flywheel with 8 non-shouldered bolts torqued to about 80 lb-ft, and no dowels or pins.

http://www.streetrodderweb.com/tech/

0307sr_transmission/

GM/Isuzu mounts the flywheel to the 400 cubic inch turbocharged Duramax diesel crankshaft with what looks like about 10 bolts. Barely any space between the bolt heads, with no obvious dowels or pins.

427 Corvettes have 6 non-shouldered bolts holding their flywheels on, with a single 7/16 or 1/2 inch dowel to ensure orientation. A specific orientation is a requirement for externally counterweighted models.

Here is how Ferrari did it on the 308

http://www.nicksforzaferrari.com/

forzaferrariwebsite1007.htm

Acura does it this way, although in the Haynes repair manual it looks like there is a single pin about half a half bolt diameter.

http://www.sounddomain.com/

shopbrand/Comptech

Air-cooled VWs and older 4 cylinder Porsches used (4) to (8) 8mm dowels with a single BIG center bolt to clamp the flyweel to the crankshaft. Here is a discussion on how well that works, from a well respected VW high performance specialist.

http://www.geneberg.com/crnkrdspg4.htm

My interpretation is that when 2 components are to be attached, and must withstand reversing and shock loads without even micromotion and fretting, friction is the way to go.
One way to provide this is with interference fits between 2 diameters. Another way this may be provided is by many big bolts torqued hard to provide large clamping forces.
Every bolt makes a contribution via clamping, when most folks consider it logical that even carefully reamed holes would leave a few dowels doing all the work.

Dis-believer will, without qualms or hesitation, loosen all their SUV's lug bolts/nuts, re-tighten to 5 lb-ft, and drive at high speed dancing on-and-off the throttel up and down steep mountain roads after pounding over a pot-holed street for 100 miles. Or, at least loosen the various pulley and sprocket bolts on their engines to allow the pins and keys to "continue" doing all the work.

 

[jetmaker]

TMoose,

Interesting articles.

Agreed that for fatigue loads, friction plays a very important part in preventing failure. However, I do not think it is prudent to trust friction effects when you are looking at ultimate loads. The reason being that at ultimate loading conditions, you have most likely overcome any friction effect unless you have extemely rough surfaces, like a clutch plate. Also, a couple loose bolt and you have a big problem.

Another thing, the bolted joints on a car/SUV are loose fit (most likely Class 3 fit), so it would be a rough ride if they were not tightened. Try that in an airplane, and it will be just as smooth, but your plane won't last many cycles....

Good discussion.

jetmaker

 

[Goosen]

I am with jetmaker. Design the hitch attachment such that the bolts *will* carry the load. Unless you have performed testing on similar components with a similar surface finish, etc, I doubt that you know enough about the coefficient of friction to depend on it for your design to succeed in terms of carrying ultimate load.

 

[nschiermeyer]

Thanks guys. I am currently re-designing the hitch. We are adding another gusset and tube, so the force won't all be acting against the bolts in shear.

Thanks again,
Nathan