Stainless steel bolt head breakage 1

[ecto1000]

Hello everyone, I have a problem to submit to the forum.
It is a coupling of the driving shaft with driven shaft, made by 2-bolt coupling at 90°, shafts in Aisi 304.
The drive shaft is solid and is inserted into the driven shaft (60 mm external diameter tube), which has flattened holes to provide a flat support for the bolt washers.
Two bolts are then inserted at 90° to each other, each consisting of:
M12 x 90 screw (shortened in the workshop to 78 mm) in Aisi 304 class A2-70, flat washer under the head, flat washer at the other end and self-locking nut.
The system works for 8 hours, with approximately 10 starts and stops. The speed is approximately 70 rpm.
The tightening of the bolts is normally poorly controlled and should be around 100 Nm.
It is clear that ideally instead of bolts two elastic pins should be applied, so to get closer to the ideal case the two bolts should not be pre-loaded.
Let's get to breakage: about once a year the first bolt, the one closest to the end of the driven shaft, has a head brekage.
Could it be rotating bending due to the lack of straightness of the shafts, or a theoretical error or a problem with the bolt material (this last hypothesis should be set aside given the repeated breakages even with different screws)?
Thank you all.

 

[dvd]

Too many unknowns - what is this shaft driving, what is it driven by, any shock loads, thermal expansion, is that 10 starts and 8 hours per year, day, etc., what does it look like (pictures and/or sketch),

 

[ecto1000]

The shaft belongs to a screw conveyor, temperature is constant, no shock loads at all, and yes it works 8h day with 10 starts inside this period, all working days of the week. I attach an example.

 

[desertfox]

Hi

Can we have some pictures of the failed parts, very often by looking at the broken parts one can tell the failure mechanism.

“Do not worry about your problems with mathematics, I assure you mine are far greater.” Albert Einstein

 

[TugboatEng]

A2-70 is the lowest strength bolt available. You could significantly improve the strength of the connection by utilizing an A2-80 o A4-80 bolt with fine threads. Fine threads significantly increase the minor diameter of the bolt.

Utilizing a body fitted bolt will provide the highest strength.

 

[dvd]

What material are you conveying?

Is this a hex head cap screw, or some other style fastener?

 

[SWComposites]

Is the head breaking at the base of the head at the fillet?
Is the washer ID larger than the bolt OD?
Please post a photo of a broken bolt.
Maybe is solution is to a) torque the bolts properly, and b) replace the bolts every 6 months.

 

[ecto1000]

It's wheat flour.
Bolts are hex head cap.
I attach a photo of the bolt body still installed, the washer was still there.
The head was lost.
In my idea I was thinking to avoid pre-loading the screw, since it must work as an elastic pin.

 

[dvd]

Curious if you may have a stress corrosion cracking issue due to chlorine in your process. It does not seem like the head should be under much fluctuating stress unless this is seriously misaligned.

 

[EdStainless]

Every time that it starts is a shock load.
Are both shear zones on the bolt smooth material (threads only external to the shaft)?
You need to look into washers with a curved inside face so that the bolts actually bear on the shaft.
And the washers need to be harder than the other parts so that they don't yield.
As it is not the bolt heads are bending back and forth.
You need cold worked bolts, or at least go to alloy 2205 bolts.
Have you measured the radius under the heads?
I think that the bearing is the real problem.

= = = = = = = = = = = = = = = = = = = =
P.E. Metallurgy, consulting work welcomed

 

[SWComposites]

Suggest switching to a fatigue rated bolt, with either external hex head or 12 point drive head. Cap heads with internal hex drives don’t have much material at the bottom of the socket which is close to the fillet between head and shank.

Are the holes on the shaft close in tolerance to the bolt shank diameter?

 

[mfgenggear]

Terrible design, complete rebuilt required.
The real question. Failure analysis required.
Whats causing the bolts to fail. Is it better the bolts fail instead of the entire shaft failing.
Causing catastrophic failure.

Some causes, out of alignment. Buckling, twisting, out of round, excessive runout between members.
Excessive shock loads, not enough stiffness, strength, wall thickness to thin.

A different joint design. Maybe required.

 

[TugboatEng]

The consequences from failure seem minimal. This is a case where GUIDED experimentation may provide a better result at lower cost.

 

[ecto1000]

Thanks to all for the great opinions.

Should not be chlorine since we are in the food sector.
Yeah ok, actually they have a soft start or inverter.
Yes threads are only external to the shaft.
Washers doesn't need to be curved since the holes are flattened.
Radius under the head yes should be analyzed, but in other equal applications no breakages at all.
By bad, actually the screw is with external hex , not internal. I made confusion.
There is approximately 0,1 mm between shank diameter of the screw and the hole.
I don't like the design too, but it must work since it is the only coupling recommended by CEMA (bible of screw conveyors). Even if the there are better couplings of course.
Difficult to analyze since the machine is too far away at the moment, unfortunately.
Yeah not big consequences, but the broken head (they say) ruined the processing machine after. Even if there should be a safety screening on top of this processing machine.

 

[Stress_Eng]

If you draw a free body diagram for the bolt, you will include a contact bearing pressure under opposite sides of the bolt head and nut, to react the moment created by the bolt to hole bearing contact loads created by tube torque. This, under the bolt head, will result in kt factored tension and bending stresses at the head fillet radius. You may need to investigate change in bolt material, size, and / or the introduction of more bolts to the tube joint. The fatigue loading could also include additional cycles as I would assume the loading from the flour is not constant.

 

[ecto1000]

Yes, I know that.
Adding more bolts is not an options since two are enough for this application, following CEMA too.
Theoretically torque is constant, but as you said investigation about material and actual application is mandatory.
But I keep thinking about the possibility of avoid tightening too much the bolts, for example 20 Nm just to bring the self-locking nuts in position. In this way the bolts should work in pure shear as an elastic pin.

 

[mfgenggear]

Ectopic
What about using shoulder bolt

 

[ecto1000]

Yes I considered it. But on other machines no issues with normal ones.
In that specific application there is something for sure not yet understood or discovered....

 

[dvd]

Should not be chlorine since we are in the food sector.

Processing bleached flour? Possibility to try a ferritic stainless steel fastener?

 

[3DDave]

There should be nearly no operating loads on the bolt head if it is used as a shear connection as the head of the bolt is separated from the shear plane.

I'm more interested by the washer that stayed in place.

Probably some part of this bolt installation holes have been damaged so just putting another bolt in place subjects it to the same load case.