Broken shaft (Tie bar die cast machine)

[tryeasy]

Dear members,
One tie bar of the 500 Ton die cast machine broke. The DC machine has four tie bars.
The fracture shows as initial fracture (+/-1.5" deep) as fatigue crack then continue the fast catastrophic fracture. See attached pictures. Dimensions of the tie bar: 136” length and 5.25” diameter.
Can the other three tie bars break any time?
Can the shaft be welded?
Does the broken shaft need additional test to determine presence of fatigue crack before weld?
Can penetrant test, magnetic particles or other test detect fatigue cracks?

Thank you for your time and cooperation.

 

[EdStainless]

If this was my machine....
I would have a new one made, with special attention to the threads. I would not try to repair one of these.
And I would remove the others and polish the threads and inspect them (penetrant).

How often are they checked for tension? Are they ever removed and inspected? How many cycles were on this one?
If you don't keep them tensioned then you could be cracking all of them.


= = = = = = = = = = = = = = = = = = = =
Plymouth Tube

 

[metengr]

If you are going to inspect the tie bars with this length and cross section I would suggest ultrasonic examination with an L-wave followed by surface inspection of the thread roots using wet fluroescent MT.

I would not recommend any weld repair. If any cracks are found, replace them.

 

[mrfailure]

Remember, components which fail under fatigue spend most of their life accumulating fatigue cycles. Any meaningful inspection plan that will catch a crack after it forms but before it has propagated the 20% - 25% needed before final fracture (tough to do). I am assuming you replaced the broken tie bar; if not, the higher loads on those remaining will accelerate their fractures.

You should try to understand the conditions which caused the shaft to fail. For example, did it see high vibrational loading? Is the steady state stress higher than anticipated? Did the shaft fail from normal conditions or did it see unusual loading? Any issues observed on the shaft surface at the origin such as fretting or overheat?

 

[tryeasy]

EdStainless and Metengr
Agree, I will not suggest repair by welding.
I will suggest removing the other and inspecting by penetrant or wet magnetic particles or ultrasound.
They check tension only when the castings show up flash or the DC machine locked up.
Unfortunately maintenance don't track the cycles. But they will start doing that.

Thanks a lot Ed and Met

 

[patprimmer]

I have no direct experience with die casting machines, but I have a lot of experience with injection moulding machines and in this regard the two seem very similar.

I would never consider welding the bar at the root of a thread which is where it seems to have cracked. Good quality machines have a threaded section bigger than the rest of the shaft or at least a section turned down to just below thread root diameter so the stress and fatigue is not all concentrated at the root of the first thread.

If I replaced all four I would try to use bars with a reduced diameter section but taking into consideration the size of shaft required for the bearings on the moving half platen.

It is important that all four tie bars stretch the same under load to prevent one side flashing and one bar breaking. This depends on:-

1) All bars being the same grade of steel or at least having the same modulus.
2) If anything is done to repair one bar that will change the effective stress strain characteristics it should be done to the other three. This pretty much makes repairs imprudent.
3) The projected area of the mould and runner being centralised so all tie bars take the same load. If that cannot be done due to part design, then a bigger safety margin of tonnes clam per unit of projected area is required.




Regards
Pat
See FAQ731-376 for tips on use of eng-tips by professional engineers &

http://eng-tips.com/market.cfm

for site rules

 

[tryeasy]

Members,
I will post the final decision about the broken and the other three shafts (upper management and machine maker). They have all applicable eng-tips members suggestions.

Again thank a lot to Ed, Met, Mr and Pat.

Tryeasy