compressor failure investigation

Maybe it got sufficient oil in it to lock it up...? Oil doesn't compress too well.
I bent a con rod in a car engine once by driving through a ford a lttle too quick .

Del

Typically, liquid carryover would cause the cylinder head to come off, or as you say would damage pistons or break piston rods... I can't tell if this is head end or crank end, but typically will a large liquid carryover to the crank end the piston rod will break right at the piston.

This case looks to me like the valve cover was simply not torqued down correctly, or had missing or improperly installed gaskets.

The other possibility is that if this is a cage retainer design, then the valve seat may have worn out allowing the valve to move back and forth with the gas. This essentially results in a very high impact stress on the cage and the valve cover eventually resulting in broken bolts as shown.

How did the valves look? Were they plugged up? Were the seat washers present? Was the hammering damage to the seat washers? How did the cages look? Cover gaskets?

Welcome back. Good to see your posts again.

Thank you for your comment.

I hope my picture clarify your questions.

The valve looks ok and there is no gasket for the cover; its "O" ring. In addition to my prior information, all valve cover bolts were found shredded except tow were found elongated.

How about the seat gasket ring?

The valve on a Dresser machine sits on a soft metal washer (often bronze) in the seat. If that washer is not there or if the valve bolts are not torqued properly, then that washer gets extruded or "coined." As the washer extrudes, then the valve has room to move and beats out the valve cover bolts..While I have seen a number of Thomasson machines, I have never pulled valves on one.

The other likely possibility is that whoever installed the covers either under torqued or over torqued the valve cover bolts. These should really be installed with a torque wrench, but it is a rare occasion when maintenance crews do so. If under-torqued the valve was loose, if over torqued the valve bolts may have yielded.

Nice observation Bob! And a good answer for you!

Do you have someone with a wrench who thinks it is their job to make sure all bolts are super tight ? I had a guy like that once in a factory. We finally figured out why so much stuff came apart under normal use. He loved to test his muscles, I think, on nuts and bolts and caused a lot of damage in only 6 months he worked there.

If that is the case, it should be apparent in the stretching of the remaining threaded studs on the other covers.

Actually this failure happened 5 days after opening that particular valve on 4^th stage to inspect the compressor for corrosion.

Further analysis of the bolts / studs will possibly determine if they were yielded before being put back in service after your inspection. (yielding = over torque) A specialist can look for plastic deformation followed by rapid failure ? (I think).

Alternatively, a fatigue type failure could have been occuring slowly before your inspection - but I am guessing this is less likely.

I'm going to preface my comments by stating openly that I am not familiar with this specific machine.

When a car engine head blows it is almost always due to pre-ignition causing excessive heat and pressure build up and leading to a gasket failure which causes coolant sucking on intake a catastrophic bolt fracture failure on next compression.

I don't believe gas pressure alone could cause bolt fracture failure like that shown. Excessive gas pressure should cause bolt stretching and then gasket failure. The explosive destruction failure shown would have to be the result of hydraulic force generated failure followed by explosive decompression failure.

I would look for signs of a 4th stage inlet valve failure causing a pressure build up that leads to a gasket leak oil sucking fault that then leads to a catastrophic hydraulic/pneumatic failure.

If that scenario is unlikely which is probably the case as you say the intake looks OK, then consider a diesel style auto ignition detonation caused by the spontaneous or even spark induced ignition of an fuel air mixture. If this scenario seems more likely carefully inspect the oil injection orifice and or needle valve to see if this could be the source of a fuel injection mixture instead of an lubrication injection mixture.

You mention that the explosion happened after an inspection for corrosion. This does not sound like a routine MOR procedure to me. If is routine I would still be suspicious. Routine maintenance is overlooked everywhere, even in a well run refinery. My suspicion would not involve evil doing. Rather, the inspection would seem to me to indicate that the compressor in question was giving signs of problems that someone perceptively noticed but was, unfortunately, not able to correctly diagnose and fix.

Given the circumstances that followed, an explosive failure, it would unlike normal human behavior to step up and say, "Yes, I knew something was wrong, so I took it apart to see if I could find the problem, but I couldn't, so I put it back together and ran it till it blew up." That conversation, ain't gonna never happen.

However, if you hand out some doughnuts along with reassurances that no one is gonna get fired and ask some questions tactfully about what was going on with this piece of equipment you might find out some good "deep cover" background information on what was going on with it before it failed.

Good luck in your investigations,

One other thought ... it appears that only your bolts / studs saw a failure. If I understand the original post correctly, the piston rod and other bits survived. This does suggest (as others who responded said) that over-torque could be the root cause. Once the head bolts / studs were weakened they began to fail rapidly. However, I would have expected a badly leaking gasket which would make lots of noise and lower peak cylinder pressure (relieving stress) - but not total destruction of most of the fasteners.

Inspect the bottom of the cover. If the retaining bolts\nuts were over torqued, the bottom of the cover should show signs of being deformed. The head appears to be aluminum. It will deform before the steel components.

I doubt it is aluminum Bob, painted silver probably. That is a solid steel cover I am sure.

It is necessary to have some more input for a really thorough analysis and a correct explanation:

- surface of the area under pressure

- bolts dimensions (thread)

- which studs where elongated and which were shredded

My supposition is that the studs were not tightened in an uniform way and some of them have been more loaded. The load went over the limits and lead to destruction.

Such an analysis requires more than qualitative estimations due to the complexity of loading of the studs pattern