Dynamically monitoring a reciprocating poppet valve? 3

[whistlemeup]

I need to instrument and dynamic monitor reciprocating compressor poppet valve motion. The poppet is mushroom shaped, made of peak material (plastic), operates in an atomosphere => 250 F, and is located internal to a gas compressor operating at 1200 rpm. Compressor valves consists of multiple poppets and it is our task to determine if they all are moving uniformly up and down, or some are floating or not moving at all which would imply too many for the given flow. I have done simular work on steel plate valves using proximitors which will not work with the plastic material. There is a 1 inch long by .385 dia. coil spring inside the stem of the mushroom that aids the return of the poppet to the seat. I am currently looking at monitoring dynamic spring force using comercial load cells with the intent of relating that to distance traveled. Any one ever do a simular project and have good advice on what technique or instrumentation to utilize?

 

[VEBill]

A synchronized strobe light (or functionally equivalent high speed camera) comes to mind. Then you could literally see what's going on. All the techniques of varying the speed and relative timing offset would allow you to to examine such repeating motion in perfect detail.

The obvious problem is, these valves are internal to a compressor. So a camera inside is probably impractical.

But prototype systems can sometimes have a window installed to confirm such one-off design questions. Literally cut away part of the system and replace it with an appropriate 'glass' window.

Or, can you view the back of the valves, the spring side? Might be able to see what's going on from that side.

If you don't get a suitable idea here, then maybe try the automotive forums. Engine designers must have similar desire to track valve motion.

 

[Thealanator]

Are you designing compressor valves or troubleshooting an existing compressor?

 

[whistlemeup]

We design compressor valves and revamp old compressors adding efficientcy as well as longivity. Mainly an engineering firm. This valve is our design with a patent pending.

 

[IRstuff]

Depending on the severity of environment, there may be other sensors that could do measurement:

> acoustic distance
> laser distance
> accelerometer directly on the poppet

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[automatic2]

you could maybe try pressure differential of the compressed meadia across the valve seat. Comparing different poppet/valves in a multivalve setup would be very interesting, to see if the load was shared, or how it was distributed due to other considerations.

 

[whistlemeup]

I had considered Lazers, but have no experience along that line. Would need a high update rate and accurrate distance measurement. Accelerometers would add mass to a very light peak (plastic) poppet. Acoustic distance I have used but feel it may be too slow. Anyone have experience with Lazers as used for very small distance measurement?

 

[MacGyverS2000]

Considering lasers can measure down to (roughly) their own wavelength (not including interferometry techniques), I'd say it's well within your limitations distance-wise... whether it makes sense from an environmental standpoint is another matter.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[itsmoked]

I think I'd really want to "see" them as apposed to "detect" them.

Can you make a see thru port out of glass or some other material? It could be an inch thick. Then you would be free to use a high speed camera and actually watch the physical dynamics. This would skip you past all the "if this - then maybe that is happening".

Keith Cress
kcress -

http://www.flaminsystems.com