linear measurement for reciprocating action

[BrianGar]

Hello guys,
A bit of a broad problem, and also a broad question at the minute, but I will refine as I go. I just need to know if this is a viable way to do this at first.

The problem in hand,

Measuring the reciprocating motion of a valve over its full cycle.

Test jig goes like this, a valve, follower, cam, spring, and housing is setup as a jig.

A spring returns the poppet valve to its seat position.

As the rpm increases, at some point the valve will start to bounce. Or not make it back to its seat in time for the next cycle.

I want to measure this. Is this possible?

Thoughts at first,

Attach a low mass linear electronic measuring device to the head of valve, and fix a pickup ring at close location?
Does this sort of sender exist?
Is the resolution good enough?
At the minute, I have 200 cycles per second in mind over say a 15mm distance.

What is the common way to do this?

Perhaps I could go with electrical contact on the seat(not sure this could react quick enough)


What are peoples thoughts?

High speed cameras are big bucks...

Im not really concerned as yet about anything other than if its returning to the seat, or_not at a particular rpm.

BG

 

[MikeHalloran]

If you can produce magnetic discontinuities in the valve stem, you can use something like this:

http://174.129.32.184/patent/US3956973

Mike Halloran
Pembroke Pines, FL, USA

 

[BrianGar]

Thats interesting Mike, and something along the lines of what Im after. Light too.

Will look further,

Brian,

 

[byrdj]

How about a proximity probe?
No contacting, but must be able to mount close to a good surface on the valve

 

[BrianGar]

byrdj,

Im guessing you can hook one of those to an oscilloscope?

Looking on google there, I presume they can work off two ref points bearing in mind this is a sliding action, and not rotary?

Valve material is Ti If that makes any odds to pickup material.

Thanks a lot,

BG

 

[byrdj]

O scope works good.

The material and size of the "target" on the valve would possibly rule out prox probes, especially if you want the whole 15mm travel for the units I've used. However there might be some special purpose probes available

the operation is the probe measures the gap between the fixed probe and the moving target

 

[Compositepro]

It sounds like you need a fiber-optic interferometer. I know they exist but don't know how much it would cost for the range you need.
You would simply need a mirrored surface on the end of the valve stem.

LVDT technology might work as well.

 

[VEBill]

Also, review this previous thread:

http://www.eng-tips.com/viewthread.cfm?qid=289831

 

[ScottyUK]

Speak to Balluff's applications guys. I think the bandwidth is your biggest problem, whether laser or inductive, but there may well be something suitable in their range. This one looks like a 'possible' with a 500Hz -3dB bandwidth.


----------------------------------

If we learn from our mistakes I'm getting a great education!

 

[VEBill]

If the bandwidth is too high and the motion repeats, then one could sample slower than the cycle time (e.g. like a strobe light).

 

[BrianGar]

Thanks folks, as I start searching the aboves, its giving me a lot to think about. No doubt Ill be back with more questions, but for the minute, its all food for thought and taken on board,

regards,

BG

 

[MacGyverS2000]

'Twere it me, I would start out with Compositepro's method... that was my first thought, too.

Dan - Owner

http://www.Hi-TecDesigns.com

 

[IRstuff]

My post from last night vanished
Sick makes a variety of laser range measurement units. This one has 10kHz rep rate and micron level accuracy/repeatability:

https://www.mysick.com/partnerPortal/ProductCatalog/DataSheet.aspx?ProductID=40643

TTFN

FAQ731-376
Chinese prisoner wins Nobel Peace Prize

 

[Skogsgurra]

Can you use a strobe light and just look at it? Like VE1BLL (almost) said.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[BrianGar]

Ummmmmmm thats a good point actually... Forgive my ignorance but isnt that simply all I need? Its early days, so buying expensive equipment is a touch out of bounds at this point.

Can you point me towards a suitable light?

Many thanks,

Brian,

 

[Skogsgurra]

It depends a lot where you are. I am in Sweden and know what to buy here. In fact, I use a very old Shimpo strobe light. Powerful and reliable. They have this solid state strobe

http://www.shimpoinst.com/dt_326.php

which goes to 120 000 rpm - or 2000 Hz. That is what I would look for.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[VEBill]

(almost)

A strobe light was the very first thing I had explicitly suggested in the similar previous thread that I had explicitly referenced above. !!

 

[jimkirk]

Just to simplify things, if all you're interested in is at what RPM the valve begins to bounce, perhaps a microphone or accelerometer and a scope with FFT capability or even a bandpass filter can detect the change in audible harmonics associated with the valve bouncing.

 

[Skogsgurra]

Sorry, Jeff. Didn't read the thread.

Gunnar Englund

http://www.gke.org

--------------------------------------
Half full - Half empty? I don't mind. It's what in it that counts.

 

[controlsdude]

I believe sick has an off the shelf linear type magnetic sensor that gives you an analog position out. I saw it in a brouchure so its not vaporware. Never used it so dont know if it fits your needs. I believe its used on cylinders to give a position analog back to the actual position not just a single point like on most cylinders.

Since its analog not sure if this would be fast enough.