high temperature glass rods

Hi,

Are you sure that you need the optical elements to take that temperature?

If the housing of this engine doesn't get too warm, couldn't you sink a hole in its wall and then bolt down a suitable window on a ductile metal or silicone gasket? Sensor stays nice and cool (rtp) on the lab side of the window, looking through at the happy gas within. Take care with signal noise (shileded cable) and pick nice amplifiers, and all should be well.

Also, what process are you trying to monitor and is optical the best way to do it?

It might be, but depending on the wavelength you're monitoring, sensors and windows may be v. expensive. I use the good people at Crystran for all my window needs (200nm and Lyman-alpha UV in vacuum systems) and would be surprised if you need an optical rod of the type you describe.

-James Garry (aka Dr easily-surprised)

You might like to try Plowden and Thompson,

who may be able to help, although for that temperature you need something very special, like fused silica or an alumino-silicate.

Like the other responder suggested, a water-cooled window may be better.

In 1958, when I was but a mere freshman at the univirsity, the mechanical engineering department had made some photographs or pictoral representations of the engine combustion front process. I have forgotten the details, but I am not certain that this was anything like cutting edge research at the time.

It seems to me that this is likely to be a fairly old technology, and need not be reinvented.

I think of things like "depth of field", focus and so on as obsticals to getting meaningful results from peering into the combustion chamber through a small rod of transparent material (not all transparent rod materials are glass rods).

I wish you luck in this project.

Indeed. I have seen similar from comparable dates. When mucking about with car engines a decade or two ago I (several times) nearly bought a widget that could be screwed into a spark plug hole to allow the timing to be tweaked so as to achieve the optimal combustion colour. (pauses) Gunson, that's the company. (googles - yep - the colourtune)

Of course, if his 'engine' is somewhat more unusual (might be an external combustion rig - fun!) then a modified gizmo (sunken hole gasket, stout window, clamp) would suffice, and certainly wouldn't need long expensive fragile rods. More info from phillippanicker would be handy (like propellant nature, engine size, chamber pressure...)

-James (now intrigued)

Hi all, thank you for the replies. The engine I am working on is actually a shock tunnel with a detonation driver mode. So a fuel air mixture is detonated in the tunnel, causing a shock wave and a flame front to travel down the tube. So my job is to get a 'blip' from the shock wave (using a pressure transducer) and a 'blip' from the flame front so as to see how much one might lag behind the other. And I want to do this cheaply ofcourse. Fuel is propane, and although flame temp for propane can go as high as

2800C, with water cooling the transducers manage to survive. I do intend to provide watercooling for the 'boroscope' that I want to build. Since all I want to get is a 'blip' transformed into a TTL signal, high tech photo analysis is not required. So all I need is the transparent rod and a few feet of optical cable. Thanks again in advance for all suggestions. Philip Our website is not the best, we haven't updated in a long time, but here it is. This project is not up on the site yet.

How is the pressure transducer coupled to the system? Is it inside the tube? Can you put the optical transducer there, too? Then you wouldn't need to conduct the light to an external sensor.

(And why apologize about watching the shopping network when you admit to watching Jerry Springer?)

Phillip,

How large a time lag to you expect to see, or want to be able to detect?

This will affect your optical arrangement.

If your silica rod is flush with the inside surface of the tube, then it will collect light over a fairly large field of view. Hence, your detector will see light from the flame/detonation front both before and after the front passes the location of the sensor.

A simple way to restrict this angle of view would be to recess the surface of the rod from the opening in the wall.

Olin Perry Norton

Or use a spark gap biased just below the arc-over voltage as the detector. It will not arc until the flame front hits it.

McMaster.com lists borosilicate rods that'll do the job for you. If you need better performance than borosilicate, ask some chemistry friends if they'll give you a bit of fused silica, fused quartz, or Vycor rod. I'd try to find some water-clear fused silica.

I doubt there'll be much thermal effect on even borosilicate glass if understand the situation; these are fast transients. The average & transient materials temperatures will not be very high.

While you now say you don't care about anything more than detecting a flash of light, I'll bet that sometime in the no-too-distant future someone will say "why not do some specrtal analysis to get the actual flame front temperature history?" I know I would. So will you most likely.

I've done things similar to what you propose. An easy way to hook things together is to use standard compression fittings*. Drill & tap the cylinder to accept a 1/8" npt - 1/4" compression fitting to hold the glass rod in place. This is a strong, secure setup that can be disassembled when needed.

You might think of using a 1/4" x 1/4" compression fitting at the other end of the glass rod to secure the photodetector or fiber optic bundle to the end of the glass rod. Use a short rod so optical absorption by the glass doesn't kill the whole deal.

In fact, you may not need the glass rod at all as others suggest. Let the photodetector look through a 1/4" stainless tube.

Good luck; it sounds like fun,

Dave

*Brass or stainless steel, it probably doesn't matter much.