Larry Hendrix wrote:
I would suggest that you might search for high temperature integrated emitter/detector packaged assemblies instead of working with fibers - only because sometimes the fibers are prone to breakage
and assembly difficulties. Also, do you have the option of fabricating the wheel with a slot (or two slots for balancing) to use transmitted light instead of reflected light? The signal will be
considerably better and more reliable over time using transmitted light, with either the discrete fibers or a packaged assembly.
The minimum frequency response should be about 10X the 2500 Hertz for pulse reproduction, but a 25 kilohertz system should be trivial to design or purchase using reverse biased photodiodes. I would
look at Newark Electronics for components (I may be able to look for some if you want assistance). Also, for the sake of reliability and assembly convenience, you would need to use several fibers
in a small bundle for both the illuminating LED and the detector diode. If you use the fibers, you do not need two holes; a small bundle of fibers can be placed through one hole in the housing
straight into the reflective surface. Divide the fibers into two small bundles at the LED/detector end. The "crosstalk" between the fibers should be negligible.
Larry Hendrix
larry_hendrix@charter.net
I am attempting to design a turbo charger tachometer using fiber
optics. The temperature of the compressor part of the turbo charger
housing can be up to 250 F so I suspect we need glass fibers rather
than plastic. There can be two holes in the housing about one or two
mm in diameter. We can use high temperature epoxy to attach the
fibers to the housing. The angle of the holes will converge on
a spot on the edge of the compressor wheel.
The compressor wheels are 2 to 3 inches in diameter.
Half the of circumference of the wheel can be painted black
and the other half can be polished aluminum.
The turbo charger rotates at speeds up to 150,000 RPM.
Therefor the pulse response of of the system must be
2500 Hertz or better.
My first guess is to use a laser to illuminate the edge of the
wheel through a glass fiber and a logic receiver on another glass fiber
to out put a TLL or CMOS pulse to a counter.
The glass fibers only need to be 3 or 4 feet long.
Can you makes some recommendations on where to start?
Thank you in advance.
Paul Lamar
Larry Hendrix wrote:
I would suggest that you might search for high temperature integrated emitter/detector packaged assemblies instead of working with fibers - only because sometimes the fibers are prone to breakage
and assembly difficulties. Also, do you have the option of fabricating the wheel with a slot (or two slots for balancing) to use transmitted light instead of reflected light? The signal will be
considerably better and more reliable over time using transmitted light, with either the discrete fibers or a packaged assembly.
The minimum frequency response should be about 10X the 2500 Hertz for pulse reproduction, but a 25 kilohertz system should be trivial to design or purchase using reverse biased photodiodes. I would
look at Newark Electronics for components (I may be able to look for some if you want assistance). Also, for the sake of reliability and assembly convenience, you would need to use several fibers
in a small bundle for both the illuminating LED and the detector diode. If you use the fibers, you do not need two holes; a small bundle of fibers can be placed through one hole in the housing
straight into the reflective surface. Divide the fibers into two small bundles at the LED/detector end. The "crosstalk" between the fibers should be negligible.
Larry Hendrix
larry_hendrix@charter.net
I agree with the light interrupter idea.
However we don't really have the option of modifying the wheel because I am not confident
we could re balance it. The optical fiber option requires very small modifications
to the compressor housing only. One side of the wheel is painted with blue dykim
or thin epoxy paint. Just two small 1 or 2 mm holes (.039 to .078 inches)
in diameter. This could be installed by anyone.
Yes I do need assistance on this as I am new to optical fibers.
I am not too optimistic on finding a high temp integrated emitter/detector packaged.
I'll check my Newark catalog.
Great idea on using one bundle. Lets work along those lines.
Any idea on the max temp of plastic fibers?
Thanks for your help.
Paul Lamar
The Rotary Engine NewsLetter. Powered by Linux.
ACRE NL web site. http://home.earthlink.net/~rotaryeng/
Copyright 1998-2004 All world wide rights reserved.