You can excite a platinum resistance thermometer with 15kHz AC and read the temperature to an accuracy of a few microdegrees if you try hard enough. You'd need a bifilar-wound transformer in your Blumlein bridge to do that well, but getting to +/-0.5C would be less demanding.
-- Bill Sloman, Nijmegen
That's an awfully high frequency for an AC bridge of ppm-level accuracy, given the approximately PTAT change in resistance of a noble metal RTD. I'd fear going much beyond the hundreds of Hz due to stray capacitances and such like. If it was not a problem, I'd suggest synching it to the modulation frequency of the SQUIDs (usually 100kHz or higher, but they're working to a null).
Best regards, Spehro Pefhany
-- "it\'s the network..." "The Journey is the reward" speff@interlog.com Info for manufacturers: http://www.trexon.com Embedded software/hardware/analog Info for designers: http://www.speff.com
t some
=A0They
king
thing
or
s I eTony Williams seems to have made it work; I though that he'd have to provide separate in-phase and quadrature detection, but I was told that he managed to get away without it.
-- Bill Sloman, Nijmegen
It is possible to vapor deposit non-ferric bi-metal plate and measure the change in capacitance due to temperature. If you are building a chip anyway, you can build the sensor together with all the electonics.
AC is fine for capacitive plate.
One idea no one has mentioned yet, is to circulate a fluid into the sensitive region, and temperature-measure it at a distance when it flows out.
Ruby fluorescence has a decay time of about a millisecond, and that's temperature dependent. You can feed a ruby chip through a fiber from a blinking blue or green LED, and look at the red light that comes back for phase shift; phase-lock the blinking LED so the red return light is (for instance) 50 degrees out of phase, and the easily-measurable frequency is a thermometer.
The electronics is all on the other end of a cable. Perhaps I can make a capacitor on the substrate of the sensor and sense it via the cable. This would be a driven shield. The other though I just had was to make a tuned circuit and measure the resonant frequency.
Where that coffee?
I like the idea that it uses two wildly different wavelengths. This means that I can have just one fiber that ends in the ruby and use a filter to keep the strong pump LED out of the detector that picks up the decay. This would mean that the op-amps wouldn't crash into the rails on the pump pulse.
Platinum resistor, PN diode, there must be some thermocouple that is non-magnetic. I've used very fine copper-constatan TC.. it's only a little magnetic.
George H.
Could you build your own out of materials you like?
It doesn't have to be two metals stuck together. That just gets a lot of change for a given temperature. How about gluing a mirror across 2 blocks of xxx and yyy? Brass and plastic? Plastic and glass?
That won't move much. How stable is whatever you would be mounting it on? Ahh... Build two of them in opposite directions. Then don't measure the position of the reflected dot but the difference in position between two dots.
-- These are my opinions, not necessarily my employer\'s. I hate spam.
This is basically the bimetal spring idea. I can't use brass but perhaps two plasics may be worth a look.
If you can through ordinary means see that it is magnetic, it must be hugely so from my perspective. A cute test requires some thread and a strong magnet
Hang the object in question from about 4 feet of thread from a fixed mounting point where there is no draft.
Bring the strong magnet to within 0.1 inches of one side of the object and wait.
If the object moves at all it is more than tens times too magnetic.
I good magnet will attract the ink used on a dollar bill.
-- These are my opinions, not necessarily my employer\'s. I hate spam.
Photocopies and things printed with Laser printers are magnetic. So are: Bird droppings Dried blood Most brass Grey PVC sometimes The hands of a machinist Lamp black Quaker Oats
Yup, I think this is smaller than that. I use it as a temperature sensor in our optical pumping apparatus. (Signal is extremely sensitive to any temperature changes of the Rubidium cell.) We get sub 1 milli-gauss linewidths at fields of maybe 10 Gauss. Any magnetic stuff screws up the homogeneity and the linewidth. I tested it (years ago) by putting lots of thicker pieces of constantan near the cell... line width was OK so a piece 10 times smaller was fine... But it sounds like what you need is even more sensitive..... Still you can get very thin wire. It's an easy solution if it would work.
George H.
without
I trust you have evaluated superfield silicon hall effect sensors. You mentioned a 1e9 field but neglected to mention the units.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.