Page 1 of 7 12345 ... LastLast
Results 1 to 10 of 64

Thread: Thermal pile

  1. #1
    Join Date
    Mar 2007
    Location
    North Iowa
    Posts
    249

    Thumbs down Thermal pile

    Ask your doctor they say - so doctor here it is ----

    SenKat sent me an old Coherent power module board today. Ass'y # 0168-661-00. On this board is a thermal pile assembly which appears to have a preheater on the rear. Output from the "pile" is through a coax cable to what appears to be an op-amp.
    The pile is about 2 cubic inches and heavy. Any knowledge about this unit ?? Is preheat typical ?? What is the output - mV or mA ? It seems to slightly detect my 150 mW red.

    Mike

  2. #2

    Default

    I don't know. Got a picture of it? Meter thermopiles usually have a heater on the front, not the rear. It is a small coil designed to take a 1 watt input for calibration. Sounds like you've got something else though. WHatever you use, the output would probably be a voltage from a low impedance source but with only a small current drive capability, enough to reliably drive the input of a voltmeter or op-amp.

    If you want a thermopile head already built, the simpler the better. Scientech made some good ones. 360001 for power reading (black coating), and 380101 for energy reading (has a volume absorber, a disk of some crystalline substance to absorb pulses). Either can be restored from their current state to read power, so long as the TEC inside them is intact and not pitted badly by strong laser pulses. If you have a lathe and a block of aluminium, you can make one, but I doubt it will be cheaper than buying on eBay.

    Output depends on the number of thermocouples, but in a Scientech head, is a voltage about 1/11th of the input power in watts.

  3. #3
    Join Date
    Mar 2007
    Location
    North Iowa
    Posts
    249

    Default

    Doc --
    I'm not good at posting pictures. The pile looks like one currently on Ebay. 3/4" input to a black surface.
    The heater at the rear is a transistor plus a three leg temp sensor epoxied into the heat sink. As it gets warmer, the output goes up almost like .1v = 1 Degree C. That outputs somewhere to a comparitor for shutoff feedback to the transistor. It appears to provide a reference temp to the enclosed block holding the pile.
    My "lesser" red outbuts 0.1 mV and the higher one reads 0.2 mV. Looks like I need an op amp for 10 and 100 to 1 boost. Just reference to room temp.

    Mike

  4. #4

    Default

    eBay listing number? Or URL?

    This thing sounds more complex than you need. Most thermopiles only need the difference between hot and cold side temperature. While the response can drift with absolute cold side temperature, it won't drift much. Most solve this by using a bulky cold side block container to shroud the TEC from ambient heat and to absorb and dissipate input heat. This changes more slowly than the TEC does, so it works like a sheet anchor, or a reservoir capacitor with a resistance. A low pass filter. So low that it works as well as having an absolute reference, so long as TEC output/temp difference ratio remains same. It should be consistent enough over the range of temperatures most show lasers can put into it, even if left doing it for hours.

    I think that anything that needs to monitor the absolute block temperature is either expecting high powers for a long time, or also needing very high accuracy. All you'll likely need is a mug-shaped block with a TEC in the bottom, and a way to calibrate it after painting the hot side with activated charcoal paint. If it comes ready made with a calibrating heater, that can be helpful, but it's not needed if you have a known reference laser to calibrate it.

  5. #5
    Join Date
    Mar 2007
    Location
    North Iowa
    Posts
    249

    Default

    Doctor --

    I think you are right about "precision". I don't need it and I can always hook that feature up. For now I'll work on an op-amp with 1:1 and 10:1 scaling.

    Thanks - Mike

  6. #6

    Default

    Here's a small board I made for mine. Might need tweaking of values. I suggest lowering from 10M to 4.7M in the zeroing offset tweak network, and possibly lower Zener diode to slightly lower voltage if you want to run this on 5V. I chose 5V1 because it has smallest temperature co-efficient, but it's not that important. The fixed scale change resistor is 88K in the layout, 91K in the schematic. Actual value depends on other parts values. I chose it by centering the scale pot then starting with the resistance closest to ideal, leaving the scale pot to fine tune it.

    One problem with this design is that the supply should be isolated from the ground because it makes a floating ground commoned to the chassis. I'm using a small 5V to +/- 12 VDC 1 watt power converter for this. That will have the advantage of a strong negative supply rail for a second dual op-amp, once I design a differentiator to accelerate meter response to changing input. Another advantage is that you get a 20V range from a 5V supply, meaning that at 1V per watt you can measure to 20 watts with this.

    I ran into problems with the LED panel meter I was using. It changes the current drawn, and I was using an LM2904 Regulator. This seems to have poor load regulation, so the meter changed the supply voltage to the gain stage, which in turn changed its output by enough millivolts to change the meter display. Nasty loop, a really vicious cycle. When I get over it I'll try separate voltage regulators for the metar and the gain stage. Or forgo battery supply capability by doing it with a small dual-winding 5V toroidal transformer. So long as the gain stage is fed a truly stable regulated supply, this thing should work fine, despite being almost absurdly simple. You might want to add a resistor across the input terminals, as I deliberately designed this to have the highest impedance input I could get, as it's a kind of universal gain stage, could be adapted to all kinds of stuff.

    If you're up for another mammoth post, I'll add my notes on painting a TEC. There's definitely a knack to that, so I wrote down my findings when I worked on it. It's recently posted again to alt.lasers too, cos someone asked there about this.
    Last edited by The_Doctor; 03-22-2007 at 01:29.

  7. #7
    Join Date
    Jan 2007
    Location
    Florida
    Posts
    4,382

    Default

    "Piles"
    uh-huh uh-huhh-huhh
    I've had one or 2
    That's English slang, right?
    Oops-gotta go

  8. #8

    Default

    This is why we say thermopiles. Hemorrhoids are best kept warm. Apparently.

  9. #9
    Join Date
    Jan 2007
    Location
    Florida
    Posts
    4,382

    Default

    Ha!!

  10. #10
    Join Date
    Mar 2007
    Location
    North Iowa
    Posts
    249

    Default

    All right -- Good laugh -- I thought I had it wrong when my sensor label was spelled Thermal Pile !!! Thanks for the laugh
    Also Doctor -- Thank you for the schematic - saved me a couple hours of "Piling" around.

    Mike

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •