Thermal pile

[marconi]

Disregard this post.
It led me to realise that my assumptions for wavelength were wrong, the diodes were running at 664 nm or longer, not 658 as hoped, and Laserben recently posted here confirming that diodes can do this:

He found that a red diode laser, strongly driven, was at a measured wavelength of >666nm, which agrees with what I saw. Not sure what the nominal wavelength spec is for Marconi's diodes though, probably 660 nm, so a 658 nm diode could be at >664 when driven hard. I only have my deduction to go on, plus the fact that if I set to 664 on a lasercheck instead of the assumed 658, the reading of the Lasercheck and my thermopile meter agree closely.

Yes, It is true that some diodes will do this..not all diodes are created equal.
Apparently., That diode in that module that Laserben has could have been up around 666nm as he claimed.I didnt check it.. As this was from an old (the first 2yr old) batch, That is one of the reasons it was in our garage sale. Plus he got it really really cheap too,.
But I am somewhat confused as he first told me it was 675nm then 666 when cooled. The diodes we use are spec'd from 657nm to 663nm.
Nominally we get 660 at 225mw. I have never seen one shoot up to 675nm.

And it is true that if you know you have a good Lasercheck you can see the variation in readings leading to what you found.
I too have one of these on my bench..It is set to 660nm and agrees with both the Melles Griot and the Scientech meters..most of the time.
If I see one that doesnt look right by doing what you have explained (not matching power) I compare these anomalies with the Ando spectrum analyzer and the diode that shows up high doesnt get used.

[The_Doctor]

I had one or two Rohm diodes really out of spec. Powers down at usual Lasercheck setting. The second I tested was after I'd got to trust my thermopile again, and that was better than I thought, implying that it had an unusually long wavelength that affected the reading more than the actual lower power did. I don't know which wavelength lengthening effect is greater for red diodes in general, that due to high drive, or that due to variance within or beyond spec. Heating will lengthen it, but there might be other mechanisms that do it with high drive.

Incidentally, think about that idea for assessing laser diode max drive, based on a curve of rising output falling relative to the rising current. If the measurement is made on any silicon detector, it might be that it is not falling as much as it appears, if at all, it might be that the wavelegth lengthens at extreme drive anyway, and that might be enough to account for the relative fall, as the calibrated meter has no way to know if that is happening.

Lastly, my post that was to be disregarded was the one before my previous one. The previous one explained briefly the two reasons I came to distrust that thermopile so much at that time, as well as this important thing it showed me about the wavelengths of the red diodes I was using.

[mixedgas]

to 1 part yacht varnish.
(I'd tried matt finish aeroplane kit glue, it was useless. I couldn't get
any butyl acetate as thinner, and other solvents made it coagulate). The
varnish is gloss, but the high viscosity and adhesion allow a heavy loading
of carbon powder (activated charcoal, politely blagged from a brewers
supplier). I can't quantify the amount of powder, as it won't suspend
evenly in the liquid. I treated it as if it was very wet sand, painted on
the ceramic surface of the thermopile, such that vibration could make it
shiny as the particles settled after painting it as flat as I could get it.
I then shook it, keeping it on a plane surface, to even out the thickness
and consistency. I dried it with forced air from a 12V 80mm fan at very
close range. The forced evaporation (and the high-frequency agitation in
the fan airflow) made sure the carbon could not settle in time to allow a
shiny surface to develop in the dried film.

Same basic paint I use. next question, how you gonna handle the memory effect (time lag) of the TC pairs? Ie you shine a 5 watt beam into it, heat the housing then try to measure a 45 mW beam, there will be a a oddball settling time proportional to previous use, this could make tuning a cavity difficult. Scientech goes out of their way to null it,
with some sort of sample and hold and subtract. Also how ya gonna compensate for ambient? A dummy TC ? Not improtant unless your going for that "pro" feel, or unless you use a very small heating on the backsides (ie coherent 212 style hockeypuck heads. )

Yeah, the ceramic is fast enough.

Steve Roberts

[The_Doctor]

Jump those hurdles as and when... Cheap too.

First, a decent thermal mass. I like the Scientech 360001 better than the hocky puck shape because it takes longer for the bulk to warm up. No heatsink vanes, so less risk of air currents affecting the outside. Being deeper, it shrouds the plate better by far, so it's very good for small powers.

TEC's vary the ratio of delta T to volts out according to actual temperature, not just the difference, but the difference is the lion's share. If it really IS critical to adjust for actual body temperature, I'd tuck a small thermistor in there and use it to provide a scale adjust to compensate that way. Cheaper than a dummy TEC, and easier to arrange circuits for. That thermistor could also be the basis of a longterm thermal overload alarm if monitoring a giant. The dummy TEC idea is nice to get a control for ambient chage, but by far the cheapest and easiest way is a deep barrel to shroud the TEC, and to avoid walking in front of it if you're trying to measure a 5 mW pointer. Another way is a tiny efficient mirror, to reflect the low power laser in off that so the heat of a hand won't get there too.

Re latency, it's not too bad coming down. A TEC 30 mm squared (as used in the 360001) quickly lets the temperature equalise when less heat is applied, so with the thermistor added, no problem with burning away at it for an hour at 5W then needing 45 mW read, it will be ready as fast as you can manually change the lasers unless you have itchy trigger fingers poised over switches. The real problem is the other way round (fast response to increased input), but a differentiator circuit adding its output to the main voltage will help there. Most of us are unlikely to need that speed. By the time the average show DPSS has stopped modehopping after startup, the meter is more than stable and reading fine. The only reason I want to try a differentiator addon is to allow fast retaking of red diode measures.

[mixedgas]

real problem is the other way round (fast response to increased input), but a differentiator circuit adding its output to the main voltage will help there. Most of us are unlikely to need that speed. By the time the average show DPSS has stopped modehopping after startup, the meter is more than stable and reading fine. The only reason I want to try a differentiator addon is to allow fast retaking of red diode measures.[/QUOTE]


having used a big scientech for about 2 years on a repump station, the thing I really hate is the need for a frequent rezero, I haven't had that issue on the two coherents (one's just a opamp with a thermister correction) , the other has a mess of analog computer gear that adds the tuning bars. Tuning bars are pretty useless, unless you wear glasses or your a newbie looking for cheap thrills.

When I get back from SELEM I'll draw out the schematic for the simple one. Too many things to get ready before selem, and too much overload at the day job. If I get around to building another power meter, and my part time business partner really needs one asap, I'd do 2 TC modules in a wheatstone bridge sort of arrangement. But thats just because my boss is a 78 year old brilliant fellow who uses bridges and lock-ins for everything.

I do have to hand it to scientech, it can easily see the heat off your hand. It helps to have a massive aluminum block around to null the scientechs to average room temp.

Steve

[300EVIL]

I do have to hand it to scientech, it can easily see the heat off your hand.

How about it! I would have never would have guessed a Peltier junction would respond like that. They are damn sensitive!

[The_Doctor]

Lots of couples between two fixed points. It's like a block and tackle system, a small amount of movement apart and if there's nothing to load the line you'll see a lot of it drawn up for a small movement. It's like a kind of passive amplifier.

The rezeroing thing doesn't bother me much. One nice thing about a simple gain stage to scale to 1V/W is that reading on ANY voltmeter will be enough. Many have autoranging, and some have autozero, and even the cheapest now have accuracies no laser sensor is ever likely to have, so they won't be the weak link in the chain. A thermistor for the sensor block temperature is nice, it can compensate for both scale and offset error. If it's good enough for Coherent it should be good enough for us.

The dual TEC in a bridge circuit is nice too, actually simpler than the thermistor idea, and more effective, justifying the cost increase probably, especially if you want to measure handheld lasers. That dual TEC could save you having to rigorously control the ambient radiant heat in front of the sensor, as well as remove scale and offset errors. The main problem with it is one of aiming, because it has to be deep set in a dark barrel interior along with the main sensor. It wouldn't matter which sensor you hit, but you'd have to be sure you only hit one of them if you wanted a quick and reliable reading. It probably couldn't entirely cancel out ambient heat either because there will be enough differences in the two sensors for it to register. I suspect it will lower it enough to read a laser pointer well enough though.

[pixpop]

How about it! I would have never would have guessed a Peltier junction would respond like that. They are damn sensitive!

I have a small one, about 12mm square. Just for grins, I wired it directly to a moving coil meter, center zero, 50 uA each side. If I put my hand near one side of the TEC, the meter would instantly indicate which side. Very sensitive. It would easily respond to a 2 mW HeNe illuminating one side.

I wonder how hard it would be to turn it into a useful instrument.

[Laser Ben]

Bit of an update, I got my two boards today. Both of the piles are read 0.01 on my meter with no light. When I shine the dual maXYZ on it(463.7mW) it reads 0.47V on the meter. I don't even know if it needs the pre heater. The second one died a fiery death at the hands of 85W of 808nm Oh well, the melting was worth $24!

[mixedgas]

I have a small one, about 12mm square. Just for grins, I wired it directly to a moving coil meter, center zero, 50 uA each side. If I put my hand near one side of the TEC, the meter would instantly indicate which side. Very sensitive. It would easily respond to a 2 mW HeNe illuminating one side.

I wonder how hard it would be to turn it into a useful instrument.

I just sent marconi a spare thermal meter board for ahem, examination.
Answer, not hard, there is maybe 20-30 parts on it to cover 0-100mW to 0-10 watt in 5 ranges. .

Steve