Thermal pile

[pixpop]

OK, has anybody analyzed their lasercheks? What's in that window???

I analyzed the heck out of mine

One rainy weekend, I decided to map out the response curve. I rigged it up so I could illuminate it with constant light, and took a reading at each wavelength. Since the lasercheck is such a pain in the proverbial to use, this took me upwards of an hour to complete. As I went, I typed the readings into a text file on my computer. This was a great plan, until I accidentally deleted the file. After that, I couldn't remember why I wanted to map the response, so I didn't go back and collect the readings again.

[The_Doctor]

The broadband thermal probes work by sensing a change in temperature, yes. Though the really good ones use a TEC rather than just a thermometer.

Smaller power meters like the lasercheck (or the light detector on an Ion laser) use photoresistors like pixpop mentioned.

Adam

If you had a bet with Steve, you lost! It's silicon in a Lasercheck. One reason for that is that the curve of its response is consistent for most (all?) silicon cells. That makes it easy to graph a compensation for. You mentioned a meat thermometer recently... I think those are actually used for strong lasers. There are high power laser meters that really are essentially the same thing, they just have a pad of some kind on the end instead of a spike. Like TEC's, they have the advantage of not caring about wavelength so no response curve compensation is needed at all.

[buffo]

I think those are actually used for strong lasers.

Um... That's what I said when I posted it:

It's a pretty cool idea, though it's only really practical for larger lasers. (You need to be able to generate enough heat to warm up the sensor block.)

It was destined for a guy with a large CO2 laser. Considering the scale reads from zero to 100 watts, I'd say that laser pointer jockys need not apply.

As for the bet with Steve Roberts over silicon vs cadmium, no - that wasn't me. I wasn't the one that posed the question (it was Steve-O and Pixpop that were talking about it).

As for Pixpop's post that it was a photodiode, I couldn't honestly remember if it was a photodiode or a photoresistor in the lasercheck. Upon further checking, he is correct - it's a photodiode. (I incorrectly posted that it was a photoresistor.)

However, even though the vast majority of photoresistors are cadmium sulphide, there are a few that are made from silicon-zinc sulphide, so it could be *silicon* either way.

Eh, at least that's my story - and I'm sticking to it!

Adam

[The_Doctor]

Sorry, not Steve, meant Steve-o.

Re photoresistors, I think they can be made out of various stuff (and old solar cells too, from zinc selenide, I think), but I remember it's not accurate, and it's slow. Silicon isn't always fast, but it can be, very. I think the large area photodiodes are slow compared to small PIN photodiodes with small capacitances, but still fast.

Re meat thermometers, ok, I was thinking that they weren't that inaccurate, is all. (I saw that post). I think they're fitted with a better sensor that couples better to the business end, and the calibration is probably better. I've seen them on eBay often enough, and the difference between those and a meat thermoter is small, they're both crude. But they don't have to be otherwise. It's like the scales at a scrap yard, so long as they pass frequent checks, no-one's going to mind if they slip by an ounce when they're weighing tens of pounds or tons.

What amazes me about black painted TEC's is if you get it right you can have something that looks crude, costs crude, but measures like a thousand bucks. And the only thing that makes it difficult for everyone to do and prove it is that the access to certified calibration is often expensive. Makers will tell you that they take extreme care with low temperature coefficients of electronics and such, but that's not hard or expensive to do. Many cheap op-amps can do it. At >20 mA a Philips BZV49 4V7 zener diode has a near-zero tempco and I found a load of them on eBay for silly money.

Just rambling now but the point is, we don't have to pay through the nose for good stuff if it really can be matched cheaply. Just another way to show how true that should be...

The photodiode in a Lasercheck is a good one. In single quantities they might amount to a quarter of the cost of the meter. (If you bought them from RS in nice windowed cans, anyway...) So I guess not everything comes cheap.

[steve-o]

Solar-cells are damn fast, but I dont know how much power they can take..hooked one up to me DMM and got an immediate reading..

[buffo]

Actually, someone here (not sure who?) was posting a while back about a TEC-based power meter that they were playing with that was sensitive enough that it picked up the radiant heat from their body whenever they got near the detector! They had to shroud the whole thing in a block of metal to shield out any IR from heat sources in the room. That's pretty damn sensitive! Pixpop's experience (above) with the bare TEC connected to the moving coil meter sounds quite similar...

So yeah, a TEC-based thermal power meter - with proper calibration - would be both inexpensive to build and *very* accurate. There's a few articles in the laser FAQ about the idea, but I haven't really pursued it yet. (It's a project that's on the drawing board for "someday", whenever that is...)

The bitch with a lasercheck is that 1) you can't measure a multi-line ion laser accurately with it, and 2) you can't go over 1 watt without getting into stacked neutral-density filters. So while my lasercheck works OK for now, at some point I'll probably want a thermal meter... It's just going to have to wait a while. (I understand that Steve Roberts and Marconi are working on something along these lines already - if they have a breakthrough I might have to try it...)

Re: the meat thermometer thing - yeah, I'm sure the bi-metalic element in that power meter is a good bit more accurate than your standard turkey thermometer! (Steve Roberts said that those things cost like $300.) But in theory you could make one yourself, assuming you had a way to calibrate it. (Those new electronic cooking thermometers that use an RTD are reasonably accurate to begin with... That might be a start.)

As for the photoresistor being slow - good point. I once played around with a cheap break-beam perimiter alarm system based on a photoresistor. (No, no laser involved this time! Visible light only.) But it didn't work worth a damn because the beam had to remain broken for around a tenth of a second for the circuit to respond, and that wasn't fast enough. I replaced the photoresistor with a phototransistor and the problem was solved.

Adam

[The_Doctor]

Actually, someone here (not sure who?) was posting a while back about a TEC-based power meter that they were playing with that was sensitive enough that it picked up the radiant heat from their body whenever they got near the detector!

Me.
Actually, someone else, but a few months after I made my long post about yacht varnish and activated charcoal and a dual op-amp gain stage. I mentioned getting a 3 mW pointer and a 5 mW pointer easily tested, but having to be careful to keep all stuff in direct line of sight of the sensor plate as thermally steady as possible because a hand could read close to a milliwatt.

I also suggested beaming the laser in off a small mirror to dramatically reduce the problem.

There's a few articles in the laser FAQ about the idea, but I haven't really pursued it yet. (It's a project that's on the drawing board for "someday", whenever that is...)

Sam put my charcoal procedure in there, I discovered that accidentally while Googling for other stuff. Only the second time in, first was the observation on visible lack of coherence seen when Opnext diodes are pushed close to breaking point.

You can wait for Marconi and Steve, but you won't find the method I posted lacking. Its basic accuracy is likely 1%, certainly better than 2%. A Lasercheck is only 5%. So all you need is calibration.

I meant to have drilled some new PCB's and tested and posted PCB graphics and parts list and schematic but some wiseguy on eBay kept me waiting 3 weeks for carbide bits that never showed, so I put the thing on hold. I paid for some off another guy after I got a refund, so maybe a week or two to go...

[steve-o]

I tried the NTC/ ohmmeter and it was VERY slow..

Heres a meat thermometer for 38 usd
http://www.techinstrument.com/acatal...FTgrOAodsScbMg

If you put a varnish/ charcoal brick on the tip?

[mixedgas]

Laser Check is a wide area photodiode with reverse bias on it. A current to voltage converter drives a analog to digital converter which talks to a microcontroller, each lasercheck has a EEPROM in it that the factory calibrates for each cell at a bunch of wavelengths to linearize the beast.

Theres a patent out there for it someplace.

STEVE-O, put the NTC in a bridge with another NTC and 2 resistors, the speed will go up.

DAMN I wish we could post schematics in line in threads.

Steve

[steve-o]

The paperclip at the top allows attachments..
Thanks for the PM Steve about the galvos.