I want to say that I am not trying to bash any products out there, I think it's great that there are low cost meters on the market made by forum members, these are a great asset to the community.

I have seen a lot of people on the forum asking how to calibrate a home made power meter or saying they have tested something with a calibrated meter made by a forum member. The truth is, it's not calibrated.

When a meter is calibrated, it is setup in a temperature and humidity controlled environment and the detector is tested at several wavelengths each at several powers using a know good power/energy meter which is certified by NIST. Every laser is checked before and after testing on the detector in question. Every detector is different and every one gets a different calibration curve, and ratios.

Sometimes you may find a meter on ebay that was calibrated several years ago and has not been used for a while, you may thing it will still be calibrated and accurate, this is not the case. There are many things that can lead to a detector becoming inaccurate. The thermal detectors have specific coatings which allow them to absorb a set amount of light, silicon detectors have a limited lifetime, parts degrade including insulation, thermal compounds, and other materials. The detectors may appear to be fine and may work fine, but they are not calibrated unless they are still within the calibration period.

Every major company understands the reasons to have test equipment regularly calibrated, it's so they can insure consistency and repeatability in their products and be able to give customers exactly what they buy. In some cases, it can cause a lot of problems if you end up selling a product that does not meat the same specifications as you state when selling it.

Some people say that just because the calibration has expired does not mean that accuracy is off. It's the same as saying, just because the safety on my gun is broken doesn't mean it's going to shoot someone accidentally, it just makes it way more likely.

I would encourage people on the forum who own meters from companies who are still around to have their meters calibrated. It usually does not cost much money, I received my meter and head back from Coherent from calibration, it cost about $250 plus shipping. It is worth having done as you can be sure of whatever you measure.

Again, I am not trying to bash anything out there, I am just saying you cannot call it calibrated unless it is NIST traceable.

The calibration of rulers is is also especially important.

Mine is NIST traceable. I should know, I did it myself

You don't necessarily need lasers to calibrate a thermal head, but like ben said you do need a temperature controlled chamber (in which the TC and temperature reading device is also calibrated) and having a calibrated power meter makes things a lot easier (because it can be used as a standard)


Just in case anyone was wondering, I am an ASQ/ISO 17025 certified calibration tech. I'm in charge of alllllll calibration for a large aerospace company.

Finally a thread where I can REALLY say I know what I'm talking about

For me having a LaserBee was good enough. If was fairly accurate and regardless of it's calibration relative to other meters it still gave me a means of comparing my different lasers.

400th post

-Tony

Only a few people even would have a need for a meter that's properly calibrated. It's not like the day the "calibration period" ends the readings go off completely. The might only be off by a tiny fraction most of the time. Professionals probably should be concerned with having their meters calibrated to this degree of accuracy, but the hobbyist really only needs a close ballpark.

let's say my meter was calibrated long ago and now is reading 5mW under the real power. is it really worth 250$ to fix it?

LaserBen;

My Scientech thermal power sensor has built-in power input terminals for calibration.

When I received it, I carefully calibrated it with a new quality multimeter.

The black velvet thermal coating on the thermopile looked very uniform.

I assumed I would have very accurate, repeatable results.

One very cold night I was testing a bunch of Red laser diodes, and getting strange results.

After pulling my hair out for half an hour, I noticed readings changed when the central heating system cycled on and off.

Ambient temp was cool, and the heating vent air flow was considerably warmer. This played havoc with my power readings.

I had learned the reason for a temp. controlled environment.

Now, I try to test when inside and outside temps are similar.

I also have turned off the central HVAC to maintain more consistent temps.

LarryDFW

When we first started with Laser and Thermal LPMs we found the
same problems with air currents... uneven heating and even lighting.

When we calibrate the LaserBee products we make sure:-
1) the test area is at a stable temperature.
2) the test equipment is warmed up for at least an hour.
3) all fans and heating systems and non-required lighting are turned off
during calibrations.
4) we do all tests 90 degrees to the beam travel to minimize body heat
pickup by the Thermopile Sensor.

Jerry

I imagine for the builders and DIY folks that calibration is important, in order to have some credibility in a market full of exaggerated claims. Thanks to Justin at LaserGlow, (who let us test my meter against his pro Coherent), I'm pretty confident that my good ol' Kenometer is at least in the ballpark. Do I need it to be precise? Not really. For me, I just need a meter to track the degradation of my lasers over time and test out their output using different batteries. When I test them, I do them all at once under the same conditions. I don't much care whether that's NIST-accurate, I just need somewhat precise comparisons between my units over time.

I have no illusions that the readings are world-class. But as a hobbyist, they serve my purpose and alert me when diodes and/or crystals are doing bad, or one brand of batteries is better than another as a power source. That's why I bought it, and I can recommend a good meter to anyone serious about this hobby.

The calibration precautions Jerry lists above are also great advice to keep your readings as consistent as your meter will allow. That's why I love this forum. We hobbyists get solid advice from pros like Gooey, Frothy & Jerry, to make the things we can afford perform better.

Kudos to all the smart guys!

Cheers, CC

What about , when you still have your LPM calibrated, to make a laser unit for test periodically the reading, so you can be sure that the meter is still working good ?

I was working on one of these ideas, just cause i had to send my old Coherent to a lab for calibration (2 years after the expiration date, poor old meter ..... but i don't have enough money for make it calibrated all the years, so i'm studying a possible DIY alternative ), and they damaged it (they said they dropped the unit ..... i don't know if it's true or if they just messed up with it and broken it, it's not a Coherent lab ..... anyway they admitted that it was their fault, and promised to get another one identical for me in exchange for free, but are passed 3 months and i'm still waiting ..... crossing all ten fingers ) ..... anyway, the basic idea was to take 3 lasers, a GGW, a LCC and a green module, made a very well stabilized thermal block, some extremely stable driver circuit with 2 different powers for each one of them, like, 15 and 150 mW, and an external feedback circuit for it, and to use it just for check the meters ..... How do it sound, as idea ?


BTW ..... for the funny part, all that calibration cycle resemble the question of the egg and the chicken ..... you need a calibrated instrument for calibrate the laser, but before you need a calibrated laser for calibrate the meter, but before you need a calibrated meter for calibrate the laser, but before ..... (ad infinitum )

Scientech has clear instructions for calibration in their older meter's manuals. Their calorimeters have a heater connection just for this purpose.

As far as cool/warm breezes in the room . . . .



The 36001 size calorimeters fit inside the units. One is solid filled with closed-cell foam to insulate the calorimeter from ambient temperature. The other's extended aperture cowl restricts air movement at the thermopile.

Peace,
dave

Yeah I even noticed this after a few hours of using my 365 + 360001, when the resolution is set to 200mW even sticking my finger about a half a meter in front of the aperture raises the reading by about a mW! That may not seem like much but I think it speaks a lot to the sensitivity and potential error of thermal meters like these.

Yeah this issue comes up in pretty much all measured aspects of science, precision is everything, but where is the reference? I imagine with lasers that they would use extremely stable (probably gas?) lasers under controlled conditions to properly calibrate meters. Someone correct me if I'm wrong though haha

Hey dave...I've seen these pictures before in another post of yours, and was wondering, are those something you acquired that originated from Scientech? They look very useful in attempting to reduce error from ambient temperature and air flow.

They are Scientech products. However, as you have discovered, this series of disc calorimeters is no longer in production. These "housings" are only available (occasionally) as used.

36-0206 (long aperture) & 36-0203 (insulated)

Peace,
dave

If you handle the thermo head, wait for it to get a stable reading before checking power. My heads can sense a momentary finger contact on the low power scales.

HMike

@ n_t.....

If you feel your Thermal head needs extra air flow protection...
you cam easily add an extended shroud to your Head made of
anything opaque...

BTW...
I see sensitivity to detect body heat a Positive... not a Negative...
since it can react to minute changes of Laser Beam power...
It is up to you to place the Thermopile Head into an environment
protected from outside interference...
I could easily make a Thermopile Sensor that will not change readings
by the heat from a human body... but it would not be very accurate
requiring many mW to show a change in readings... which would skew
the actual power of the Laser's beam...


Jerry

As I suspected, thanks for the clarification and information.
Yes, I noticed this too, before I take any readings I let the meter "warm-up" and make sure the ambient reading isn't fluctuating before I zero it out and actual measure the laser's power.
Thanks for the tip! I might try to fashion some protective contrivances in the future .
And I must not have been clear, my bad; I didn't mean to suggest that the heads incredible sensitivity was a negative, but rather a testament to their engineering. Like you said, it's only under controlled conditions (which is the responsibility of the operator; I should have specified "user error" instead) that they can truly live up to their potential and give astoundingly accurate measurements.

Exactly....

Jerry

I have a question when it comes to thermopile heads.

Why do the readings change according to relative angle of the beam?
The reading is lower when shining the laser at a 90 degree angle,rises by 10-15mW if I angle the beam (20-30 degrees) and shine it toward the edge of the sensor?

I just received my first thermo. LPM and noticed this.
All my other LPM's are photodiode, which does not have this effect.
I read something about this effect somewhere on here, but just did a search and could not find it.

EDIT: I assume this is not caused by body heat, as Jerry stated above, as this also occurs standing 10' away from the LPM.

You Kenometer is designed to shine only on the center of teh thermopile at a 90 degree angle. If your getting a higher reading at a different place on the thermopile it's not accurate.

Here is a picture of the inside of your Thermopile.


As you can see the junctions are arrayed around the perimeter. This is designed to have the beam in the center and have the heat radiate in the thermopile equally to all the junctions. If it's not centered its going to be skewed a little.


As to Ben's statement in the OP, I agree that having a calibrated LPM is desirable, however this is one of the major differences between having a $1500 LPM and a $200-$400 LPM. You PAY for the nist tracking and to have the thing, to quote the movie "My cousin Vinnie" "Dead on balls accurate" This just isn't feasible for the lower end laser power meters.

Thanks Ken.
That answers that question. As I stated, I knew I had read the reason on here somewhere, but couldn't remember (or find) it.

There are basically 2 types of Thermopile Sensors...

The Disk Thermopile



"The disk is made of one set of junctions laid out radially. One set of junctions is
arranged under the aperture while the alternate set is near the edge of the disk
which is attached to a massive heat sink."

Like Kenom posted....For accurate measurements of this Disk type just aim your
laser's beam onto the center of the Active surface...

The other is....
The Wafer-Type Thermopile



"The second one resembles a wafer, or sandwich, with thermocouples running between
the two sides. One rectangular face of the wafer thermopile receives the heat.
That creates a large temperature gradient across the small distance to the other
face that is in contact with the massive heat sink. The array of solid state
thermocouples in the thermopile generates a voltage proportional to this gradient
just like in the disk thermopile. Because of the close spacing of the thermocouples
to each other, the resulting large number of thermocouples in the wafer, and the
large temperature gradient across the two surfaces, the output voltage of this
thermopile is the most sensitive to laser power and the least sensitive to beam
position and size. "


These Wafer types do not exhibit the non-linear response across it's surface
associated with the Disk type.
This is the type used in the LaserBee Thermopile heads...

BTW...they are both accurate Thermopile Heads when used properly..


Jerry

So what happens with the disk type thermopiles if you dont hit the center?
Does it not read or just not read correctly?

I was looking at a used meter on ebay made by Ophir which was disk type,
maybe a good thing they didnt accept my first offer?

What is the best way to measure the power for each head type as far as beam width?

For example.
- Take a perfectly calibrated laser that outputs precisely 100mw.
- Take a NIST calibrated high end meter with the two types of thermopiles
- Adjust beam size to 2mm (perfectly centered on the active sensor at 90 degrees)
- Shine it on each thermopile and note the power
- Adjust beam size to 10mm (perfectly centered on the active sensor at 90 degrees)
- Shine it on each thermopile and note the power

Which thermopile head would be more accurate over the range of beam sizes? Is there an optimal beam size when measuring?

I don't need NIST accuracy as long as they are somewhat close. I do hope to find something to detect if the measurements are drifting though over time. Maybe some kind of heat source that is very consistent or something? This sounds like a good project for those with LPMs to at least know if their meter is getting too far out of calibration. You would at least know if the readings are changing over time.

The Disk type Thermopiles are the most accurate at the center of the Disk...
if the Sensor's Active surface is small... veering off center will produce
a differential output from thermocoulples around the circumference of
the sensor area..
My Kenometers using the Large Coherent thermopile heads do not exhibit
a large variation swing... it is negligible while the smaller heads exhibit
a slightly larger variation..
(I know this because I spent 3 days testing all of my LPM heads. That
includes Coherent... Ophir... Scientech... Newport and LaserBee I..)

Unless a Disk type Thermopile head is physically or electrically damaged
they will work just fine..

@ xStatiCa...

The best way is just off of 90 degrees to not get reflections back
into the Laser...

A good Thermopile Sensor will read a 100mW Laser with a 2mm diameter
beam the same as that same 100mW laser with the total beam covering
the whole sensor... For the Disk Type... due to the above variations...
There is usually a plate with a proper size hole to keep the beam in
between the inner thermocouplles..

With a Wafer type you can place your beam anywhwere on it's surface
and the reading will be the same... with the beam at 2mm or the beam
large enough to cover the entire Wafer Thermopile's surface...

We do our testing with beams no smaller that 2mm and as large as 8mm
diameter on the LaserBee Thermopile.

The case of the LaserBee I... the problem of beam size is only due to
the max Power density/CM2 allowable on the Thermopile's coating to not
damage it.

We prefer the Wafer Type... that's why we chose going that route..

As to drift... If the Thermopile's Hearsink is at ambient temperature
and you are not pumping watts of Laser energy into the head heating
it up way over ambient... then there should be no drift.

We have used the same Shop LaserBee I for over a year and it is still
spot on to the calibration we did back then... But we take care not to
touch or damage/burn the Thermopile's surface and store it in a dust
free drawer when not in use..

Like I said... we have both types in the shop... If they were no good..
we wouldn't have them...

Jerry

Jerry, your explanations (especially concerning LPMs) are always very informative and a pleasure to read. That cleared up a lot of the questions I had concerning the difference in sensors...and I didn't even ask the question! I would rep you but apparently I have to "spread it around" first /