LPM Design: A Beginner's Guide

[Blaster]

Most admirable, thanks just given actual data logs posted of a ophir and laserbee 3.7 usb even at average and max power they show 10 % down i thought of response time but the peak never reaches the maximum on an ophir LPM or others 10% not life and death but noticeable ...my Laserbee is very new and I'm happy with it for sure, just other graphs show it down on other kmown LPMs by about 10%

 

[Trevor]

Most admirable, thanks just given actual data logs posted of a ophir and laserbee 3.7 usb even at average and max power they show 10 % down i thought of response time but the peak never reaches the maximum on an ophir LPM or others 10% not life and death but noticeable ...my Laserbee is very new and I'm happy with it for sure, just other graphs show it down on other kmown LPMs by about 10%

Compared to an Ophir LPM, the 3.7W LaserBee will have a much lower peak power - every time. That's because the peak for diode lasers is always at the beginning of the test. As the test continues, the power drops as the laser heats up.

This is also responsible for a lower average power - the low readings as the sensor heats up are included in the average, thus pulling it down.

Make sense?

Trevor

 

[Blaster]

Not really as it always takes like 20 seconds plus to get a max or nearly maximum reading where as the other LPMs seem to immediately in 1 second jump to peak and then gradually drop off, i was putting this down to battery gradually going flat.I suppose I'm saying if the maximum of a laser is in first few seconds as i actually suspected except with DPSS that need to warm up, my laser bee after a few seconds has only barely climbed to about 1/4 of what it will max out at. i see what you mean by the longer its on the less max power. Question is why is the laserbee in that case have such a slow sample rate meaning you cant really get a accurate reading and have to assume and add about 10% to the shown readings on the data logger. To get what an ophir does in one second or two at maximum .....takes 20 to 25 seconds on a laserbee 3.7 usb to eventually get to maximum ..is it just price or can the sample rate be improved, or is it the thermopile that just doesn't respond fast enough

 

[Trevor]

You may benefit from reading this thread.

http://laserpointerforums.com/f70/lpm-response-time-who-needs-84428.html

The sampling rate is fine; it's the sensor that is slow and is causing the slow ramp up to the actual power.

Trevor

 

[Blaster]

Thanks Trevor i will, you do great work for the forum ...hate slow sample rate that's all ...

 

[Trevor]

Thanks Trevor i will, you do great work for the forum ...hate slow sample rate that's all ...

Ordinarily I'd move discussion over to that thread, but I think since this is an important aspect of design, it's good to have over here too!

I did a quick test of the same laser on an Ophir 20C-A vs. a LaserBee 2.5W USB. The test duration was 60 seconds and the laser was allowed to completely cool between trials.

I've marked the averages with horizontal lines.

Ignoring the overshoot, the Ophir 20C-A read a peak of 1351mW.

The LaserBee read a peak of 1300mW, but as the sensor continued to warm, this was kicked back down to 1288mW (this knee is, I believe, part of the LaserBees sensor nonlinearity correction). Because of the nature of a knee in the curve, it's difficult to know whether that is a "true" peak or not.

Thus, the LaserBee read (at best) a peak 4% lower than the Ophir 20C-A. Not awful, but it's not exactly a selling point either.

Now, on to the averages.

The Ophir 20C-A averaged 1290mW over the duration of the test.

The LaserBee averaged 1182mW over the duration of the test. The lower value is due to the the amount of time it takes the sensor to respond to the full power of the laser.

The difference is approximately 9% - which backs up what you're saying about a ~10% difference.

Also, note:

Sampling rate should NOT be confused with sensor response time. The LaserBee I used samples at around 17Hz - while the Rubicon samples at just under 10Hz. The "slowness" you see in your LaserBee is due to the high thermal mass of the sensor, not because of a slow sampling rate.

Hope that helps!

Trevor

 

[Spooky]

Trev, Arg,

Ever looked at doing LPM's for higher power lasers?

I have a good one but at $4,000 not cheap, there's an aweful lot of mid hobby, light industrial users that would love an LPM that can run to 200 watts?

I guess my question is

Is it possible?

cheers

Dave

 

[ARG]

Definitely possible, but the problem is finding a material that can handle that power density without degrading over time.

I have looked into it in the past, probably should revisit it and look into it throughly.

 

[Trevor]

Trev, Arg,

Ever looked at doing LPM's for higher power lasers?

I have a good one but at $4,000 not cheap, there's an aweful lot of mid hobby, light industrial users that would love an LPM that can run to 200 watts?

I guess my question is

Is it possible?

cheers

Dave

200W? It would be stating the obvious to say that the main challenge would be a sensor. :tinfoil:

That power level couldn't use the hobby-level-TEC-glued-to-a-heatsink model; we'd have to go full-on graphite-disc-with-radial-thermocouples. At that level of fabrication, it could potentially be more cost effective to simply use a professional sensor (potentially water- or fan-cooled Ophir OEM sensor).

We'll talk about it, I suppose.

EDIT: He beat me.

Trevor

 

[Spooky]

Very very few pro LPM's for industrial lasers have data logging the cheapest chinese POS runs £700 ($1200) and uses up sensors like they are free.

Coherant do a small laser wizzard but it's noted for not being that accurate (and no data logging)

Even a "magic box" that attaches to a commercial head would be viable for the number of laser users in the world with machines under 200 watts.

Power density in the 200 watt tubes is over an 8mm diameter circle (the power is measured at the combiner) so density shouldn't be too bad.

Just some thoughts about a market nobody has really exploited

cheers

Dave

 

[Atomicrox]

Shouldn't be hard to make a DIY LPM for a 150W Ophir head (which is about the same price as the 20W head, but has less precision). The hard part would be heatsinking it properly..

 

[Trevor]

Shouldn't be hard to make a DIY LPM for a 150W Ophir head (which is about the same price as the 20W head, but has less precision). The hard part would be heatsinking it properly..

Yep! But to get to 200W we're getting into water-cooled or fan-cooled territory.

Trevor

 

[Atomicrox]

Indeed! IMHO having the full 20W range of the regular heads would b good enough
The only issue AFAIK is getting decent ADC.

 

[ARG]

Yeah, plus noise is a problem at .25mV per 1mW ratios.
Also, I don't fancy having to generate a +/-24V supply, would be best to have it run off of AC mains at that point.

 

[Atomicrox]

Why would we need .25mV per 1mW? It should still work above 5W at 1mV/mW, as long as the supply voltage is high enough. But I agree the supply might be an issue..

 

[ARG]

Well, I don't know of any (reasonably priced) 16-32 bit ADC's that measure 0-20V, would have to change the voltage to power ratio with a voltage divider.