I'm Almost Ready For 20W!!!

[scumbagatheist]

Sorry, I'm not good at explaining things in words but I'll try. If this still doesn't come clear, sorry, I just woke up

The op amps inside the head can swing its output to the plus or minus voltage minus 2V. Since the output is ground referenced (brown wire) you can only go to the plus or minus voltage. If you're using two 9V batteries, that would be plus 9V or minus 9V. Subtract the 2V dropout of the op amp's output stage, you get 7V maximum output swing.

Now, you're planning to use a 36V supply, which you will then split in two to get plus and minus 18V. So now the ophir output will be referenced to the new "ground" which is the midway or the "split" which is at the middle of that 36V supply. So now the op amps have plus and minus 18V supplies. With zero input, your output will be at the midpoint or split of the 36V. So you have 18V minus the dropout which results to 16V swing. To have a 20V output, you'd need plus and minus 22V rails or 44V total which is well beyond the max supply voltage for the op amps.

The 20W rating mentioned is for the 20CA-0.1-Y which has an 0.1V output per 1 watt input which would only need a 2V swing for 20W input. I'll try to dig up where I read it and link it here.

Wow..Thank you djQ for the very concise explanation. In that case, I guess I'll just push it as far as it is capable. Dave can always hook me up with a far more powerful meter when the time arrives.

PM me and I'll walk you through it if you want.

I've needed a test subject to explain things to such that I can write an effective "how to" tutorial.

ARG, I'll PM you in the morning about this. I would love to be your guinea pig. This sounds like a match made in someone's heaven.

 

[djQUAN]

Wow..Thank you djQ for the very concise explanation. In that case, I guess I'll just push it as far as it is capable. Dave can always hook me up with a far more powerful meter when the time arrives.

No problem. The 2V dropout is actually a bit conservative. With a high impedance load (like a DMM input which is about 1-10Meg Ohms) The op amp can swing closer to the rails. My head used a TL081 on the output which the datasheet mentioned can swing to about 13.5V (with 15V supplies - 1.5V dropout) with a 10kohm load. So you can probably measure close to 18W with your Ophir head with a DMM.

 

[scumbagatheist]

Thank you again...Does that mean I can get closer to the 20W measurements if I use a real LPM system like the ARGMeter?

 

[ARG]

Thank you again...Does that mean I can get closer to the 20W measurements if I use a real LPM system like the ARGMeter?

Mine is only rated for 4-5V (depends on the reference) measurements, else it would require a voltage divider on the ADC such that 0-20V would be mapped to 0-5V for the ADC to read. This would require some high accuracy resistors and a good OP amp such that the accuracy would be preserved.

The rail system is also limited to +/- 9V so you'd be limited to 7W there
I do have schematics for +/-6V (low draw) and +/-12V systems.
The designs are being changed ATM, that's why the site is down. So many ideas. I just have to put them down on paper.

 

[lasersbee]

Sorry, I'm not good at explaining things in words but I'll try. If this still doesn't come clear, sorry, I just woke up

The op amps inside the head can swing its output to the plus or minus voltage minus 2V. Since the output is ground referenced (brown wire) you can only go to the plus or minus voltage. If you're using two 9V batteries, that would be plus 9V or minus 9V. Subtract the 2V dropout of the op amp's output stage, you get 7V maximum output swing.

Now, you're planning to use a 36V supply, which you will then split in two to get plus and minus 18V. So now the ophir output will be referenced to the new "ground" which is the midway or the "split" which is at the middle of that 36V supply. So now the op amps have plus and minus 18V supplies. With zero input, your output will be at the midpoint or split of the 36V. So you have 18V minus the dropout which results to 16V swing. To have a 20V output, you'd need plus and minus 22V rails or 44V total which is well beyond the max supply voltage for the op amps.

The 20W rating mentioned is for the 20CA-0.1-Y which has an 0.1V output per 1 watt input which would only need a 2V swing for 20W input. I'll try to dig up where I read it and link it here.

Finally someone that sees through the clutter and has read and
understood the data sheets....:gj:

Jerry

 

[Leodahsan]

Does the 20C-Y-1 laser head has some gain setting inside? So it could be changed from 1V/W to 0.5V/W?

I just had op-amp lessons @ electronics course. Maybe I could help with the circuitry.

 

[ARG]

It does have a potentiometer for the gain, but changing it voids the calibration which is no good.

 

[scumbagatheist]

Yeah, I am pretty sure I read that somewhere.

 

[djQUAN]

I did a quick search and found the post in the first result that had the various internal pots labeled. And it's ARG that posted it. http://laserpointerforums.com/f70/diy-ophir-lpm-78856.html#post1133802

The pot will only have minimal adjustment range. Ophir will change a resistor to alter the gain so that the pot will reach the 0.1V/W output for the 20C-A-0.1-Y, I just don't know which one as I also haven't seen the schematic.

added note: JP2 is connections to the thermopile. JP1 connects the ground to heatsink but is a poor one. I tried shorting it out of curiosity and it resulted to an offset when someone is touching the heatsink.

Although that head shows a CA3140E op amp, mine used a TL081. The other op amp is the same.

 

[Leodahsan]

It does have a potentiometer for the gain, but changing it voids the calibration which is no good.

Idea; Can't some one use a analog controlled, stable laser, point it at the ophir and increase the power up to 1W, so the ophir read 1V, then adjust the gain pot down until it shows 0.5V? If the laser won't vary, the calibration will be accurate no?

EDIT:
anyways one can't measure the gain pot resistance for 1v/W, so you can adjust it back if needed?

 

[scumbagatheist]

IIRC, the range isn't linear, so while it may seem calibrated at the given power of the laser in your scenario, it wouldn't remain accurate across the sensor's range. I'm sure someone can explain this better than myself.

 

[Leodahsan]

IIRC, the range isn't linear, so while it may seem calibrated at the given power of the laser in your scenario, it wouldn't remain accurate across the sensor's range. I'm sure someone can explain this better than myself.

It should be linear... I may be wrong but, it isn't just a gain circuit connected to a voltage divider in a thermopile? I'd like to see a schematic for the ophir.

Edit: You're telling me the thermopile don't have a linear curve?

 

[scumbagatheist]

I'm on my phone, so I don't have the datasheet on the Ophir, and I'm certain I don't have the schematic.

ARG has an older thread, discussing the accuracy of the LaserBee LPM, in which they discuss accuracy of various sensors. IIRC, there is some discussion there, regarding your suggestion.

 

[Leodahsan]

I'm on my phone, so I don't have the datasheet on the Ophir, and I'm certain I don't have the schematic.

ARG has an older thread, discussing the accuracy of the LaserBee LPM, in which they discuss accuracy of various sensors. IIRC, there is some discussion there, regarding your suggestion.

I found something. The circuit inside the ophir has 3 pots and 2 ICs. One is the op-amp, connected to a offset pot (zero) and gain pot (1V/W default on your head). The third pot some said its a "curve" adjust, but I read on eBay sellers description the ophir has a time response improvement circuit. That makes me think the thermopile has a LINEAR curve... I'll be searching more about that, as I dream of having an ophir I'll read about it first

Edit: My info is confirmed. The third pot is time response adjust. So we have offset, gain and time response. IMO the curve IS linear!
Read here


It may be of your interest SBA...:

The sensor can measure up to 5 watts on its own
20 watts with a heat sink "BUT AT 10mw RESOLUTION" with part changes on the board

Keep that in mind for the future when were blasting 8 watt handhelds!

 

[scumbagatheist]

I am so thankful to have good technicians, such as yourself. My background is in botany and 1990's era biotech. I always wanted to learn more about CE and such, so here I am.

Thank you. When you're ready to buy yourself an Ophir, I have one NIB from the fleabay seller we all got them from.

 

[Leodahsan]

I am so thankful to have good technicians, such as yourself. My background is in botany and 1990's era biotech. I always wanted to learn more about CE and such, so here I am.

Thank you. When you're ready to buy yourself an Ophir, I have one NIB from the fleabay seller we all got them from.

Thanks for the kind words bro. When I have the money I'll buy a laser head and play with it.

Let's make your laser head measure 20W then... do you have 10 2W 445nm lasers for da testz? :eg:

cheers

edit: regarding the gain lowering, I had another idea...
You measure the gain pot... For 1V/W... the original setting.

Lets see, you measure 5 of your lasers...

100mW (.1V out)
360mW (.360V out)
675mW (.675V out)
2W (2V out)
3.2W (3.2V out)

So you said.. "...so while it may seem calibrated at the given power of the laser in your scenario..." to confirm that the curve is linear (or not linear...) you lower the gain until your 2W reads 1V out... so all your table should read 50% lower...:

100mW (.050V out)
360mW (.180V out)
675mW (around .337V out)
2W (1V out)
3.2W (1.6V out)

if the amplifier is a simple non-inverting configuration, the pot should now have half of its original resistance...

it's trial and error. If error occur, you should adjust the pot back to its 1v/w gain resistance, measure before.

if the response is linear up to 3.2W, most sure it will be linear up to 20W... where it will read 10V. If the op-amp inside the ophir can swing +18-18, lowering the gain to +10-10 will work.

We need some thermopile specialists to confirm that, I can only advice about the circuit.


edit: you have 2 ophirs uh? You can set up both and measure a laser on both, and slowly lower the gain of one. So one reads the full value (1W lets say) and the another half the value (0.5W lets say). I hope you understand me :beer: