Homemade High Power Diode Laser

[koliaco]

Re: Do you mean Sony SLD1239-JL54? If so this LD does have a window.


I will check with a needle to see if mine has one, cause to me the junctions looks too clear to have any windows on it.

[a_k]

I had several of the Sony LDs and they all were windowed.
My windowless ones (Samsung i think) don't even have a complete casing and the laser crystal is exposed.

dscn2576-1.jpg (30.43 KiB) Viewed 2331 times

The MAX1978 you can order as free sample from Maxim.

Ahmet

[koliaco]

I had several of the Sony LDs and they all were windowed.
My windowless ones (Samsung i think) don't even have a complete casing and the laser crystal is exposed.

I bought my LD from ebay - a korean guy.the data sheet he gave me is SLD-1239JL-54 but I am very sure the diode has no window - I could touch the junction with needle point. So I am not sure if I was sold an item and given specsheet of another product or its just variety with the same product code? Anyway it give very good quality beam without optics, the only concern is its temperature stability.

I could achive 0.001-0.002 V stability at thermistor (MAX1978) which will correspond to 0.07 to 0.14 C stability (base on EVAKIT dataheet, 14mV~1C). so this figure is by far not anywhere close to 0.001 as said achievable with MAX1978.

Perhaps the LD housing needs to be smaller? My curent housing is 20x20x20mm aluminium block and the LD fits slugly into it and has Arctic Silver 5 grease. the NTC is located as in previous post picture and I have shielded cables for NTC, Peltier and LD. The ground planes for LD and TEC are separated.

The NTC and Peltier are not soldered to MAX1978 but have 3pins connection, maybe that could cause instability as well.

Power supply for TEC is regulated solid state 5v 3.65A
LD - regulated 6v 1.5A actual LD: I=90mA, V=2.34V


LD current stability is 0.1 - 0.2 mA with single LM317, i tried with two LM317s as current limiters on 12V and they seem to be less stable (same 0.1-0.2 mA but they jump more frequently plus voltage starts to fluctuate as well)

[wler]

I could achive 0.001-0.002 V stability at thermistor (MAX1978) which will correspond to 0.07 to 0.14 C stability (base on EVAKIT dataheet, 14mV~1C). so this figure is by far not anywhere close to 0.001 as said achievable with MAX1978.

How did you measure the mV stability against the roughly 1V background? If it's done with the multimeters shown in the background, I would think they are the source of instability (moreover, electrical noise can easily induce a few mV of fake signal). A better option would be to use a 16 bit ADC which derives its reference voltage from the same source, in this case the on-chip reference.

Put in another way: the chip can be so stable because the setpoint and thermistor voltages derive from the same source and thus a drift largely cancels out ratiometrically. If your voltmeter has its own reference, then you fold in the drifts of both voltmeter and the on-chip reference, and the combined drift is likely to be much larger than the precision you want to measure (unless they happen to cancel out accidentally).

I think the easiest way would be to trust the circuit and just go on; assuming that you use high quality components. Most important is the resistor in series with the thermistor, if that has a temp coefficient of say 100ppm, you can't expect to get better results than that obviously.

[koliaco]

I think the easiest way would be to trust the circuit and just go on

The max in picture is MAX1978EVAKIT, it has the Sofware to come with 25 pins connector for DAC, but my computer has no 25 pin connector so I was unable to use it. Perhaps I should just trust it and make holo to see how its goes as you say.

I do notice improved stability when I unplug the set point measurement outlet and just monitor Thermistor voltage