2v 28A high power diode laser supply based on old PC AT PSU / General Science and Electronics / Forums

Forums

4hv.org :: Forums :: General Science and Electronics

« Previous topic | Next topic »

2v 28A high power diode laser supply based on old PC AT PSU

previous 1 2

Move Thread

LAN_403

Carbon_Rod

Sun Sept 02 2007, 07:51AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Having more than a few KA7500B based old PSU, I wanted to use I0JX's method of paralleling the PSU's to build a high current supply (30A adjustable as regulated with TIP36C and a Zener on each.)

The issue with the following mod should be obvious in regards to stability. However, during shutdown the spike will not exceed 2.0043vDC. The problem is under load the voltage drops significantly and can no longer exceed 5A plus without falling below 1.3vDC. This smaller unit is rated for 20A at 5v... The water circulation cooling pump draws 2A... Any ideas?

Carbon_Rod

Sun Sept 02 2007, 07:52AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Another aside, one of the diode lasers appears dead as it only draws 0.7A when the regulated current source can provide 4.6A at 1.7vDC. The threshold current for this unit should be around 3.5A (assumption based on this units fiber count and manufacturer implying it should be 10W.) The supply design should be stabilized with the cheaper one before trying a 35W module or the mystery unit... Do you think this behavior is consistent with a dead unit or is the 10W identification (and threshold estimate) incorrect?

1188719666 65 FT1630 7f

Old Newport unit (10W?)

Steve Conner

Sun Sept 02 2007, 10:02AM

Registered Member #30 Joined: Fri Feb 03 2006, 10:52AM
Location: Glasgow, Scotland
Posts: 6706

Hi Carbon_rod

If you were modifying the PSU to produce 1.994V from the 5V rail, why did you leave the feedback to the 12V rail connected? You should disconnect it and take feedback only from the 5v rail. I expect that would improve your results.

I don't know anything about safety clips. Do you mean some kind of anti-static shorting clip that needs removed at installation time? I've seen many lasers that ship with metal clips shorting their pins together like this.

Finally, about your laser that only draws 0.4A, just try cranking the voltage up a little more! Remember laser diodes are current operated devices: you use a regulated current source to force through the desired level of current, and the voltage just settles at whatever the diode and connecting wires want it to be. This depends on temperature, length and gauge of wires, etc. If you put a fair amount of current through it, and things aren't bursting into flames in front of the output port, *then* I'd consider it might be dead.

A different approach you might want to consider is to add a linear current regulator (with a bunch of hefty transistors or MOSFETs) to an unmodified 5 or 3.3v supply. Linear regulators are less efficient, but they're easier to design and stabilize, and they need no output capacitor, so the charged-capacitor-loose-connection-laser-trashing events that ... hinted at are much less of a threat.

If you make the current sense resistor in a linear current source big enough, you can be pretty sure it won't trash your laser even if the feedback loop messes up completely. I used this approach in drivers for some semiconductor optical amplifiers that sold for about $10k each, to add some fail-safeness. (Unfortunately the manufacturer made a mistake in the pinout diagram on his datasheet, so we ended up wrecking two of them anyway!)

Finally here's a gratuitous pic of the output from a SDL 980nm pump laser running off one of my drivers -_-

1188727374 30 FT30059 20070327192936214 Image233

Carbon_Rod

Sun Sept 02 2007, 02:41PM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

Thanks for the PSU tip, it will be interesting to see if a more stable configuration is available (IIRC one can't safely float a side of the SMPS... still, it may be worth a try.)

Both the Zener and LM317 based current regulator designs suffered stability issues when they dropped below 3vDC. But otherwise such designs work like a charm for 2-9A and 0-20A accordingly.

I couldn't find the model number for the 10W Spectra-Physics module. The Absolute 019-5016 Diode Pump Power Monitor removed the housing with the real model number (the PCB replaced the common metal lid.)

The idea thus far was to regulate V to 1.99vDC, and simply add the appropriate resister in series with the bar.

$10K... ouch... did it have 250watt OSRAM pump stacks in it? ;]

Carbon_Rod

Mon Sept 03 2007, 06:09AM

Registered Member #65 Joined: Thu Feb 09 2006, 06:43AM
Location:
Posts: 1155

The direct method of optical output monitoring seems easier. The idea was to use the threshold properties and power output to identify the modules' compatible replacement number. The V drop across one of the 35 watt bars is off spec while still at the 9A threshold... so it would likely exceed 2v limits at higher output levels without V regulation (weak element perhaps?)

Pic of the simple test unit with a 1ohm 50watt inline resister running off the PSU with reverse diode protection.. Characterization of 10watt CW unit is as follows: 1.70v @3A, 1.82v @4.7A...

Does anyone have data plots for the Newport modules?

previous 1 2

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob