Constant Power vs Constant Current

On: Thu, Apr 18, 02 12:42:30 AM

Colin Kaminski wrote:

I have had very experienced laser designers recommend both constant current and constant output power. Tom, can you provide me with some guidance? I have been using Frank's circuit with constant power (APC). Should I try constant current? What driver would be versatile? I am looking to start trying different diodes.


Tom B. - Thu, Apr 18, 02 04:45:21 AM

I'm in the constant current camp myself. The weak link in the APC scheme is the photodiode, which is quite sensitive to back-reflections and other nastiness. One perceived disadvantage of constant current drive is the need to hold the diode temperature constant, either by actively controlling it or running in a constant temperature environment (like my basement). But you have to hold the temperature constant anyway to avoid mode-hopping. My reasoning is that if you have to hold the temperature constant anyway, and at a given temperature output power is constant for a given drive current, why go through the rigamorole of amplifying a weak constant signal (and added noise) from the photodiode to produce what should be a constant current, but won't be because of extra noise and back-reflections. Worse yet, the thing most likely to change the amount of light relected back to the laser is opening the shutter, and you don't want the APC circuit to start responding to an apparent change in output power just as your exposure starts. The simplest constant current source is a stable battery and a variable resistor. Or a well filtered variable lab supply and a fixed resistor. I'm using both, and they seem to work well enough. But I would really rather have a low noise variable current source that's a little more high-tech. Eventually I'll get around to designing or buying one. In the affordable category, I like the Thorlabs LD1255 0-250 mA driver - the 1 microAmp RMS noise spec looks very good. With the cable and power supply, it runs to a total of $140 US. ( I seem to recall looking at some drivers and chips from other companies and finding that they didn't bother to provide noise specs - not a good sign.

Colin - Mon, Jun 03, 02 01:27:13 PM

OK, you talked me into it. I just ordered the LD1255. I'll let you know how it works.

Tom B. - Fri, Jun 07, 02 02:52:12 AM

Cool - I'd be interested in your impressions as I've only seen the catalog description. For general purpose testing, I'd be tempted to put the thing in a ferrous metal box to shield it from stray magnetic fields and replace the little trimmers with good quality front-panel pots (multiturn). Also add a digital panel meter module to monitor drive current. And one for temperature, too, of course. And so on. Thus do simple projects grow into multi-weekend monsters.

Colin - Fri, Jun 07, 02 11:22:55 PM

You would save about $120 to just start from scratch. Low noice circuitry is not that much harder than a simple current source.

Use this form to add your comments to the current forum entry above
Your Name:
The Text...

HTML, or
Plain text

Action or

| Main Index | Reload | Help? |