Blown diodes and questions about the Badpip Quad driver

[robvdv]

My setup is as follows:
1 x Badpip quad driver running off a 24V 9A supply
4 x M140 blues in series per channel on two channels (8 total)
4 x P73 reds in series per channel on two channels (8 total)
CM106 like sound card and correction amp

I've managed to blow a blue diode on each chain of 4. I'm not exactly sure how or when I did it. All diodes have a 47mF and 10nF capacitor plus a reverse polarity diode in parallel. Both blown diodes are on the negative side of the series.

When I turn on the power supply the lasers flash brightly for a few hundred milliseconds before settling down to the brightness determined by their modulation voltage. I'm not sure if this is a normal thing with the Badpip driver? I would have assumed a slow ramp up would be more sensible. Is there any kind of soft start built into the driver that anyone knows about? Would a voltage surge from the 24V power supply be blocked by the driver?

Additionally, if I put more than 5V signal voltage through the driver would a greater than configured current go to the diodes, or is there a ceiling? (I suppose I should test this myself, but asking anyway).

The most plausible explanation for me blowing these diodes is (stupidly) unplugging and replugging the leads between the driver and the diodes while the driver was powered, but I just want to get as much information about possible about this. I'm kinda eager to avoid further damage.

Here's a pic for the curious:

[chad]

I don't have a bad pip driver but I wouldn't recommend any caps in parallel with them. The diode shouldn't hurt anything.
Out of curiosity why the scanner sets?
And how much current are you running?
Chad

[robvdv]

I wouldn't recommend any caps in parallel with them

Why not? I thought it was a pretty common thing to do to avoid damage from spikes.

Out of curiosity why the scanner sets?

It's kinda like three spiros overlaid. This isn't a conventional projector. The spiros illuminate a vibrating water lens. It's for an art installation at AfrikaBurn (South African Burning Man regional event). I'll post the full build soon.

And how much current are you running?

I can't remember exactly but it wasn't too crazy for an M140. Maybe 1200ma or so. The P73s are at 800ma or something. I'll push them up when I get the TEC installed.

[chad]

Why not? I thought it was a pretty common thing to do to avoid damage from spikes.

I don't know anyone who does. I have run lots of diodes in series and I have never added caps. The drivers Aren't designed to handle a capacitive load and that is 188uf. A pretty substantial capacitance. What voltage rating are the caps?


Current sounds fine.

Chad

[-bart-]

First off, plugging on and off while powered is a big no-no!

Then, don't put anything across the diodes(except for a suitable lasorb).
dlp manufacturers, put 8 of them in series, nothing added, no problems.

The capacitors in fact make the diodes a far more difficult load to drive. The badpips want to drive an ohmic load, adding capacitance confuses the current drive.

By 47mF you mean 47µF, electrolytic?
To make matters worse many electrolytic are -10% +50%. By putting them in series they distribute the voltage accordingly when modulating.

[robvdv]

I don't know anyone who does.

I've seen this advice repeatedly, here for example: http://laserpointerforums.com/f51/in...ion-55979.html. Are caps the right thing to do when you're driving the diodes off a non-modulated driver for a pointer, but the wrong thing to do if you're driving them off a proper driver like Badpip's?

First off, plugging on and off while powered is a big no-no!

Yes, it was a moment of stupidity.

Then, don't put anything across the diodes(except for a suitable lasorb).

Should I remove them, then? Are they actively doing harm? Do you think they could be responsible for the blown diodes, or is it most likely that was caused by the hot plugging?

[-bart-]

Remove them, nobody "here" does it that way. The only thing I've encountered professionally across laser diodes, is zeners and lasorbs.

The diodes briefly lighting upon power-up is definitely a BAD sign, and likely caused by the capacitors.
Although, you should rule out the soundcardcorrectionamp too.

Switching converters *can* benefit from *the right* added capacitance across the load.
But even then, I would apply them across the output of the driver, and not across each individual diode.

[robvdv]

Okay, dank je wel, Bart. Groetjes uit Kaapstad.

[kecked]

Could not upload the images.

[Badpip]

My setup is as follows:
1 x Badpip quad driver running off a 24V 9A supply
4 x M140 blues in series per channel on two channels (8 total)
4 x P73 reds in series per channel on two channels (8 total)
CM106 like sound card and correction amp

I've managed to blow a blue diode on each chain of 4. I'm not exactly sure how or when I did it. All diodes have a 47mF and 10nF capacitor plus a reverse polarity diode in parallel. Both blown diodes are on the negative side of the series.

When I turn on the power supply the lasers flash brightly for a few hundred milliseconds before settling down to the brightness determined by their modulation voltage. I'm not sure if this is a normal thing with the Badpip driver? I would have assumed a slow ramp up would be more sensible. Is there any kind of soft start built into the driver that anyone knows about? Would a voltage surge from the 24V power supply be blocked by the driver?

Loose the capacitors - what you descripe sounds like an RC-constant playing tricks on you. As others have said, the drivers are expecting a resistive load.

Surges from the power supply will be handled by the LC filter formed by the power supply wires and the input capacitors.

Additionally, if I put more than 5V signal voltage through the driver would a greater than configured current go to the diodes, or is there a ceiling? (I suppose I should test this myself, but asking anyway).

Inputs are clamped to 5V (+/- a couple of percent)

The most plausible explanation for me blowing these diodes is (stupidly) unplugging and replugging the leads between the driver and the diodes while the driver was powered, but I just want to get as much information about possible about this. I'm kinda eager to avoid further damage.

This will certainly kill diodes. With no diodes connected, the driver is trying to push current through an infinite resistance. This causes the output voltage to go all the way to your input voltage. This will then discharge through the diode, when you connect it.

/Thomas