A little on my Novalux 488nm DPSS. *PICS.*

[aryntha]

My 'sweet spot' seems to be 530mA. It'll 'flash bright' at 600mA but quickly drops down, so I haven't kept it up there too much. My "start of lasing" point is 410mA. It gets brighter and brighter until 530mA, but only lasts a few minutes there. Anything higher than that, it dives very quickly.

I seem to get much better stability with the 'dumb' supply. I wish I knew which leads here were essentially power output, so I could measure the output of the 'dumb' supply.

But, yeah, I'm not spending too much time at the higher currents just in case.

 

[Trevor]

Not long at all. I seem to get "action" between i 0.41 and i 0.6 ... Hopefully we can't go high enough to blow these things.... it's weird, it seems the "dumb" supply regulates it better... I'd love to see that "Laser Lock" LED go on, and somehow have it automate the current.

Send it the cp> (constant power, as opposed to ci, constant current) command then p [desired power in watts]> and that'll happen, but if you get a bad power fluctuation, it'll ramp up the current to make up for it. And it will keep ramping it up until it reaches 800mA, at which point the driver will shut off. I think even at that power the diodes will probably be fine, since they seem to have to push pretty hard to get the crystal to lase.

-Trevor

 

[Captain Dan]

I think I feel a tear coming. That is such a beautiful wavelength!

 

[aryntha]

hmm. Got some 'checksum error' returns on "p 0.05>"

This is what I did:

cs0>
cp>
p 0.05>
l1>

the "laser lock' light blinked a bit, no output, and it went into an error state.

EDIT: Okay, I'll want p0.005 that is, heh, whoops.

 

[Trevor]

hmm. Got some 'checksum error' returns on "p 0.05>"

This is what I did:

cs0>
cp>
p 0.05>
l1>

the "laser lock' light blinked a bit, no output, and it went into an error state.

You tried to set it to 50mW...

Not sure about the checksum error bit though; I'll look into it?

-Trevor

 

[aryntha]

Have you had any success with a sequence like this?

cs0>
cp>
p 0.005>
l1>

?... Yeah, I went too high a few times here but everything seems to be OK still on the dumb supply. (fingers crossed)

 

[Trevor]

Have you had any success with a sequence like this?

cs0>
cp>
p 0.005>
l1>

?... Yeah, I went too high a few times here but everything seems to be OK still on the dumb supply. (fingers crossed)

Momentarily, but it seems to lose lock and ramp up the current out of control until it lights the error LED and turns off.

I wouldn't worry about going too high. Mine seems to have stood up nicely to the high (23mW) output - though I haven't managed to get it to be quite as solid at ~7mW as I did before that experiment.

I'll work with it more when I get home from work tomorrow, but I think I'm going to go ahead and head to bed for the night.

-Trevor

 

[aryntha]

Okay well, either way, thanks, this is a significant amount of progress it seems... Still not sure where you found these commands, but perhaps it's a trade secret

Anyhow I think this thread is going to be a good resource going forward, to say the least.

 

[Trevor]

Okay well, either way, thanks, this is a significant amount of progress it seems... Still not sure where you found these commands, but perhaps it's a trade secret

In my research on these, I've found a list of commands that mostly work. I'll post that when I get a chance; I hadn't because when sending individual signals, it would be quite easy to accidentally damage the head. Now that it seems these heads can even take max current from the driver, I'm less worried about a slew of accidental head deaths. Just... please don't kill your head. I'll feel really bad.

Bruteforcing has also yielded several more commands that do things - but are undocumented. Those are a bigger issue. ><

Lastly, I've found control software that should work - I just need to get it to properly connect and stop faulting and crashing all over the place.

Anyhow I think this thread is going to be a good resource going forward, to say the least.

Hopefully I can write some simple custom software to control these; give me a few minutes and I'll get those commands up.

EDIT: Here's the command lists.

These are for the current firmware: http://www.vuemetrix.com/support/tech/dpss_commands.html

These are for legacy firmware (more relevant to us): http://www.vuemetrix.com/support/tech/old_stuff/sw_commands.html

Current control software version can be found here: http://www.vuemetrix.com/support/

-Trevor

 

[Trevor]

Big update!

After a dialogue with Vuemetrix, turns out Novalux did indeed write their own software for these. However, no software shipped with the units. The computer-controlled power supplies we have are likely for testing the heads before loading the "dumb" power supply with the settings.

That being said, Vuemetrix's software, WinVue, does not play well with these drivers.

I'm going to start working on controller software. It'd be interesting to see about arduino-controlling these, so they're not tethered to a computer... :thinking:

-Trevor

 

[aryntha]

See, there are rumors of the correct software floating around on alt.lasers on Usenet. I haven't been able to find it. But apparently it was there.

Someone out there has to have it, somewhere.

I've got an old Win2k machine that should work better with older software, if that'll help.

 

[Trevor]

See, there are rumors of the correct software floating around on alt.lasers on Usenet. I haven't been able to find it. But apparently it was there.

Someone out there has to have it, somewhere.

I've got an old Win2k machine that should work better with older software, if that'll help.

Hmm, so that may be Novalux's original software. I'm not sure I entirely believe the person I talked to... I think Novalux probably did indeed distribute software, but I can't be sure.

Unfortunately, it's more a problem of the software not being able to interpret the output of the controller (it'll complain about the input and eventually something will snap and it crashes), not that my machine is having trouble running it. :undecided:

I guess my next hurdle is to figure out the second TEC in my head, and make sure it's working properly. Then I want to try and get a stable higher output. After that, I want to see about getting the "dumb" supply to behave better.

What's the latest on your system? :yh:

-Trevor

 

[aryntha]

I'm getting significantly better output, than out of the black "test" supply, using about 535mA. It outputs around 10-12mW vs the 7-8mW on the "test" supply. But it seems to heat up quickly and dim. No damage though.

However I still haven't been able to get any of the constant power requests to do anything but blink the "Lock" light and error out. For safety I've been trying p 0.002> to p 0.007>. But not one will light it so I assume there's probably something else I need to do , to set up the head that way.

I found the command to reset faults in there, so that's good. Don't have to unplug the supply anymore after each command. The "Checksum" error was caused by not issuing cs0>, which apparently turns some sort of checksumming off.

Here's a question, could these supplies be flashed to newer firmware -- and would we even want to do that? (Probably not.)

... I'll try to track down that Usenet copy of the original software, too, but I remember trying last year and I had no hits.

 

[Trevor]

I'm getting significantly better output, than out of the black "test" supply, using about 535mA. It outputs around 10-12mW vs the 7-8mW on the "test" supply. But it seems to heat up quickly and dim. No damage though.

I got stable output when I had it sitting on top of a HUGE heatsink. I'll post photos this evening when I inevitably spend the whole evening working on this.

However I still haven't been able to get any of the constant power requests to do anything but blink the "Lock" light and error out. For safety I've been trying p 0.002> to p 0.007>. But not one will light it so I assume there's probably something else I need to do , to set up the head that way.

I don't think these are well-suited for that... or, it's why they're on eBay.

I found the command to reset faults in there, so that's good. Don't have to unplug the supply anymore after each command. The "Checksum" error was caused by not issuing cs0>, which apparently turns some sort of checksumming off.

Sorry, I meant to pass that along to you... but now you have the list. The checksums are to catch typos (transpositions, etc.) from what I understand. It's the Verhoeff algorithm.

Here's a question, could these supplies be flashed to newer firmware -- and would we even want to do that? (Probably not.)

No way we'd be able to get it.

... I'll try to track down that Usenet copy of the original software, too, but I remember trying last year and I had no hits.

As will I.

More info to come!

-Trevor

 

[Trevor]

Okay... latest news. I got my system to stabilize quite nicely.

• The onboard power meter was reading 0.74 - which I believe is 7.4mW.
• This is the second time I've got the system stabilized at this power.
• Both times I had it sitting on a large heatsink.
• The primary TEC in my head will not heat or cool, all I can do is get a thermal reading off of it - about 35.5°C both times. Sitting on the heatsink, the pump diode heats up to this temperature then stays there.
• I had the crystal heated to 29.35°C.
• The current was at 466mA.

Here's the setup:

Near the end of the run this time, the heatsink was slightly warm to the touch, and the power dropped to a reading of 0.72 over 3-4 minutes. I think if I actively cool the heatsink, the power will be rock-solid stable. Unfortunately, this hyothetical 7.4mW is nowhere near the 15mW rating of this head, but we'll see.

-Trevor

 

[aryntha]

Trevor - did you get ahold of the software to do this? Or are you using the ci / constant current mode at 0.46? Judging by your lock light I'm guessing CI with i 0.46?

I'll give this a shot tonight too. I'm guessing "adp?" reads the current power.

Either way not bad. Not 15mW, no. But not bad for what it is.