MUFF WIGGLER Forum Index

 FAQ & Terms of UseFAQ & Terms Of Use   Wiggler RadioMW Radio   Muff Wiggler TwitterTwitter   Support the site @ PatreonPatreon 
 SearchSearch   RegisterSign up   Log inLog in 
WIGGLING 'LITE' IN GUEST MODE

Information
Working on fixing the 'session logged out' issue - please log out and log back in as there is a new cookie name. It MAY be worse for a while, sorry. We'll see... I'm working on it. Thanks

Laser woes - help!
MUFF WIGGLER Forum Index -> Video Synthesis  
Author Laser woes - help!

johyde

I’ve been trying to make some audio-driven lissajous figures with a laser - like this:

https://vimeo.com/176440302

(only managed a short phone video)

Unfortunately, I think I’ve just blown it up. Not literally, but I’m pretty sure I’ve totally broken it. I wonder if anyone can help me try and work out what I did wrong.

Here’s what I did: I connected my soundcard (RME FirefaceUC) directly to the ILDA input of my laser . Very straightforwardly - just tip and sleeve of jacks to x+ and x- (ie pins 1 and 14), y+ and y- (pins 2 and 15), r+ and r- (5 and 18) and b+ and b- (7 and 20). For no particular reason I didn’t connect the green yet. I also worked out that I needed to connect Interlock A and B (pins 4 and 17) together to actually make the laser work.

And it did work! Brilliantly! However, after about 20-30 mins of fairly random messing about the pattern started to fragment, and what I’m left with is a laser that appears to have had a nervous breakdown - every second or so the red laser pulses and the two galvos twitch. That’s about it. None of its internal functions (patterns etc.) work (even with the ILDA disconnected). Oddly enough, if I feed it my X and Y signals they do still kind of work, but I can only get the intermittent red and the ‘twitch’ of the galvos is added to the patterns I’m sending the unit. Pretty comprehensively busted I’d say.

Obviously I’ve been an unbelievable idiot and I should have been much more careful. I really didn’t know lasers were so fragile (I’ve been doing a lot of too-late Googling and found out that they are). However, I’d really like to narrow it down to what specifically I did wrong. Here are my ideas:

1) You shouldn’t just connect Interlock A and B (unlikely I think)
2) The voltages I’m pumping in are too high (again, unlikely as I believe I strictly should have needed to amplify them - I did read Robert Henke said that audio interface signals are often hot enough though).
3) The voltages are too low (again unlikely I think - I mean I wouldn’t have thought that would damage the thing).
4) Just simply connecting an AC audio signal across the + and - terminals is bad. I believe they’re supposed to be ‘mirrored’ (ie one literally the inverse of the other) - I guess an audio signal isn’t quite that in that the positive and negative are ‘out of phase’ (if that’s the right way to put it)
5) The r g and b signals are supposed to be positive only, and I might have accidentally have fed one of them a +- AC signal (possible)
6) A short. The only possibility I think might be between two of the negative signals since these were close together on the breakout board I was using, and it was the copper strands of the audio cables connected (I did try to physically separate them and used some insulation tape, but it is just about possible. I’m absolutely sure there won’t have been any other shorts, between positive and negative signals or between positive and positive).
7) I was feeding the thing too high frequencies. I tried to play it a bit safe but to be honest I think I got a bit over excited, and I don’t really know how high I went. I think this is by far the most likely fatal error (with the x and y signals - with the colour signals it maybe shouldn’t matter?). The laser is listed at 40kpps at 4 degrees with a maximum scan angle of 40 degrees. I’m fairly sure this was the (main) thing I did wrong. I really have no idea how to derive a sensible top frequency from those figures - does anyone know?

Sorry for the long post - despite a bit of a disaster here I want to learn from it and move on!


Jasper

I'd suggest you Pm gearhead61, he's a laserhead. Otherwise I'd try asking on photonlexicon, they're quick to reply and it's a forum full of laser cats.

Sorry for the complications Dead Banana


waveglider

So here are some possible failure modes you may have inadvertently created:

1- Bipolar voltages into the RGB inputs. The color channels are only supposed to get 0 to +5v, if you put negative voltage in there you can damage the laser drivers.

2- Overdriving the scanner. The XY inputs are bipolar and accept audio signals no problem, but if your frequency is too high and you are pumping 10v p/p, you will smoke the scanner. I usually limit my scanner driving to 2khz and plenty of attenuation until I can dial in the maximum safe levels to drive the scanner.


The best thing is to wait for the LZX Cyclops module to come out in August. It has all the protection you need to use your modular to send ILDA signals to a projector.


johyde

Thanks guys - appreciate the help. Waveglider - yes, I'm pretty sure it was one of these - probably number 2. I'd used another laser that had protection for this and this one didn't, hence me being a bit under-cautious with it.

One thing - I don't think I broke the laser this way but I'm curious - what's the difference between a bipolar signal and an AC signal? Are they different at all? And will using an AC signal make the results different in any way?

I will (/am) for sure looking out for the LZX module!


mckenic

Sorry I cant help but Im watching this with much interest.
The video you posted is exactly what I would love to do - so kudos for the brilliant stuff. Hope you get sorted - fingers crossed!

Hoping the Cyclops (and a recommendation for a cool Laser) can scratch the itch for a dumbo like me!

thumbs up


johyde

I'll keep you posted mckenic!


lionelfischer

you can still damage a laser galvo system with cyclops, but the module makes it more difficult to break one. use your ears, if you can hear it squealing too much you most likely need to back off a bit on the frequency or complexity. also even at lower frequencies you can still potentially damage/"smoke" the galvos, if your doing big jumps (like square waves and stuff) instead of smoother sweeps you can exacerbate the galvos. its more involved than just using a low pass filter, use your ears and also common sense in terms of how complex the wave forms are, if you're mangling a wavetable osc or doing gated or squareish things you may be stressing out the system. you will probably break a few laser systems until you get better and being mindful of the whole thing and not breaking it, its all part of the process though.

you can also do everything thru some kind of low pass protection and slowly increase the cutoff frequency when you find something you particularly like and see how far you can push the cutoff until you think the galvos are screaming a bit too much.

you can also try bitcrushers either in hardware or software, but soemtimes those can cause big jumps for the galvos even though they are technically downsampling the signal which you'd think would make it more 'galvo-friendly' etc..

there is no "dumb" solution where you can "set and forget" like the cyclops has on its back but it certainly does help. the only way you can be sure not to really mess anything up in your laser hardware is if you use a software designed for it like pangolin software or LSX or something which keep everything kosher and optimize all the shapes your doing to work with your galvos and eliminate any jumps


Ty

Great advice. Thank you. thumbs up

What speed laser do you recommend? Obviously the faster the more $$$. By the way you make it sound - If I'm new to lasers I should get a less expensive one to "learn" on?


zilzilzilzil

Be careful with high frequencies. 40kps probably shouldn't go too high...10k has been a limit for me in the past but I wasn't using lzx. I think lasers and oscilloscopes need to be treated differently. Also, I've noticed that lasers don't like, or don't seem to like, being used for extended periods of time. I'd stay below 5k if possible.

MUFF WIGGLER Forum Index -> Video Synthesis  
Page 1 of 1
Powered by phpBB © phpBB Group