Newbie Intro ^.^;

[ParoXoN]

@buffo:

Thanks very much for the clarifications!

For my part, the goal of the project is to interface and control the galvos and laser with an embedded Atmel AT91EB5 board. This is convenient on the one hand because the board has its own PWM channels which are controlled by a very precise onboard DAC.

It's inconvenient, however, since it also means I have to write all of the software that will control the projector myself. I poked around for a doc and found (this and this) but the note in the second one seems to indicate that the document's contents may be deprecated.

If the above docs are accurate/close, the frame parsing doesn't look very difficult. I'm certain I could write up a uC/OSII (the OS I suspect we'll be using) process to parse the frame data into memory and then cue the driver to process to pass that on to the galvos at a given rate using internal interrupts.

Regrettably the limiting factor for this project is price I was talking with Marc about the ScanPro 30s and at the moment they're out of my alotted budget.

Just to give you all a notion of how the funding works: the University will dole out ~$100 CAD to each of us for whatever we want and the Uni gets to keep the property... I figured it would be best to use their money to buy an inexpensive set of galvos (was considering Spacelas SP20K or Phenix PT-20K for their incredibly low price point; < $100 before shipping.)
If it turns out I enjoy this (and I suspect I will) I can always sink more of my own money into a higher quality set later. My only fear is that these lower end galvos will crap out on me and leave me in an uncomfortable place.

For the purposes of my project I don't require complex imagery or even complex geometric shapes. Basic writing and simple geometric shapes should be sufficient to woo my audience at a tech demo, given that my peers are designing things much less flashy. (Bwahaha.) My main goal here is to learn enough to be able to recreate the project on my own time, with my own (hopefully higher quality) parts and the second time around produce something with greater capability.


Thanks for bearing with my newbness everyone ^.^;;

[buffo]

the goal of the project is to interface and control the galvos and laser with an embedded Atmel AT91EB5 board. This is convenient on the one hand because the board has its own PWM channels which are controlled by a very precise onboard DAC.
It's inconvenient, however, since it also means I have to write all of the software that will control the projector myself.

Ahhh... Ok, that clarifies things a lot. And it also complicates them. You're going to have a lot of work to do in the coming months! I hope you're a hotshot when it comes to coding for that Amtel microcontroller... There should be a few folks here that could offer you at least some assistance with that, but sadly I'm not one of them.

I poked around for a doc and found (this and this) but the note in the second one seems to indicate that the document's contents may be deprecated.

Use the proposed 2004 standard. It's what almost everyone uses these days. If you want more history on the subject, you can search the forums for "ILDA format 3" and get more than an ear full, but suffice to say that most frames out there are format 4 or 5 these days. If you follow the specs in that document, you'll be able to accurately display nearly all of the artwork that is available today.

I'm certain I could write up a uC/OSII (the OS I suspect we'll be using) process to parse the frame data into memory and then cue the driver to process to pass that on to the galvos at a given rate using internal interrupts.

Ok, that will allow you to put up a static frame. And if you can control the frame rate, you could even handle a multi-frame animation. Are you planning on having a set of frames pre-programmed then? Something like a set of (pick a number - maybe a dozen) frames that could be pre-loaded and then selected at will? If that's sufficient to satisfy the needs of the project, then that's good, as it sounds quite simple, and much less work than re-creating a true laser show editor / playback engine with it's complicated user interface and multiple tools for frame editing, animation control, image manipulation, timeline editing, etc...

Regrettably the limiting factor for this project is price

Ahh yes. Well, the problem with lasers is that nothing is ever cheap. This is a money pit hobby, no doubt about it. It's very hard to do things on the cheap... Still, there are a few ways to economize.

the University will dole out ~$100 CAD to each of us for whatever we want and the Uni gets to keep the property...

Wow! If that's the entire budget for this project, you're in trouble, because you won't be able to afford a set of galvos and a laser for that sort of money. (Well, you could purchase a bare red laser diode and build it into your own module, and then assemble your own constant current driver, but even if you do that, the parts are going to run you close to $40.) I had figured you'd be spending at least $100 on the laser, and around $125 to $150 on the galvos. Can it be done cheaper? Yes, but not by much...

I figured it would be best to use their money to buy an inexpensive set of galvos (was considering Spacelas SP20K or Phenix PT-20K for their incredibly low price point; < $100 before shipping.)
If it turns out I enjoy this (and I suspect I will) I can always sink more of my own money into a higher quality set later. My only fear is that these lower end galvos will crap out on me and leave me in an uncomfortable place.

I would be less worried about them failing than of them not performing well in the first place. I'm not familiar with either set, but if they're under $100, they're not going to be great. At the very least, they'll probably need to be tuned after you get them to get decent performance. (It would be great if there was a senior member close to your area who would be willing to assist you. Tuning galvos is tricky to begin with, but tuning with an unknown controller is a nightmare...)

There are a couple sets of scanners on the low end of the price curve that have been proven to be worthwhile investments. The Scaneco20's, the LSP20's, and the old DT-25's were all decent scanners for the money. Maybe the Spacelas SP20's would be good enough too. But I haven't tested them. (Admittedly, they do look similar to the ScanEco20's though...)

You still need a laser though. And you're going to have some other hardware involved in this build too. A metal baseplate, for one. Some wiring and connectors too. Is the university going to supply that stuff? Or is it all on your own dime after the initial $100? (It might be worth it to spend another $100 of your own money to make the project more solid, even if it goes to the university afterward.)

For the purposes of my project I don't require complex imagery or even complex geometric shapes. Basic writing and simple geometric shapes should be sufficient to woo my audience at a tech demo, given that my peers are designing things much less flashy.

OK, if you could produce images like this:

or this:

Would that be enough? Because that's about as complex of an image as you're going to get from a pair of sub-$100 scanners. Is that what you had in mind?

My main goal here is to learn enough to be able to recreate the project on my own time, with my own (hopefully higher quality) parts and the second time around produce something with greater capability.

If you are *sure* you want to roll your own software again once this project is finished, that's fine, but know that it's a problem that has already been solved for you several times by both professional laser show software companies and hobbyists alike.

I strongly suggest you attend a Laser Enthusiast's Meeting at some point after you finish this project. That way you can see what's out there in terms of off-the-shelf control software before you embark on a quest to write your own from scratch.

Don't get me wrong: writing software is a journey which bears it's own rewards, but if you're really into the artwork (or just the laser light!), you'll be able to get started much faster with a commercial solution.

Adam

[ParoXoN]

So, some updates:

1. My budget has been increased. I can probably get away with spending ~$200 plus whatever I'm willing to contribute personally. Bearing that in mind, would you suggest the Scaneco 20s or the LSP-20s??

2. Re: Laser: for the purposes of this project I'm limited to a class 3a laser. This will be replaced once the projector is out of the lab. (The lab coordinator is a little concerned about safety.) The plus side is I don't have to spend much on the laser atm. I'm concerned about the performance but a) this is a prototype and b) the laser can be replaced fairly easily (since I'm using purely solid state diodes, swapping one for another shouldn't be too bad...)

3. As far as the controller goes, this is the interesting part for me XD In the future I might choose to use a commercial piece of software to control the galvos/laser for me but for the moment I'm looking forward to the software/firmware side. I'm sure I'll be cursing myself for saying that in a few weeks but we'll see xD

4. Frame/animation: The plan is to have several pre-programmed files rather than writing a more sophisticated control suite. Mostly static images are fine but small animation would also be cool. If, though, at the end of the day, all I can get is simple images, that will be enough.

5. Metallic pieces/machining: I have access to a machinist via the school and can have more or less anything I want made (provided it's not mechanically complex.) I understand the basic layout of the projector, but does anyone have any specific schematics I could use as a reference? (If that's even worthwhile?)


Thanks for all your help XD

PS:
Project status:
Sourcing the galvos, working on interfacing external memory to the microcontroller.

[buffo]

would you suggest the Scaneco 20s or the LSP-20s??

Either one would be more than adequate. The ScanEco-20's are slightly cheaper though, and since cost is an issue I'd probably go with them. But both these sets offer amazing performance for the price.

for the purposes of this project I'm limited to a class 3a laser. This will be replaced once the projector is out of the lab. (The lab coordinator is a little concerned about safety.)

That's actually smart. So long as you use a green laser, the results will still be quite visible. And a 4.9 mw green is going to be very affordable.

I'm concerned about the performance but a) this is a prototype and b) the laser can be replaced fairly easily (since I'm using purely solid state diodes, swapping one for another shouldn't be too bad...)

Well, a solid state green laser is a lot more than just a diode. It's a diode-pumped Vanadate cavity with a KTP doubling stage, and there's often a TEC in there to cool the pump diode as well. Still, the entire module can be replaced without too much work, and for a class 3A unit, you won't need the TEC. You might be able to get away with a green pointer module mounted in a large heat sink, but you'll still need to buy or build a constant current driver with modulation support to make it work with your scanners. (You could use one of Dr Lava's Flexdrive drivers, which sell for around $35 or so, or you could build your own for a bit less in parts.) Ask around here and you might find someone with a low power green laser that you can pick up for next to nothing...

As far as the controller goes, this is the interesting part for me. In the future I might choose to use a commercial piece of software to control the galvos/laser for me but for the moment I'm looking forward to the software/firmware side.

As you wish. It's your sanity, after all. But yeah, at some point you'll probably want to switch to using a regular PC to control the projector. So long as you wire it to the ILDA standard, you won't have to change anything on the projector when you upgrade.

The plan is to have several pre-programmed files rather than writing a more sophisticated control suite. Mostly static images are fine but small animation would also be cool. If, though, at the end of the day, all I can get is simple images, that will be enough.

That should be fairly easy then, assuming you can handle the coding... There are enough frames in the public domain that you'll be able to pick out a few dozen that you like and pre-program them into the controller.

I have access to a machinist via the school and can have more or less anything I want made (provided it's not mechanically complex.)

That's very convenient. Will the resulting projector be yours to keep, or will it stay with the school? (Reason: if it's yours to keep, make the housing larger so you have room to add a red and a blue laser later on to get full color capability. If it stays with the school, then make it just large enough to house the green laser module and the scanners.)

I understand the basic layout of the projector, but does anyone have any specific schematics I could use as a reference? (If that's even worthwhile?)

For a basic layout diagram, have a look at this drawing in my gallery. (Click on the image for a larger version.) It's not a true schematic, as it lacks any reference to interlocks, keyswitches, or emission indicators, but it does show you essentially how things are connected.

If you're wondering how to wire an interlock, this circuit may be of use to you. Or, for one with remote and local controls, have a look at this schematic, which is how my current projector is wired.

Finally, this article on the ILDA website (written by Bill Benner of Pangolin) explains projector wiring in considerable detail, and also discusses the benefits of a single-point grounding scheme. Good reading.

Adam

[no-esc]

For what you are doing:

I have the spacelas 20K and they are good for me atm. (98 bucks) Say hi to Sam at Spacelas for me.

Pick up a DR lava Flexmod (30 Bucks)

a 5mW -50mW dpss green module www.o-like.com 20 to 30 dollars.

Have your machine shop guy build you a simple heatsink attach a 5v wall wart for lasers pwr suppy.

Figure out your wiring (I grabbed the dmx control board from spacelas for 25dollars.)

check here for pics on my own progress if you want more pics.

http://akrobiz.com/laserboy/forum/in...pic,252.0.html

Good luck.

v/r

Len

[ParoXoN]

Hey, thanks for the reading XD All extremely helpful! (Esp the Bill Benner article... So much information.)

Re: Laser:
I think I've managed to locate a convenient Class 3a green laser (my bad in referring to it as just the diode; I meant the entire module.) The reading I've done indicates this should be a 5V unit?
Regarding the power source; what exactly is the purpose of a modulator? Does that adjust for changes in the light output characteristics as the laser runs (changes due to heat, etc.?)

Re: The coding:
Keeping things ILDA-standard will be convenient; it'd be neat to hook the projector up to a regular computer later and see things continue to function.

Going to be ordering the parts this week, hopefully! Yay ^.^

@no-esc:
Thanks for the info and links XD (Though o-like.com link seems broken for me...)

[Edit]
Regarding the differential inputs:
The plan is to use a unity gain op amp in the inverting configuration (I was actually looking at the TL084, the same one mentioned in that article you linked me to) to generate the diff signals. I understand that I can keep -X/-Y at GND and then modulate +X/+Y by 0-10V but it would be easier for me to keep things in the 5V range.

For the laser power supply, since I'm using a very basic laser unit (that isn't configured for differential inputs), is single rail input sufficient? I don't see the laser as being as susceptible to noise as the galvos.

Thanks!
[/edit]

[buffo]

I think I've managed to locate a convenient Class 3a green laser (my bad in referring to it as just the diode; I meant the entire module.) The reading I've done indicates this should be a 5V unit?

Yeah, just about all DPSS green lasers use a 5 volt supply. It should be marked on the driver though. (Are you buying a complete module with it's own driver, or are you buying a CW-only laser module for which you'll need to build a driver to modulate it with?)

Regarding the power source; what exactly is the purpose of a modulator?

In order to "draw" pictures with the scanners, you need a way to lift the "pen" off the "paper" so you can re-position the scanners. This means you need a way to turn the laser on and off rapidly. (very rapidly) This is modulation, more commonly referred to as blanking. Without some form of blanking, you'll be limited to continuous line drawings only. But if you want to be able to create pictures like the peace symbol graphic I posted above, you'll need to be able to modulate the laser while the scanners are tracing out your image.

This modulation (or blanking) capability is normally built-in to the laser driver module. A driver module is typically powered by +5 volts DC, and it will have a separate input to accept a modulation signal from your controller. (This signal is normally single ended, btw, with +5 volts being max power and 0 volts being off.) You can find modules that support TTL (exclusively on or off) or analog (continuously variable output) modulation. In your case, TTL will be sufficient, but when you get into higher power lasers (and especially with multi-color projectors), you'll want analog modulation for greater control and a wider color palette.

For the laser power supply, since I'm using a very basic laser unit (that isn't configured for differential inputs), is single rail input sufficient? I don't see the laser as being as susceptible to noise as the galvos.

Nearly every commercially-available laser module uses single-ended inputs for the modulation input. Granted, the ILDA spec does mention differential signals, and there are + and - signals for red, green, and blue, but I'm not aware of any laser that supports differential color signals. (I do have a PCAOM that supports differential color signals, but it's rather uncommon.) Also, nearly all of the commercially-available laser show controllers use single-ended signals for color control.

Bottom line: Single-ended signaling will be fine for the color signal. Definitely use differential for the X and Y though.

Adam

[ParoXoN]

So the modulation is for blanking (and light intensity in general?) Got it.

As far as modulating driver goes, would it be sufficient to have a transistor switch driven by TTL controlling the voltage supply to the laser or is it more complex? (Would a TTL-5V linear amp work?) I've already got a 5V CW 5mW green laser so I was considering using that but if it's a significantly better decision to use with a built in driver/modulator I'll go with that.


Back to the galvos:
It seems LaserShowParts ships out of China. What sort of experience have people had with delivery times/stock/etc.? Are there any retailers in North America you all know of?
Also, for the ScanEco20s, are the signal connectors 2.54mm molex? It'd make interfacing more smooth for me.
[Edit]Right! Dave's associated with LSP.com XD I'll PM him [/Edit]

If I wind up purchasing a new laser, where would be a good place to order it from? (Again, preferably within the continent...) I'll poke through the forums to look for retailers, but I just thought I'd ask

[buffo]

If the driver for the CW laser you have has good constant current regulation, then yes, you should be able to modulate the laser externally with a transistor (or a mosfet even) between the +5 volt source and the laser driver. It's not a perfect solution, but for the purposes of your project it should work OK. (After all, TTL modulation is all you really need for this one anyway.)

As for the galvos - don't worry about the fact that they come from China. Dave and Adrian at LaserShowParts have an excellent reputation for getting things shipped out on time. They actually have an associate who works in China as a liason between the factory (where the goods are shipped from) and the staff in Australia (where the orders are processed).

Don't worry about the signal connectors either. The scanners will come with a set of pig tails that plug into the connectors on the amps and have bare wire ends for you to connect the control signals to.

If you need to order your parts from a stateside vendor, you might want to check out ctlasers.com. This site is run by Marc Arotsky (gottaluvlasers here on PL), and he also has a good reputation among the members here. Unfortunately, he does not carry the ScanPro, ScanEco, or ScanLSP line of scanners. He does carry the Dragon Tiger units (DT-40's and DT-30's), but they are a bit more expensive than you were looking to spend.

Adam

[ParoXoN]

Ahaha I'd taken a look at ctlasers; one day... One day XD

I'm not terribly concerned about the parts coming from overseas Just my due diligence to my lab coordinator who's a bit leery about overseas transactions (given her previous experiences with students in this lab.)
I'll hopefully place the order before the end of the week. I'm going to take the module I have to the lab tomorrow and see how it responds to a handcrafted modulator.

Drafting the precise project proposal and schematics atm. Will post when they're complete