TTL/Analoge drivers?

[dkumpula]

Lots of great responses and detail in this thread. One thing more regarding having drivers that are rated for much higher current than the diode; Adam said it was not ideal - a common analogy is using a sledgehammer to drive a nail.

Specifically what occurs in this situation is you have a driver/diode combination that doesn’t respond as quickly or accurately as you would want, making it challenging to get accurate low power response resulting in whites fading down to one of the colors or having one or more of your diodes flickering when at lower levels. Like most lessons in life, I learned this the hard way. . 😂

David

[lasermaster1977]

... some vendors use a capacitor to tweak a PWM-output to "cheap anlogue" - it's simply "averaging" the PWM-pulses with the capacitor, so the pulses and pauses between average to a (rippled) voltage, defined by the PWM ratio.


Viktor

Yeah, that's like using mineral oil as a substitute for water.

[buffo]

... some vendors use a capacitor to tweak a PWM-output to "cheap anlogue" - it's simply "averaging" the PWM-pulses with the capacitor, so the pulses and pauses between average to a (rippled) voltage, defined by the PWM ratio.

While this doesn't work very well (for the exact reasons you explained), the other, more immediate problem is that most laser show controllers do not output a pulse-width-modulated color signal in the first place. So even if you have a driver that supports PWM input on the modulation line, your laser controller likely won't be able to send such a signal.

So yeah, buying a laser diode driver that supports true analog modulation is always the right move.

Adam

[lasermaster1977]

your laser controller likely won't be able to send such a signal.
Adam

...unless "your laser controller" is "his" laser controller, his solution for XY DACs, etc.

[buffo]

...unless "your laser controller" is "his" laser controller, his solution for XY DACs, etc.

I don't think he has a controller yet. He mentioned that he was just getting started and was shopping for diode drivers to build a projector. If you were planning to suggest a controller to him that uses PWM for color, I'd be interested to see what it's capable of.

To be honest though, I can't think of a single laser show controller - be it a commercially-available product or a home-build project - that uses pulse-width modulation to approximate analog color modulation. The only time I've ever seen PWM color control used in a projector is in some low-end stand-alone projectors from China that had their own built-in pattern boards and only ran in auto mode. (And importantly, they could not be controlled by a computer.)

You could, of course, build your own laser show controller that uses PWM for color control, but 1) it would give you poor color control, 2) it would cost nearly as much as a proper analog color implementation, and 3) building a controller from scratch is hardly a project for a beginner in the first place.

By far the cheapest means to get computer control of a projector is with a modified sound-card DAC and some free laser show software. And of course, a sound-card DAC will support analog modulation.

I stand by my initial advice: any laserist building their own projector should insist on analog modulation support for all laser diode drivers. If you don't, you will be disappointed later on. The cost difference is trivial (at most $10 per driver) and the performance difference is *huge*.

Adam

[lasermaster1977]

I don't think he has a controller yet. He mentioned that he was just getting started and was shopping for diode drivers to build a projector. If you were planning to suggest a controller to him that uses PWM for color, I'd be interested to see what it's capable of.

To be honest though, I can't think of a single laser show controller - be it a commercially-available product or a home-build project - that uses pulse-width modulation to approximate analog color modulation. The only time I've ever seen PWM color control used in a projector is in some low-end stand-alone projectors from China that had their own built-in pattern boards and only ran in auto mode. (And importantly, they could not be controlled by a computer.)

You could, of course, build your own laser show controller that uses PWM for color control, but 1) it would give you poor color control, 2) it would cost nearly as much as a proper analog color implementation, and 3) building a controller from scratch is hardly a project for a beginner in the first place.

By far the cheapest means to get computer control of a projector is with a modified sound-card DAC and some free laser show software. And of course, a sound-card DAC will support analog modulation.

I stand by my initial advice: any laserist building their own projector should insist on analog modulation support for all laser diode drivers. If you don't, you will be disappointed later on. The cost difference is trivial (at most $10 per driver) and the performance difference is *huge*.

Adam

No, no, not suggesting that, just that some people may use their own DAC design or have an old DAC design (like me) that predates today's technology. I agree with you entirely, analog laser diode control is the ideal solution. Back in my time we only cared about on/off, TTL did the job, and still does when followed by linear gain adjustment for each diode color.

That said, a poor man's manual RGB PWM controller for dimming only would cost maybe $5 using two 556 timers or only three 555's, one for each color, followed by a couple of LM348 quad op-am packages used to set the min and max voltage gain for each RGB channel. The 3 potentiometers for manual control would be the most expensive thing.

But like you said, it's a lot easier and you get more bang for the buck simply buying something like the FlexMod P3's.

[Shysk]

Oh, thanks to all for the replies and your time for this!,
So, i do not take in mind the mW for each diode?(With a little bit of difference)? I can configure the color balance from a software?
I will consider buying an Helios Dac, or making a OLSD one, but, while i decide and they arrive i will practice with the sd card .ild stuff.
Is there any way for making my own analoge drivers? So i can learn more about them, and, save a little bit of money for buying better diodes. With this, i also start the next answer, any good diodes recomendations?
Also, is there a limit for the projector while representing a laser show? The more points to show slower it gets?

Have a good day!!

[dkumpula]

So, i do not take in mind the mW for each diode?(With a little bit of difference)?

Buffo covered this in detail earlier, but the basic idea is to get diodes that are approximately in the right range for the ratios you want to make up your white. As Buffo also said, we all have our personal white preferences. For the record, mine is 4 parts red: 3 parts green: 7 parts blue when using single mode laser diodes.

I can configure the color balance from a software?

For most of the software packages out there, yes.

Is there any way for making my own analoge drivers? So i can learn more about them, and, save a little bit of money for buying better diodes.

I doubt you are going to save anything by making your own and the chance of you blowing up laser diodes attempting to make your own is very possible. If you don't have a large budget in the first place, I'd recommend just buying a good driver like the Simpledrives sold by BBE / x-wossee or the Lovell driver sold by Stanwax. I've used around 150 of the Simpledrives and love them. They are excellent drivers and they come in several different power ranges to properly match with your diode. They have bias control, standby-bias suppression, start up delay and overheat protection for the diode (a thermister is included), and several more features all in a tiny package.

With this, i also start the next answer, any good diodes recomendations?

Check out DTR's site. He has a good range of diodes available for fair prices.

Also, is there a limit for the projector while representing a laser show? The more points to show slower it gets?

The simple answer is your scanner/galvo speed has a maximum number of points it can draw accurately at a particular 'size'. We refer to this as deflection such as "30K points per second at 8 degrees optical deflection". In general, better quality scanners (and more expensive) can scan faster and/or wider. You can scan any number of points that you want to make up a 'picture' in a graphics laser show, but if you have more points to scan than your scanners can scan in around 1/30 of a second, then your eyes start to see a flickering image. This is less of an issue if you are making beam shows as beam show patterns are generally simpler / have fewer points.

You can learn everything you would want to know about laser scanners from this excellent book by Bill Benner for the low, low price of $0.99.

I hope you have fun exploring this fun field in the months to come!

-David

[myself]

This has been a super interesting thread, and as a laser newbie with an electronics background, I'm gonna try to clarify my understanding by re-stating things in my own terms:

Show controllers provide a voltage output for each color channel, which indicates the intended brightness of that channel. It's called "analog" because the voltage varies throughout its range (probably 0-5v in most cases) according to the requested brightness at any given point in time, so for instance, 2.5v would mean "half power". This isn't useful for driving lasers directly, because it has weak drive strength, and because diodes are current-driven. It's just a signal.

Diode lasers, like other diodes, have a "forward voltage drop" when they're operating. Varying the voltage across them is nonsensical, the voltage is basically flat even across wide variations in current. (This is the flat region after the "knee" in the I-V curve.) So the function of a laser diode driver is to take that voltage signal from the controller, and use it to control a current with some chunky drive strength behind it. At its heart, a laser diode driver is probably a big fat transistor and sense resistor for feedback, with an op-amp varying the drive current according to the signal voltage.

There's more to it, because neither the analog color signals nor the diodes themselves are ideal. The color signal may not go fully to zero during times when the laser should be off. The laser's output may not be linear over current. Lasers need overall brightness tweaking across channels to get a good color mix. The laser may overheat, and it can't protect itself. So a good driver probably has adjustments (trimpots?) to scale both the top and bottom of the input signal to match what it gets back from the sense resistor. It hopefully has some sort of temperature feedback, which down-regulates or outright inhibits the drive current with increasing temperature. It might have thermal protection for itself, too, since that big pass transistor will be operating in its linear region and dissipating a bunch of power when the laser is on.

When we buy the cheap RGB TTL laser modules, there's a tendency to think of the whole package as a monolithic unit. Four mounting screws, a power connector, 3 color connectors, done. But really, there's nothing special about this cheap 3-in-1 TTL driver board, it can be removed and replaced with better drivers. The lasers themselves are just 2-wire devices, and can be driven as we please.

Setting a new driver to work with a laser carries the hazard that, if it's set for too much current initially, the laser may smoke instantly. So you probably want to dial the driver way down (perhaps while connected to a dummy-load resistor or something), and then adjust it upward with the real laser until you reach the power you want.

But how do you know when that is? Seems to me that you'd want a cheap laser power meter (another $60 to $150-ish), and write down the power of each channel while it's driven by the factory-adjusted board, then later, tweak the new driver to hit those numbers. In this case, absolute calibration of the meter isn't important, as long as its results are reproducible from one day to the next. Keep in mind that these power meters appear to be thermal devices and probably have a slow response time, so adjustment may take quite a bit of patience!

Or instead of trying to measure the output power, actually measure the current through the laser. That's probably a lot more accurate, and faster, and uses the multimeter everyone probably already owns....

Do I have that approximately right?

[lasermaster1977]

[QUOTE=myself;347638]But how do you know when that is? Seems to me that you'd want a cheap laser power meter (another $60 to $150-ish), /QUOTE]

You start with the glass empty, then very slowly poor water into it. :-)