I agree that the chances of a perfect first harmonic match between blanking modulation and pattern scan speed are slim, but they will occur...
And I suppose a few edits on the frame would be enough to change the draw speed slightly, thus causing the blank spot to move again. But that sounds like a pain in the rear. (Maybe I'm just lazy...)
Are you going to try a real-world test? (Break out the breadboard?) If so, keep us informed of your progress!
Adam
Sorry to reply to my own post, but with the new information Bill Benner posted about scanner speed, I think it's time to re-think this whole thread.
Bill mentioned that for scanners tuned to 30Kpps, your maximum small signal bandwidth would be roughly 2.5 Khz. Ok - that's a far cry from my original estimate of 10Khz. So, assuming that we still need our pulse-width modulation frequency to be at least an order of magnitude faster than our small signal bandwidth, we take our 2.5 Khz and multiply it by 10, and suddenly we're talking about a 25Khz PWM rate, which is well within the limits of a diode laser that supports TTL blanking at 30Khz.
You've still got the first harmonic match to worry about, but for sure you don't need to buy a new diode driver to make this work...
I dunno, Eyecage... You might be on to something here.
Adam
Hi Aaron;
If Eyecage's idea works, it would be an external circuit, not a plug-in. It would take the analog signal from the output of the controller and convert it to a series of TTL pulses that would then be sent to the laser driver's TTL blanking input.
Adam
O, I C, I thought it would be software...
I like the idea of a portable hardware gizmo, but software is a good idea.
if you were controlling blanking and galvos, I think you could eliminate the dark areas altogether by automatically drawing the dark part of the lines outside the current pattern, then resuming from the point where blanking began. You would have an even trade-off between brightness and frequency - you produce less brightness by sacrificing pps. But no dark patches! Right?
Another thing to consider (whether hardware or software) is only attempting to produce a limited number of brightness "levels", since any increase in the number of simulated levels is beneficial.
With an RGB setup, TTL can only produce 8 combos (including black) - 7 actual colors.
This number (8) is the number of brightness levels attainable (off + on = 2) raised to the power of the number of lasers in the system (3).
If you just simulate one more brightness level, this gives you 3 attainable levels and 27 combos.
Go up to 8 levels and you get 512 combos - a pretty decent palette.
What?...
Actually, I like the hardware idea. Maybe make something that could work all three lasers in one little box. You plug your laser TTL wires into that and the analog hook ups from the ILDA plug into that also...
Well, I just tested several lasers that I have around here on a signal generator
capable to several tens of megahertz and watched the effects with a scope and my eyes.
What I found was what I pretty-much knew but wanted to share.
Since many of these lasers have different freq responses and effects of brightness vs. frequency , none were the same unless they were the same model. Diode lasers were predictable but the DPSS were not.
I seen a difference of maybe 50 to 80 percent brightness using the same frequencies!
The DPSS lasers actually dim on there own with increased frequency even beyond what we use for lasershows..
And for almost no light or close to off to simulate the lower end , the duty cycle for recovery of the crystal set is too slow to come back to almost full brightness when asked to do so. Like at the next pulse of long duration compared to the previous of a very short duration.
However, With that in mind, the level of brightness change we are looking for would have to be so great the effect from no light to almost 50 to 75 percent.
There is no such thing as 100percent as we are modulating the diode. And even we could change the duty cycles and durations it would be quite a trick to control each of the three lasers with software to do this.
Still, an interesting concept
It would have to be done with the lasershow s/w otherwise it wouldnt know when to do it.
but I feel this would only eat up precious CPU time while trying its best to do a lasershow while trying to do all the conversions.
A simple way of frequency to voltage conversion would only work if the duty cycle was somewhat constant (50% duty cycle) as the Freq to Voltage converter wouldnt be able to recognize the difference in freq. with differing timing between pulse durations and duty cycles.
Hope that made sense
Last edited by marconi; 06-05-2007 at 18:27. Reason: eccentric spelling problems
"My signature has been taken, so Insert another here"
http://repairfaq.ece.drexel.edu/sam/laserfaq.htm
*^_^* aka PhiloUHF