10 kHz analog vs. 30kHz TTL modulation

[Zoof]

Hey group,
I just discovered this forum - great to see so much info on this hobby!

For a long time I've been playing with home built DAC, speaker scanners and homebrew galvos.
Now I'm looking at putting together a RBG projector with some real components.
This question is on the choice of lasers - whether to get 10kHz analog or 30 kHz TTL modulation on blue and green DPSS. Looking at the 50 mW range for blue. Scanning will be around the 30kpps mark. Analog would be preferred but is 10kHz fast enough for 30kpps scanning?

thanks.

[Pangolin]

Hi Zoof,

You pose an interesting question, and you are not the first person to ask such a question. The problem is, we don't know HOW these laser companies have spec'd their lasers. We don't really know what the heck they mean by 10kHz and 30kHz respectively. I mean, is this the -3dB bandwidth? Is this the true maximum rate? What happens if you feed in a modulation signal that is 11kHz or 31kHz respectively? And are these specified with a 50% duty-cycle? So you see, without more information, these specs from a laser company are pretty much useless...

There is other confusion over exactly how scanners and scan speeds are specified as well. For example, we euphemistically use the term "30K", but this doesn't really mean 30kHz. This means that the points are coming out at 30,000 points per second. There is a feature in the ILDA test pattern (a central circle), and this feature has 12 points in it. This feature must be in a certain relationship with another feature of the test pattern (a central square) to be in compliance. So therefore to find the actual bandwidth of a scanner capable of going "30K", you divide by 12. So the -3dB bandwidth of a 30K scanner is really 2.5kHz.

Now, with that in mind consider this. In the bad old days of laser shows, many people used scanners for blanking as well as scanning. So in other words, the scanning speed and blanking speed were the same (if you have 30K scanners, this means 2.5kHz bandwidth for both the scanning and blanking). So this implies that if the -3dB bandwidth of your laser is really 10kHz, then this is four times as good as if you were to use scanner blanking. Thus, the implication is that 10kHz should be sufficient.

But again, all of this discussion only means something if their spec of "10kHz" is really the -3dB bandwidth of the laser...

Best regards,

William Benner

[Zoof]

Hi Zoof,

You pose an interesting question, and you are not the first person to ask such a question. The problem is, we don't know HOW these laser companies have spec'd their lasers. We don't really know what the heck they mean by 10kHz and 30kHz respectively. I mean, is this the -3dB bandwidth? Is this the true maximum rate? What happens if you feed in a modulation signal that is 11kHz or 31kHz respectively? And are these specified with a 50% duty-cycle? So you see, without more information, these specs from a laser company are pretty much useless...

There is other confusion over exactly how scanners and scan speeds are specified as well. For example, we euphemistically use the term "30K", but this doesn't really mean 30kHz. This means that the points are coming out at 30,000 points per second. There is a feature in the ILDA test pattern (a central circle), and this feature has 12 points in it. This feature must be in a certain relationship with another feature of the test pattern (a central square) to be in compliance. So therefore to find the actual bandwidth of a scanner capable of going "30K", you divide by 12. So the -3dB bandwidth of a 30K scanner is really 2.5kHz.

Now, with that in mind consider this. In the bad old days of laser shows, many people used scanners for blanking as well as scanning. So in other words, the scanning speed and blanking speed were the same (if you have 30K scanners, this means 2.5kHz bandwidth for both the scanning and blanking). So this implies that if the -3dB bandwidth of your laser is really 10kHz, then this is four times as good as if you were to use scanner blanking. Thus, the implication is that 10kHz should be sufficient.

But again, all of this discussion only means something if their spec of "10kHz" is really the -3dB bandwidth of the laser...

Best regards,

William Benner

Hi William,

I see your point, thanks for that insight. Before I just assumed that the 10kHz was the frequency of a sine wave blanking input that the laser could follow with 'any' accuracy. But there is really no ground for that assumption.
However, assuming that 10kHz is a measure of some variable and laser manufacturers are consistent with that measure, would 10 kHz blanking be fast enough to scan the Ilda test image at 30kpps? I guess I'm asking for some emperical data from people here having experience with Chinese lasers.

cheers,
Matthijs

[buffo]

Bill's point is valid; since we don't really know what the manufacturer's are claiming when they state 10K blanking, it's hard to do a comparison.

HOWEVER... Assuming that they all use the same definition, even if it is incorrect, then we can make some comparisons based solely on our experience. And while I have no evidence to back up my assumption that they DO all use the same definition, I can say that lasers I've seen that were advertized as 10K blanking-capable started to show tails at between 12 and 15kpps speeds.

Astroguy has a blue laser in his projector that supports *analog* blanking to 20K, and it starts showing tails on some files at around 30kpps speeds. So a very dirty rule of thumb would be rated blanking speed times 1.5 = max galvo speed.

Note that this is only on *some* ilda files. I'm not familiar enough with the ilda test pattern itself to make comments about it compared to other files, but doesn't it have multiple blanking points built-in to it? (Someone may be able to chime in here.)

Then too, there are software settings that you can tweak in order to get rid of the tails. (Color change look-ahead, blanking lead, etc...) But it can be a problem with some files.

For the record, however, Fred can display the test pattern at 30K with no tails, even with that laser he's got that only supports 20KHz analog blanking.

I would say that for best performance, try to match the manufacturer's rated blanking speed to the speed you want to run your galvos at in kpps. And when it doubt, faster is better. (At least for graphics, that is... Beamshows aren't nearly as critical.)

Adam

[dave]

I can say that lasers I've seen that were advertised as 10K blanking-capable started to show tails at between 12 and 15kpps speeds.

Showing tails is not a issue to do with blanking speed. Showing tails in a scanned image, means the blanking shift in the laser software is not set correctly for the scanner speed.

The problem you are going to experience with running a dpss at a higher blanking speed than its designed for, is decreased brightness

[Pangolin]

Hi Zoof,

I am afraid there are only two possible ways to get your answer. First, you could purchase this laser and just "risk it"... Second, you can try to find someone who has this very same laser, or a similar laser from the same manufacturer, and ask them.

For what it's worth, I do believe there are people who are successfully using lasers with claimed "10kHz" bandwidth, but I am not sure if it is the same manufacturer...

Bill

[buffo]

Showing tails is not a issue to do with blanking speed. Showing tails in a scanned image, means the blanking shift in the laser software is not set correctly for the scanner speed.

Dave;
Adjusting the blanking shift in this case just moved the tails from the end of the scan to the begining. (IE: The tails went away, but then the start of each line was also dark.) I guess if you had really good software that had separate ofset settings for blanking on and blanking off, so you could make one lead and the other lag, then it would work... But then that assumes that all your ilda files have the exact same number of blanking points too...

The problem you are going to experience with running a dpss at a higher blanking speed than its designed for, is decreased brightness

That is a completely separate issue, and yes, blanking-induced power loss is also a problem with many lasers made in China. (I have that problem with the laser installed in my Alphalite projector.) But I also have problems with tails. And if I dial back the scanner speed to the point where the tails go away, I can still end up with the blanking-induced power loss issue if I try to display an image where every other point is blanked, for example. (There are pictures in my gallery that illustrate this effect.)

Adam