idea for ultra fast power level setting

Electronics related to CNC

Re: idea for ultra fast power level setting

Postby steppenshoe » Thu Oct 27, 2011 1:34 am

So by new scheme you are saying the DSP still does not cut it(a pun) and better can be achieved.

I would think the guys that make the DSP would have the resources to do as you suggest to make the engravings better.
Last edited by steppenshoe on Thu Oct 27, 2011 3:11 am, edited 1 time in total.
steppenshoe
 
Posts: 83
Joined: Sun Oct 16, 2011 3:59 am

Re: idea for ultra fast power level setting

Postby steppenshoe » Thu Oct 27, 2011 1:35 am

Double sorry
Last edited by steppenshoe on Thu Oct 27, 2011 3:10 am, edited 1 time in total.
steppenshoe
 
Posts: 83
Joined: Sun Oct 16, 2011 3:59 am

Re: idea for ultra fast power level setting

Postby fullspeceng » Thu Oct 27, 2011 2:29 am

Hello self-builders! It's been a while since we posted but we are always interested in what people are doing to improve the world of laser engravers. Here's some useful information:

Here's a page out of a 50W RF Universal laser OEM module. Notice 50us on delay before the laser even does anything then 126us delay before it reaches full power = 176us on delay.

Here's a measurement we took with our 10mhz high speed optical detector designed for 10.6um of a typical DC glass CO2 tube. This is a 5ms pulse (yellow). ON=LOW, OFF=HIGH. Notice it takes 500us from totally off to hit the peak then an additional 1ms before the tube stabilizes at stable power. High peak initial power is from plasma ignition which stabilizes after overcoming the initial resistance barrier.

After doing many pulse waveforms of various on/off times, we conclude a rise/fall time of around 500us to 2ms on most typical glass tubes/power supplies depending on past history, temperature, power supply, etc.

Let's do some calculations:
So 1000dpi at 30ips=30,000 on/off pulses/sec= 33us per pulse. But typical reaction time is 500us... so power must be averaged over several pixels. You may also know that dithering + averaging filter = gray.

We allow you to come to your own conclusions given these experimental results.

Don't believe everything sales people tell you :)

Best of luck in your project!
Attachments
universalRFlaser.png
5msPulse.png
fullspeceng
 
Posts: 38
Joined: Thu Dec 17, 2009 10:13 pm

Re: idea for ultra fast power level setting

Postby lasersafe1 » Thu Oct 27, 2011 2:48 am

Personally, I'm waiting to see prices come way down on RF lasers now that the original patents have expired for SYNRAD & company. I understand that FullSpectrum will be making some, or at least I hope. The controller is a whole other issue. The 8 bit grey scale controllers aren't cheap, but what is the big deal. It's a printer for gods sake! Apply X shade here and XX shade there during the sweep of the print head. Of all people on this site, Henry at FSE will likely be the first to nail the true grey printing issue.

For now I'm going to stick with what I know how to do with my existing system until something better comes along that is affordable. I see what you are hinting Henry. Very interesting indeed, but hard to control.
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

Re: idea for ultra fast power level setting

Postby steppenshoe » Thu Oct 27, 2011 3:25 am

So wait, this means that 8 bit gray scale is impossible with the lasers we use no matter what we do, correct. So companies like Kern that tout their system is nothing more than marketing because most peoples lasers can not take advantage of it at all. They are just to slow , no matter what we send the tube the tubes just can not react fast enough. It seems the systems we have now just about take advantage of everything we can get out of these CO2 laser tubes already, correct?
steppenshoe
 
Posts: 83
Joined: Sun Oct 16, 2011 3:59 am

Re: idea for ultra fast power level setting

Postby lasersafe1 » Thu Oct 27, 2011 3:53 am

No, Kern uses RF lasers, that's one of the reasons they cost $25K and up. And I believe that the hint Henry was saying is that you can get effective grey levels if you control the length of the pulse on a DC tube because it has such a slow rise time. Perhaps I'm wrong. He is laughing at us noobs trying to figure out what he has known all along. His data proves something else to me that I hadn't thought of. I have always noticed a quality difference with 30ips raster scanning when I choose X-unipolar over X-sweep. I have always attributed it to the slightest amount of belt slop, but now I realize that if there is a delay in firing the laser, then X-unipolar will always keep the delays lined up from left to right. When I go X-sweep, they work against one another and make a fuzzy image.
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

Re: idea for ultra fast power level setting

Postby steppenshoe » Thu Oct 27, 2011 4:41 am

If the Rf has a delay of 176 us that is still a huge amount away from the 33 us, it seems RF can not get the full potential out of the 8 bit gray either. The lag is less than Co2, but still way to slow to keep up with the speed in the example he posted of 1000dpi at 30 ips. So strictly from this data Kern is over selling their software though the result of the RF laser will clearly be better..

Maybe I should have said the software that companies try to sell saying they are 8 bit can not really work to their potential for the Co2 lasers even though they claim the software is better, it's overkill for a Co2 laser.

It seems we want to do anything to keep away from patterns the eye can see so would sweep be the preferred method or do I have to go on a case by case basis.
steppenshoe
 
Posts: 83
Joined: Sun Oct 16, 2011 3:59 am

Re: idea for ultra fast power level setting

Postby lasersafe1 » Thu Oct 27, 2011 12:05 pm

It doesn't matter if the rf still has a delay. Smart software can account for delays as long as they are consistant. You just fire the laser sooner than you would otherwise call for it. What really matters is whether or not you can predictably turn the laser on in a fixed time at 21% power or 25% power or 82%, etc. Obviously with rf you can control power very predictably and because they are kept in a plasma state with the tickle pulse, you don't have the issue with a power spike at startup.

I seem to recall seeing very good 3D engravings from the Kern laser on their web site. They must be doing something right. And Gantryco, the image provider uses a 400w laser, obviously RF, to check their image results.

So I am a bit puzzled about Henry's statement about not believing the sales people, unless he is talking about companies selling DC tubes. RF lasers do have power control. Obviously the DC tubes have very limited power control.

http://www.kernlasers.com/3d_laser_engraving.htm
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

Re: idea for ultra fast power level setting

Postby r691175002 » Thu Oct 27, 2011 5:00 pm

My point is that grayscale and dithering are really the same thing.

Consider: All higher end lasers (such as the synrad firestarters) use exclusively digital control. Using a grayscale control program, a 128 gray would be produced by pulsing with a 50% duty cycle. This is identical to the behavior one would see with dithering.

Regardless of what kind of grays you feed into any controller at some point in the process that grey is becomming a series of pulses.
The main differentiator here is the speed of the pulses vs the rise time of the laser. Naturally if you are pulsing faster than the rise time the pulses will blend together. I would argue that we are already pulsing faster than the rise time on our chinese tubes so technically we are already grayscale engraving anyways.

I am almost 100% sure that the cheap power supplies we use do not have an rc filter behind the TTL input. That would be very strange circuit design and they would behave differently since the current TTL input is very clearly digital.
r691175002
 
Posts: 233
Joined: Mon Dec 13, 2010 5:05 pm

Re: idea for ultra fast power level setting

Postby bdring » Thu Oct 27, 2011 5:10 pm

My power supply is analog or TTL PWM on the "input". I assume it does have a PWM on that circuit. The TTL enable is definitely a digital (on or off) circuit.
Bart
"If you didn't build it, you will never own it."
bdring
Site Admin
 
Posts: 2966
Joined: Sun Nov 22, 2009 7:33 pm
Location: Chicago, IL, USA

PreviousNext

Return to Electronics

Who is online

Users browsing this forum: No registered users and 1 guest

cron