idea for ultra fast power level setting

Electronics related to CNC

Re: idea for ultra fast power level setting

Postby dirktheeng » Wed Oct 26, 2011 11:21 pm

The real question is how fast can we get the laser to change current with a digital pot. The amplifier cirquitry speaks voltage in the power supply. Thats why we can use PWM to set the power... inside the power supply there is an RC circuite that converts the PWM signal to a voltage the same as you get if you send a current through the pot. The problem with the PWM is it is slow in responding to the change. I am willing to bet that the laser will respond much quicker to the change of the input voltage signal via a digital pot rather than the PWM.
Last edited by dirktheeng on Wed Oct 26, 2011 11:46 pm, edited 1 time in total.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: idea for ultra fast power level setting

Postby dirktheeng » Wed Oct 26, 2011 11:26 pm

r691175002 wrote:The retina is fully capable of grayscale engravings and they have addressed it in the past:
http://www.fullspectrumengineering.com/ ... rt=30#p568

The short version is that nobody does grayscale engraving since materials don't respond as consistently and to your eye there would be no advantage (any form of printed media is dithered). Inkjets use dithering (its not like you can apply a "lighter" dot of colour, the droplet is either there or it isn't) and the visual quality is fine.


What they actually say in that post is that they handle greyscale by dithering which is not really an analog operation it's a digital on/off
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: idea for ultra fast power level setting

Postby dirktheeng » Wed Oct 26, 2011 11:41 pm

r691175002 wrote:You could alternatively use the power supply the way it was designed, namely use the potentiometer to set the maximum current and then provide a 20-50KHz PWM signal through the TTL input as specified in the manual. If you do the math, with a 1ms rise time you can shove a crapload of pulses into the power supply and even a single pulse would be enough since hopefully one would have 8 bit or higher control.


Actually, the manual says that the response time is LESS THAN not equal to 1ms. lasersafe1 measured the current response to be 15micro second, and one would have to believe that the tube can respond to a speed of at least 60 micro seconds for laser safe to be able to engrave 250dpi at 30ips. Therefore, one is left to the conclusion that the manual grosely overstates the response time and that the 1ms time is way to large. If that were the case, one would have to reduce the engraving speed to less than 2ips at 250dpi. And by the way, the frequency which the head travels over a 250dpi image at 30ips is 7.5khz, so at a standard 30khz PWM signal, you can only get 4 pwm cycles in which is not enough to change the voltage on the internal RC circuite. It typically takes 20-30 cycles for the value ot reach a steady state. There is simply no way that you can get an effective change in laser power that way within the time it takes to pass over 1 pixel.

That is why I am proposing setting the analog value via a digital potentiometer. That change in value happens as fast as the digital pot changes. The amplifier circuits in the power supply are pretty fast.
Last edited by dirktheeng on Wed Oct 26, 2011 11:57 pm, edited 1 time in total.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: idea for ultra fast power level setting

Postby dirktheeng » Wed Oct 26, 2011 11:55 pm

Anyhow, the only way to settle the issue is to buy a ditigal pot and scope the current. I will assume that the current value matches the time of the laser within a factor of 4x as demonstrated by the actual results that lasersafe1 got. I am going to ask my wife for an early christmas present and buy a scope as I am getting to the point that I actually need one. These forms are filled with much opinion and not backed by much in the way of measurment so it's time to change some of that. I'm not really interested in peoples opinion unless that opinion is backed by some measurment or observation of quantifiable fact. I'm going to get a scope that will allow me to measure and post images of traces so that these issues can be settled once and for all. I need one to start working on engravers anyhow.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: idea for ultra fast power level setting

Postby lasersafe1 » Thu Oct 27, 2011 12:24 am

It is very easy to tell the difference between wood that has been engraved by dither mode and wood that is engraved with 256 levels of grey power control. I submit for your observation the following images. The dithered laser looks like it is full of pin pricks. The power control burns to a certain depth and stops.

We had an extensive discussion about this in another thread. http://www.buildlog.net/forum/viewtopic.php?f=6&t=36&start=30
Attachments
eagle.jpg
power control
eagle.jpg (92.87 KiB) Viewed 1324 times
jamison.jpg
dithered laser
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 1:02 am

The difference is easily apparent. I only wish there was an image where both processes worked on the same exact file.

I guess for 3D I have to stick to my router than because I definitely get results like in the non dithered picture using the micro bits and time.

Plus, I would like to see the same carving done with say a DSP verse another system. For me that is what really matters. I would like to get the best a home laser can give now. Going for a system to give the results shown in the better picture is great, but if it is beyond what we currently can do I sure would like to know what is the best out there right now so I can have that while we work to get better.

I assume the best engraving right now is using the DSP with 4 drivers. If any Chinese laser can not give me the results I want with the best system I can make at home I guess than that is the time to go to a different type of laser, correct?

I have a few more questions if someone has time to answer.

Will a larger watt laser using the same motion and laser control system improve the engraving in any way? Will better optics improve the engraving? Will a non Chinese, higher quality power supply improve the engraving? Will the long term laser tube over the cheaper shorter term laser tube improve the engraving in any way?

I would like to start with the best of everything within reason. It seems a lot of work being done is on the laser controller side, but if any of the other items can improve the engraving and are currently available wouldn't it make sense to maybe get those things first?

Thanks for any comments
steppenshoe
 
Posts: 83
Joined: Sun Oct 16, 2011 3:59 am

Re: idea for ultra fast power level setting

Postby dirktheeng » Thu Oct 27, 2011 1:03 am

for lasersafe1:

I agree that doing an analog raster is going to be better than doing a dither (in the special case of getting high quality 3d reliefs). I have agreed and appreciated everything you have said. You have been extreemely helpful and presented hard evedence and information that is valuable.

I would like to know if you thingk this is worthwile to persue. My understanding, is that the PWM input has an RC circuite in it that produces a voltage that is used as an analog signal to an amplifier that sets current levels. My supposition is that the long understood "slow" response to setting power levels is due to this response. I believe that there is no inherent slow response in the amplifier/tube that causes this because the on signal allows one to get to a set power level very quickly. As you have measured within about 15-60us. I cannot think of a reason why there should be an inherent speed limitation in an analog amplifier circuits to produce a current to send to the laser tube. Therefore, my supposition hangs on the fact that it takes a long time (relatively speaking) to get a 30kHz PWM signal to settle on a voltage through a RC filter. I think it may be possible to get the whole system to respond more quickly by skipping this step and going to a digital method to control a risistor which will set a voltage quickly and hopefully eliminate the lag we have been seeing in the power change.
Last edited by dirktheeng on Thu Oct 27, 2011 1:18 am, edited 2 times in total.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: idea for ultra fast power level setting

Postby dirktheeng » Thu Oct 27, 2011 1:10 am

steppenshoe wrote:The difference is easily apparent. I only wish there was an image where both processes worked on the same exact file.

I guess for 3D I have to stick to my router than because I definitely get results like in the non dithered picture using the micro bits and time.

Plus, I would like to see the same carving done with say a DSP verse another system. For me that is what really matters. I would like to get the best a home laser can give now. Going for a system to give the results shown in the better picture is great, but if it is beyond what we currently can do I sure would like to know what is the best out there right now so I can have that while we work to get better.

I assume the best engraving right now is using the DSP with 4 drivers. If any Chinese laser can not give me the results I want with the best system I can make at home I guess than that is the time to go to a different type of laser, correct?

I have a few more questions if someone has time to answer.

Will a larger watt laser using the same motion and laser control system improve the engraving in any way? Will better optics improve the engraving? Will a non Chinese, higher quality power supply improve the engraving? Will the long term laser tube over the cheaper shorter term laser tube improve the engraving in any way?

I would like to start with the best of everything within reason. It seems a lot of work being done is on the laser controller side, but if any of the other items can improve the engraving and are currently available wouldn't it make sense to maybe get those things first?

Thanks for any comments


Well... if this idea works out we may be able to improve the quality some, but traditionally speaking, you have to get an RF tube to get true analog depth for each pixel which is really expensive... many times the cost of the entire laser system. If you want good quality relief carvings, you are better off going with a router any day... even with an RF laser, the material plays such a role in the depth you get that you have to do a lot of tests to get it to be consistent.
dirktheeng
 
Posts: 616
Joined: Thu Sep 09, 2010 4:49 pm

Re: idea for ultra fast power level setting

Postby steppenshoe » Thu Oct 27, 2011 1:21 am

Got yah, the laser will be good for some materials my router is not that great on though. Plus, I just love making stuff.

Are any of the hardware items I mentioned worth upgrading to improve the cutting or engraving>?
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 1:22 am

Both images above are essentially the same file. They both came from GantryCo.com and are undithered files with 256 levels of grey (8bit). Jamison Laser purchased the file from GantryCo, so they were then free to put their text on it.

The good engravings can be done with a Chinese tube and power supply but it needs a whole new control scheme. You would have to process the image and consider what you are going to do on line 1 of the raster scan. You would then make a pass and hit everything that has a color depth of 255. You would then pass again and hit everything that has a color depth of 255 plus the the pixels that have a color depth of 254. You would then pass again and hit every pixel that has a color depth of 255,254 and 253. Etc. Etc. Etc. Perhaps there would be some variant where you change power at the end of one pass.

Although I complain, I have indeed had some specimens turn out much better than the Jamison sample shown above. It all depends on the type of wood and density of the grain.
lasersafe1
 
Posts: 599
Joined: Fri Nov 27, 2009 8:23 pm

PreviousNext

Return to Electronics

Who is online

Users browsing this forum: No registered users and 1 guest