2012 DSP Feature Request : PPI

[twehr]

Guys:

I just talked to the programmer and he said that he could add this function to the LaserCad. But he was wondering what's advantage to have PPI included? In what suitation is the PPI most significant? What is the densitty of the PPI wanted? He want to understand it completely before adding this new feature to the card. Both software and hardware need to updated (through patch)

Marco,

PPI is important for handling different materials. Some materials will cut with less (or even no) charring when the PPI is reduced. PPI needs to be adjustable down to about 100 PPI.

As for density - I guess you are asking about the "spacing" of the pulses. Think of one pulse as being a complete cycle of ON and OFF, where it is ON for half the cycle and OFF for half the cycle. So, if set for 100 PPI, each full cycle would cover the head movement along a path of 0.01 inches. So it would be ON for 0.005 inches and off for 0.005 inches. That cycle repeats 100 times - giving 100 cycles (Pulses) Per Inch.

As the PPI goes higher, the length of each pulse (and the length of each space between pulses) becomes shorter. As some point, there system is no longer able to send signals fast enough to make a distinction between ON and OFF, so it becomes ON continuously. So, the maximum PPI (before going continuous ON) is HALF the maximum frequency that the system can send signals. Example - if the system can send 20,000 signals in the time necessary for the head to travel one inch along a path, then the maximum PPI would be 10,000 - that's 10,000 ON signals and 10,000 OFF signals for 10,000 complete cycles.

[Gadroc]

Nice explanations Tim.

[Engineerable]

In addition to the reduced charring mentioned above, another benefit when cutting thicker material is that it reduces distortion of dimensions. The burning is more controlled, especially in changes of direction, where the laser stays for a longer period of time due to deceleration and re-acceleration of the head.
It also reduces the extra wide kerf where the laser starts and stops.

It's important for engraving glass, in order to reduce the production of shards due to excessive heat. With PPI, the laser is blasting small pits in the glass, rather than a long trough which tends to create shards. So PPI should be available for both cutting and engraving modes.

I also believe that PPI controls heat in the corners better than adjusting the "Corner Power", because the laser is being pulsed by distance traveled, so the amount of time it takes to decelerate and accelerate in a corner is irrelevant. Anyways, I've noticed that adjusting "corner power" to the point that the corners aren't burning, will then result in inconsistent power between curves and straight lines. I think PPI should replace "corner power".

Tim, are you certain that the "Duty Cycle" spacing that you described is what you want. My understanding is that you want to pulse the laser for the shortest amount of time (1-2ms, or whatever the fastest response time is). So for 100PPI, the laser would be pulsed (turned on and off as quickly as possible) every .01in. This would create a perforated pattern, rather than a dash pattern, and distribute the heat better. The first 2 images below show what the "Duty Cycle" pulsing would look like with a laser beam diameter of 0.005in.

The images below represent examples of cutting a corner using PPI, at 200 PPI, and at 500 PPI. The important thing to notice is how the pulses allow the heat from the laser to be varied along a path, independent of the speed along the path (to an extent). Theoretically, each pulse should go about the same depth, so the part can be cut by increasing the number pulses per distance until the pulses overlap.

[twehr]

Tim, are you certain that the "Duty Cycle" spacing that you described is what you want. My understanding is that you want to pulse the laser for the shortest amount of time (1-2ms, or whatever the fastest response time is). So for 100PPI, the laser would be pulsed (turned on and off as quickly as possible) every .01in. This would create a perforated pattern, rather than a dash pattern, and distribute the heat better. The first 2 images below show what the "Duty Cycle" pulsing would look like with a laser beam diameter of 0.005in.

I have learned to never be "certain". And in this case, I am certainly open to learning from those who may have more experience. Your method may be easier to implement, as it is a bit less demanding of clock cycles - keeping the "on" time constant and only calculating the time between blips.

In a version of the software that came with a much earlier version of the card, they did, in deed, have perforation mode available. I really liked it - made great dot-to-dot drawing for the grandkids when spaced at about 1" apart. I think I mentioned that mode in an earlier post on this topic.

Regardless of the method chosen to implement it, PPI will definitely set the DSP apart from all other controllers available to the home builder. Normally, it is only available on the high end machines (Epilog, etc.).

[twehr]

Here is a great reference article that was just published today. Explains PPI better than I ever would be able to.

http://www.buildlog.net/blog/2011/12/ge ... er-system/

[Engineerable]

Hi Marco,
Has anything happened with integrating PPI in the controller. I can't wait to cut wood without charring the edges.
Thanks!

[Tech_Marco]

The PPI is not added yet. It could be possible on next release. It's Chinese New Year break, folks there want a break
Will keep the programmer pressure on the PPI issue. It's one of my job

Marco

[Engineerable]

Sounds good, thanks, and wish them Happy New Year for us!

[artofmystate]

I just want to mention I am looking forward to having this extra feature, and I am hoping there is an update on the status of this addition.

[Tech_Marco]

Just talked to the programmer and he said that he would try to get one out in about a month. He was just too busy to do it in last few month.

Marco

[baccus61]

The programmer is working for a new software after the V5.77 and it will show job time on the software. It will give an estiamted completed time and I hope this is what you guys want. He said that the software should be ready in next 7~10 days.

Regarding the PPI, I need to discuss it with him as he seems don't understand what exactly you guys want. May be someone can setup an example and I'll talk to him see if he can add that to the new software before he roll it out

Marco

This works quite well Marco and is pretty accurate.
Let the programmer know he's doing a very good job with all these alterations. The DSP is slowly evolving into a great combination of Hardware and software.
Rich.

[Tech_Marco]

Update: I've talked to Li and he said that he was working on the PPI after spending time to fix some bugs on the card and the software.

As most of you can tell, the 2012 AWC-608 have gone through a rough path as it's a new developed card for about 6 months. Now it's getting closed (very closed) to trouble free and more stable than it's just came out. Plus, software is more stable and enhancement has been done for last two months. Personally speaking, I think the card is very good now and I really don't see any major trouble even though I won't say it's 100% bug free. Since I"m not using the card for production, I may not be able to see the trouble you may have. But I have been testing for at least 30 cards so far, what the problem I seem so far is that it's more like 'confuse' other than a real 'trouble'. This is due to the lack of supportive manual for reference. First draft of English manual is ready but it still need some works before I will release it.

The PPI should be available in about a week. Plus, I've make a request that the new LaserCad to support Illustration for version CS. Currently it support up to ver 8.0 which is way behind the current version CS5.5.


Thank you for patience with us

Regards,
Marco

[Tech_Marco]

Update: V5.90 with PPI is ready. Li is doing test on the upgrade patch for the DSP card. In order to support the PPI, the mainboard need to be upgraded. I'll post both software and the patch once he completed the test.

His forumula:

For every 25.4*1000)/(2*PPI) um (distance), signal level reversed.

PPI = Pulse per inch
inch = 25.4mm
mm = 1000um


Marco

[Tech_Marco]

PPI completed and available to download from there:
viewtopic.php?f=55&t=1095&p=4030#p4030

Please try it out and report any issue to me


Marco

[bluewave65]

Works fine, but i don`t understand why is ppi setting located in that area ???????
It should be in lasercad layers section to set each layer with another ppi setting ...........

Dave