Hello Guest it is March 19, 2024, 01:20:28 AM

Author Topic: Mach4 Printer Port Discussions  (Read 1243667 times)

0 Members and 1 Guest are viewing this topic.

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #40 on: June 16, 2014, 07:25:51 AM »
Tweaky:

   Thank you so much for the test. Its good to hear your result, it matches mine with the arm, it does run here much better than Mach3 did as well.
Im not totally sure why darwin is smoother, the code is totally new and has no old mach3 driver code in it other then the system internals of the
interrupt and even that was gone over and redesigned for better standards.
   Sorry for any errors, Im sure as Mach4 matures towards completion they will go away. Its a slow slog so we'll see where we're at soon. As to how to
control your laser, since Im now hooking up one of my own I think youll be presented with a method...though it may not be a method from before :), I
havent quite decided as yet. Keep up the great work, I appreciate the feedback. You have officially become the first person to ever actually cut with
Darwin, all my testing has been in motion and math only. :)

Art

Offline Tweakie.CNC

*
  • *
  •  9,194 9,194
  • Super Kitty
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #41 on: June 17, 2014, 06:04:25 AM »
My first laser test with Mach4 and Darwin (using a well trusted GCode toolpath that has been run many times before). Although this is not a method I would prefer to use the laser switching was done by using the A axis direction pin (A1=on and A0=off) and everything ran exactly as expected with no errors.

The problem mentioned in my earlier post regarding the final G00 X0,Y0 home move has now been identified as being the M30 command which, on my old and perhaps rather slow PC causes, both motors to stall during the GCode rewind. Removing the M30 cures this problem.

The ‘run today and not tomorrow’ problem I was experiencing has also been cured so it’s onwards and upwards with perhaps some more adventurous testing.  :)

You need to get that Synrad hooked up and running Art.  ;)

Tweakie.
PEACE

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #42 on: June 17, 2014, 07:00:36 AM »
Wow, thats excellent Tweak..

   Question:

     I know the synrad ( and most RF lasers) requires a tickle of 1us at 5Khz.. The most Mach3 could ever do
at 5Khz was ( kernalspeed / 5khz ) was 5 power levels at 25Khz. The synrad has a pulse former in it that cleans up the 1us pulse so it probably worked fine for a pre-lase condition..my question is "Was 5 - 10 levels of power enough for you?" Looking at the situation with mine Im wondering if Id rather have 100 levels using a small micro hooked up and commanded by Mach in some way to produce the required levels.
  In chinese lasers you have only power, but RF lasers have so much more you "could" do. A maximum of 20Khz of control could be done..and still maintain the tickle properly if a small micro was used to control the trigger in conjunction with mach.

    Mach3's engraver plugin used a dithering grey code for photo's, this stretched any pixel by generating a grey code equivalent to the grey scale of the pixel involved. Does that work real well? Would it be an improvement if the actual power changed for each pixel? I never actually tested that grey code scale, I added it on faith at the time as a bit stream that alternated in a bresenham way as a function of photo density..

 Just wondering...

Art


 

Offline Tweakie.CNC

*
  • *
  •  9,194 9,194
  • Super Kitty
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #43 on: June 17, 2014, 08:20:34 AM »
Hi Art,

For the Synrad I use an external tickle pulse generator. It is a very simple circuit and has approx 10% of adjustment for both the frequency and pulse width – this enables it to be set to just below the point of lase. The Mach PWM and the tickle pulse are then switched in or out by a separate trigger circuit (diagrams attached).

In order to get the best laser versatility (never know what we will want to do tomorrow) I think you need as many different individual levels of power as you can achieve (I believe Synrad use 5% step variation of the duty cycle between 0% and 95%).

You will have to be careful if operating the Synrad above the recommended 5kHz PRF – for example at 20kHz then a 95% duty cycle will exceed the maximum recommended internal photon density for the tube (overheating the output coupler and / or the rear mirror). The Synrad controllers software limit the maximum available duty cycle as PRF is increased thereby negating the problem.

I am no expert here but it is my understanding that the Mach3 engraving plugin allocates a set number of ‘dots’ per each individual pixel value and because for high pixel values the dots overlap each other many times during one X axis scan it produces ‘dot-gain’ and thus a shade of burn in accordance with actual pixel value. In my opinion this process works extremely well and I think I am able to get results that are up there with the professional laser systems. Overall the engraving plugin is going to take a lot to better but you are the expert and I am really looking forward to trying the Mk II.

On the other side of the coin those using diode lasers have found that DC variation in actual laser output power level in accordance with pixel value is the way to go.
I can actually do this with my DC excited laser but I still prefer to vary the output power level by varying the PWM in the same way as the Synrad is controlled.

Tweakie.
PEACE
Re: Mach4 Printer Port Discussions
« Reply #44 on: June 17, 2014, 09:27:36 AM »
Hi Tweakie

trying to follow your diagrams, is this correct? Not at all an electronics expert.
Plus when I see this, I get a little confused why you pass a single signal to both sides of an "and gate", when you already have the logic.

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #45 on: June 17, 2014, 10:31:20 AM »
Tweaky:

  I wasnt aware of the power issue, Im glad you mentioned it. Its surprising as they do allow a CW operation, youd think the power of 100% wouldnt be that much less than 95%, though I reliase its a matter of photon reovery allowing higher power at 95% than at 100%.  Ill keep note of that..

>>Mach3 engraving plugin allocates a set number of ‘dots’ per each individual pixel value and because for high pixel values the dots o

  Was it recoded since the original I wrote? As I recall I adopted a grey scaleing of the distance between pixels..but I wasnt sure if I put the code out for others to modify. ..

Art

Offline Tweakie.CNC

*
  • *
  •  9,194 9,194
  • Super Kitty
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #46 on: June 17, 2014, 11:32:37 AM »
@ Craig,

I understand what you say, however, the logic gates invert the signal - connecting another gate in series restores the state. As there are 4 'and gates' in a package it makes sense to use one of them this way rather than use a separate inverter chip.
The output shown from the tickle pulse generator is signal 1 and GND. Signal 2 is fed from the Mach PWM or a stand alone PWM generator then the output goes to the laser control circuitry. Hope this makes sense.  :-\

@ Art,

I think the Synrad CW operation may only apply to the relatively new models of their RF lasers. I don’t know for sure but I suspect that externally we can apply 100% duty cycle while internally the electronics is still pulsing at a reduced duty cycle and I would be concerned that older models do not have this safeguard. Perhaps we need to contact Synrad for more information.

As far as I am aware the grey scaling was not incorporated into the version of the plugin that you released.
If it’s any help this is typical of the shades of grey that can be achieved from the plugin as it stands http://openbuilds.com/attachments/a1-jpg.1862/

Tweakie.
PEACE

Offline ART

*
  • *
  •  1,702 1,702
  • Tough as soggy paper.
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #47 on: June 17, 2014, 11:46:02 AM »
Tweaky:

  Interesting...Ill have to take a look at the current code to see whats changed...

The J48-1 manual says wiring for CW is fine ( and seems to be here) but that you
get slightly less power than you do at 95% as the gas recovery rate is less. I dont think
its pulsing it when CW'ing.. the latest manual shows a pretty good explanation of the
method used , the waveforms seem convincing. Also, with a 10 wattt I doubt the
density gets high enough to hurt the mirrors or the lenses , though with 100 watt Id probably
worry about it. I did wonder about how they cool without water sufficiently, but then aluminum is a very
good conductor.

  I saw a photo of a J48-2 laser the other day mounted vertical, so failure in isuch a situation must be statistical,
your fine or your not..or your fine till your not. :)

  Very interesting world to play in..

Art

Offline Tweakie.CNC

*
  • *
  •  9,194 9,194
  • Super Kitty
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #48 on: June 17, 2014, 12:05:29 PM »
Hi Art,

I have been searching my PC but cannot find the link to it - The Synrad UC2000 controller data sheet gives a lot more information on Duty Cycle and Pulse Repetition Frequencies and how they should be managed - perhaps may be worth looking for it if it is still available on the Synrad site.

I have no practical experience of mounting a tube vertically I am just going on the manufacturer's advice trusting that they are correct. :-\

Tweakie.
PEACE

Offline Tweakie.CNC

*
  • *
  •  9,194 9,194
  • Super Kitty
    • View Profile
Re: Mach4 Printer Port Discussions
« Reply #49 on: June 17, 2014, 01:58:26 PM »
Just thought I should mention this...

I don’t know if this is a Mach4 v.1767 issue or a Darwin issue but for those like me, that work in metric, the motors Counts per Unit : Velocity and Acceleration all have to be entered in Imperial units irrespective if metric is selected in general config. Not a problem once you know but it was puzzling initially. :)

Tweakie.
PEACE