M3 stops Pokeys and Auggie

[GlennD]

I got most of my honey do list for the weekend done early so I thought i would hook up the K40 to the pokeys, while i was waiting for the parts for the bigger table.  I already had changed to a DSP so the change was easy to do.  I didn't plan on doing this but can't leave well enough alone.

I don't have a wired network out in my work space so i connected up to the laptop via usb. MSI laptop Win7 64bit home edition.
  I can run the thing around, home, and do pretty much everything.
I haven't checked if i have the step count is 100% right but I have the motors running ok with the test code and the distances traveled are visually close.  (Used the step value I saw for one of the smoothie boards sites).

I don't have the laser hooked up was checking output first.  However every time  I run the M3 Auggie disconnects from the Pokeys.
Pokeys did an update yesterday afternoon i believe it was 4.1.61 but don't remember 100%.
I don't remember the version of Auggie but I downloaded Gearotic yesterday around lunch time.
I haven't licensed Gearotic on that PC yet.

I am curious on which pin out of motor 8 i use to drive the firing of the laser.  I was thinking it was the step pin but not sure now since I can't find it mentioned.
I multiplied the current Pulse frequency value originally at 5khz by 4 since the CO2 runs at 20khz.
I am using PWMpin 2 for the PWM to set my power level so changed that value from 4 to 2.

What else should I look out for?

Thank you
Glenn

[ArtF]

Hi Glen:

  The way I have it set up, Axis 8 controls the PWM, you dont use its step or dir, but the PWM pin selected by the script. PWM#2 in my case. ( Pin 3 of the lcd connector. ).

  While the step/dir likely still run on axis 8, ignore them, control is actually that PWM pin.

Now, I don't use any firing logic other than pwm, a pwm of 1% is tickle, so no radiation comes out. This
is typical of RF lasers, there is no fire TTL, just pwm.

    Glass lasers, with high voltage supplies are up in the air as to how to control them. There is little
published. I have however, done some reading. At least one manufacturer of high voltage supplies
says to use 20Khz PWM, and leave the TTL input  floating.  The PWM actually is changed to a
voltage average internally so when it goes to zero, radiation stops. This would make it work the same
as my RF laser does. Turn on time is widely reported to be all over the place, from claims of 10us to
10ms. I believe from testing on my workshop one that its much faster than 10ms, but perhaps it varies..

The hookup with no TTL makes sense, normally the PWM input is simply fed a voltage from the control
board, and the TTL then HAS to control it, but if what Ive read is true, PWM alone is enough.

  Let me know how any experiments work out. I can make up a signal control of some sort if it turns
out to be needed, we just need to figure out what IS needed. My RF laser is working awesome, better
than Id hoped, so Im very interested in how you make out. Id set your master PWM on screen to
25% or so, and do an M3 , and see if the laser is happy doing nothing with the 1% minimum,
then see if it triggers when you do a g1 move..

  I can give you access to the min and max pwm settings, I haven't added a hook to them yet,
Ive been waiting for a high voltage co2 to hook up so we can figure out the best protocol..

  One idea for the TTL is to use it as I use my Laser button on screen, it activates relay #2,
which turns on my power supply, you could instead hook that to give 5 volts to the TTL input
with a pulldown resistor to kill it if you kill the relay. The ideal situation is to be able to
turn it on, have it sit not radiating when still, and have the pwm properly trigger it as it moves..


Art

[GlennD]

Art

Gregg Holt over at SawMill creek has mentioned these DC CO2 laser require the TTL signal.
The PWM is used to set the power put the TTL signal fires it.
Does look like it is pull down since it is currently 5v on the PS.

http://www.sawmillcreek.org/showthread.php?203722-PWM-setting-on-Chinese-lasers
  See #7

I included a picture of the wiring diagram for the DSP that I have. 
The RF wiring for it shows exactly what you are talking about.

To reiterate your thoughts just pull that TTL low, leave the PWM value low and only bring it up when you want to fire.

I will give that a try tonight. 

If I can get the darn thing to stay connected with M3 active.
If I can't get beyond that it will be a few weeks for me to get back at it.


Glenn

[ArtF]

Hmph.. I would have thought TTL high would be fire.. However, yes, good test,
just to see if PWM alone can fire it.

Im not sure why the disconnect on M3, Id think the only thing changing is the
PWM rate ..

Ill give it some thought, preciate the tests..

Art

[GlennD]

Sorry for muddying the waters earlier I should have stuck to one question.

Here is the log file for the M3 issue I am having.
I could run the M3 command no problem while in my office.
The version of Auggie is 1.44
The version of Pokeys is 4.1.61

Not sure what I am missing for a setting.  

The differences between the two set ups(office versus my workspace),

PC versus laptop, laptop runs the test program in Auggie and machine responds fine.
Only when I try to run a m3 do I get an issue.  .

Connection network versus USB.

It did update the Pokeys from 4.1.60 to 4.1.61 when I installed Pokeys on the laptop.

Like it just quits talking to the Pokeys.  The buffer bar quits moving and If I hit estop button several times the buffer load bar goes to one green and can't get it back until I restart Auggie and power down the Pokeys.


Glenn

[GlennD]

Alright let this be a lesson for all involved.
Don't do as I do.
I wanted to copy the values from X to the Y axis.
So I thought not really using the rest of the axis so copy all axis will work.

It also copies the soft limit check.

Palm on fore head. 
Glenn

[ArtF]

No errors showing up in th elog by the look. Was it the softlimits kicking you somehow?

It may be the libraries have changed, Ill update in the morning to a new compilation to
see if its connected. Odd that its m3 only...

Art

[GlennD]

Yep set softlimits to false on axis 8 and it works.
The TTL does need to be pulled low for the laser to be on.

Glenn

[GlennD]

Ok so PWM worked.
Wire the TTL through the Laser power relay you have set up.
One side of the relay is to ground.

I am thinking change that and put in something in the spindle on/off to turn another relay on or off.

I used stock settings in Augs, I only set the output size to 150 in X.

Started at 47% laser power then put it down to 28% on the laser power after a little bit of running.
Not sure if it picks up the changes I am guessing it does because it didn't seem to be burning as much.

I had 2000 in the feed rate but it sure seems slow.  Run time 50 minutes
Every once in a while the motors makes a noise and it kind of jumps but as you can see it didn't appear to loose steps.

I cant rule out the laptop it wanted to do more updates.


Glenn
The extra line in the photo is because this was a scrap piece of wood that I was testing an LO-R5/RDworks with.

[ArtF]

Awesome...

First laser image other than my own. Good quality as well. So it seems your fine at 1ms control, since I can see 4 grey
levels in there, ( and thats all there is in that image.). This means your laser is running a fast PWM just fine. If 28% did
that at 2000, you can probably do 6000 easy at 90%..

  The system shoots that selected 28% only when the axis is doing 2000feedrate you selected. And it is realtime, it adjusts as
you do.

Small steps though.., great job, I appreciate the confirmation, this photo shows a lot about how the
process can work. TTL can be left on, and PWM will control on/off fine.

Thx Glen,
Art