Cannot get arc-ok in Mode 0

phillc54 wrote:

rodw wrote:

viesturs.lacis wrote:

Unfortunately that is not going to happen. We already had the discussion and they want to go as cheap as possible so using existing plasma source is a must.

Thats the problem. Really the words CNC Plasma cutting and as cheap as possible don't really go together.

I disagree, cheap as possible is the reason I came to LinuxCNC originally.

So did I and I learnt. I'm on my third Plasma cutter now. Its one of the major brands designed for what I ask of it.
At least you started with a Hypertherm plasma with a blowback start even if it was not CNC capable out of the box.

If I was building a machine for a client like the OP is, you would have to assume its intended for industrial use. There is no way I would agree with a course of action that sets me (and the client) up for failure. Sometimes the customer is not always right.

Just make sure you are paid for your time trying to get it going.

rodw wrote:

If I was building a machine for a client like the OP is, you would have to assume its intended for industrial use. There is no way I would agree with a course of action that sets me (and the client) up for failure. Sometimes the customer is not always right.

Just make sure you are paid for your time trying to get it going.

My mistake was not inspecting the plasma source more carefully. I did not take the cover off to check inside, if there are some inputs/outputs to be used for cnc control. I just relied on the fact that it is a Telwin plasma source and I have heard lots of good feedback about that brand.
The customer wanted to go even cheaper - with linear rails made from square tube and simple roller bearings. I convinced them about using Hiwin rails and better steppers/planetary gearbox solution. I could not convince them about acquiring different plasma source.
But at least I managed to get fair budget for my time that I have to spend to set the machine up.

Yesterday I was back there and I have to admit that I have messed up somewhat big time about the THCAD board. When I wrote that I specifically checked for the board to be compatibe with HF plasma source, I was right. Because it is 10V version. But it has lots of resistors soldered in, where I expected to be the voltage divider resistors (I will make actual voltage divider, no problem with that). I checked the invoice from the order and I also checked with different voltage on the board input.
First of all, good news is that the board is still working.
Second of all, I would like to ask, if somebody can explain to me, what does the "arc voltage" number on bottom left corner of Axis UI represent.
I have attached 4 pictures that show encoder velocity and plasmac.arc-voltage-out pins as well as some of the parameters and also you can see that "arc voltage" number on Axis GUI.
THCAD is set for 1/32 frequency division and encoder_scale = 1, so:
1) with 0V input I get encoder velocity = 3730, which matches 120,5/32 = 3,76 kHz
2) with 5V input I get encoder velocity = 17548, which matches ((961,5 - 120,5)/2 + 120,5)/32 = (841/2 + 120,5)/32 = 541/32 = 16,9 kHz
3) with 12V and more on input I get encoder velocity = 31250, which matches 961,5/32 = 30,04 kHz
But plasmac component reports different values on plasmac.arc-voltage-out pin and in Axis GUI.
Plasmac.arc-voltage-out matches the encoder velocity value, but I do not understand, what is that number on the screen. Can somebody, please, explain?
At 0V GUI shows 737, at 5V it shows 545 and at 12V it shows 250. What is that about?

And one more question I have is about reed relays. I was thinking about the arc-ok signal and found this:
linuxcnc.org/docs/2.8/html/plasma/plasma...ide.html#reed-arc-ok

Based on this post:
forum.linuxcnc.org/plasma-laser/35449-an...nt?start=1380#141083
I ordered 2 pieces of these:
www.lemona.lv/?page=item&i_id=c963acd47680841e49df390cf8cccbb1

I just would like to make sure that I understand correctly. 2 inner pins of that relay are for the coil. Do I understand correctly that wrapping around the cable from plasma source will create magnetic field strong enough to switch the contact? And so I should only connect 2 wires to the relay on the pins of the contact?

And one more question I have is about reed relays. I was thinking about the arc-ok signal and found this:
linuxcnc.org/docs/2.8/html/plasma/plasma...ide.html#reed-arc-ok

If you are using a reed switch like the one in the picture it is from an alarm contact. But I believe you need MODE 1 rather than mode 0

You have a THCAD-10 by the sound of it. It can only see 0-10 volts. You need to go back to the Mesa THCAD manual and add some scaling resistors in accordance to the provided formulas.

Its all in the plasmac manual as well
linuxcnc.org/docs/2.8/html/plasma/plasma...ng_the_divider_ratio

The following section (9.64) also explains how to set the offset and the scale in the plasmac GUI to correctly report the scaled arc voltage.

The THCAD is a voltage to frequency converter. Using an encoder input is a convenient way to allow that frequency to be read by Linuxcnc. The next part of the puzzle is to use the thcad calibration data to scale that back to a voltage. 3760 is your 0 volt offset, so you need to calculate the scaling value as described above.

Ok, thank you, I will try tp set the scaling and offsets and then check, if the value gets reported correctly.

Any comments about the reed relay? Will this one work for the "arc-ok" signal? (Yes, I know, that I will have to change to mode=1 for this).

viesturs.lacis wrote:

rodw wrote:

And one more question I have is about reed relays. I was thinking about the arc-ok signal and found this:
linuxcnc.org/docs/2.8/html/plasma/plasma...ide.html#reed-arc-ok

Based on this post:
forum.linuxcnc.org/plasma-laser/35449-an...nt?start=1380#141083
I ordered 2 pieces of these:
www.lemona.lv/?page=item&i_id=c963acd47680841e49df390cf8cccbb1

I just would like to make sure that I understand correctly. 2 inner pins of that relay are for the coil. Do I understand correctly that wrapping around the cable from plasma source will create magnetic field strong enough to switch the contact? And so I should only connect 2 wires to the relay on the pins of the contact?

Thank you for your patience for bearing with me about the incorrect arc voltage readings. Today I made voltage divider and tested and the reading seemed more or less correct.
The thing is that I am back at the situation described in the topic - I cannot get "arc ok" signal.
First thing I did after setting up voltage divider and plasmac voltage scale and voltage offset, was manually firing the torch and plotting the arc voltage readings in HalScope (picture "plasma-voltage-manually" in attachement). That way I obtained arc voltage for pilot arc and approximate arc voltage for established arc. It was 135V for pilot arc and 55-65V for established arc.
I also have:
1) in HAL file: setp plasmac.arc-fail-delay 5
2) in Config tab: Fail timeout = 4

At first I tried setting "OK high volts" and "OK low volts" range so that pilot arc is not included, but established arc is included in the range. Then I get "arc lost" error 0,1 seconds after starting pilot arc (pictures of test1 in attachment). That does not make sense to me!

Then just for testing purposes I changed "OK high volts" and "OK low volts" range to include also pilot arc. Then I get "arc lost" error precisely after 2 seconds. That also does not make sense to me. It is obvious that arc was not established at all in this case, but IIRC there is another error message for that. Why did I get "arc lost"? And why after 2 seconds, if I set it to try for at least twice as long?

If the pilot arc gets established, I get "arc lost" immediately also when "OK high volts" and "OK low volts" range include pilot arc voltage.

I would appreciate any ideas...

EDIT: only now I noticed that the file names of pictures are not displayed...