CIgweld Cutskill 60 - adapting to CNC use

After hunting about the web re this local brand of plasma cutter and finding nothing I decided I would start a thread to both document what I find and act as an anchor for asking questions of this community.

Background:
Cigweld is an ESAB sub-brand and they have an interesting range of manual plasma cutters called Cutskill in amperage models of 35, 45 and more recently 60 all operating off 230V single phase supply. Supply amperage ranges from 10A to 15A depending on model. They use a blow-back style pilot arc starting feature similar to Hypertherm and others. So no HF in the mix thus making CNC'ing (in theory) easier to do. Air pressure is self regulated by the cutter based on the amperage setting you dial up on its UI.
There is no CNC port on these models and given the market they are aimed at that makes sense.
Manual cutting on the Cutskill 60 that I have is very enjoyable. The hand torch has a supplied stand-off. It rips through 3mm steel and aliuminium like its not even there and savages 18mm steel. The 60A version is rated for 20mm steel but that wont be under CNC conditions unless you have drilled entry holes and programme for cooldown pauses to accomodate the duty cycle.

Current Status:
As mentioned hand cutting works extremely well and on thinner metals you really need to move the torch to keep up or dial the amps right down.

Arc Ok: successfully tested a reed switch inside 3 turns of the work lead. Proven using a digtal continuity meter and watching the reading on torch fire. Looks to be consistent.

Torch On/Off: The manual has pins for what I think is a euro style connector documented. So I know what pins the torch switch is linked to. It should be easy to hook up an external connector for CNC use to those.

Torch Height Control: Given no CNC port I will need to either (a) make a voltage divider that could be part of a cutom CNC port or (b) use a THCAD that can directly connect to the torch and work "pins" so as to directly read the cutting voltage which from looking at the manual looks to be under 300V.

Questions: I'm used to linuxCNC and Mesa cards so I guess it's really a choice between THCAD-10 or THCAD-300.
THCAD-10 with some resistors in front of it to split the voltage so as to get max 10V at max voltage from cutting? or

THCAD-300 with what would seem to be a simple direct connect with no other extra parts and then measuring actual readings and tuning settings as needed?

If I discover there is some voltage divider going on inside the machine as it does its own internal measuring for its LCD display could this impact the THCAD in some manner?

Views? Advice from real world experience? Links to other threads that basically spell this out that I have missed?

Cheers - James.

tommylight wrote:

THCAD10 and two 1MOhm resistors are much safer and easier to wire.
THCAD300 has only one "place" where it can be safely connected for use with HF plasma cutters, so complications.

...

One more reason to get the THCAD10, it might have an impact, not much and easily corrected with scale and offset.

Thanks! I was doing some reading of the THCAD manual and from an adjustability perspective was coming to the same conclusion that the THCAD-10 was probably the way to go. Gives me a lot more confidence in my thinking having you confirm things.

Fortunately my cutter is not HF, I was careful on that front and have it confirmed by Cigweld tech support. But I definately like the phrase "much safer and easier to wire"

Is the general wisdom to wire/house the THCAD inside the plasma power unit or in a plastic housing on the back/side of the plasma power unit? It would seem safer to keep all that voltage isolated away and only have a conenction to the mesa control "box" for the differential signal and +5v/GND link.

What THCAD to use and where to mount it is covered in the plasma primer. I would mount it at or preferably in the plasma cutter and use a 4 wire shielded cable to power a THCAD-10 from the control box and bring the frequency back on another pair. Then I would take your calculated scaling resistance, split it in two and put half on each leg to the thcad. The frequency should be more immune to noise and there is no high voltage outside the plasma case.

rodw wrote:

.... Then I would take your calculated scaling resistance, split it in two and put half on each leg to the thcad...

Rod - being a complete noob on electronics (I tend to lean on a mate who is a trained electrician) I take it you are are referring to the INPUT voltage into the THCAD? i.e. the resistance needed to scale the THCAD-10 to cope with input volatge up to say 200 or 300?

Re primer, found it and reading.

For a small machine, you could get away with 200 volts. The thing is over that value, we never need to know the voltage > 200. There is a thread in the plasma section with a recent plot showing what goes on in a piearce.

Thanks Rod.

Found this in plasma cnc primer. So very clear now on what that means. It is referring to an HF start model but it seems the more conservative path.

If you do not have a voltage divider and you do not have room inside the plasma cutter, install a THCAD-10 in a metal case outside the plasma cutter and install 50% of the scaling resistance on each of the IN+ and IN- inside the plasma cutter case so no lethal voltages come out of the case.

I'm going go over the plasma unit looking form more data plates and will have a look inside. I might just send a question to Cigweld and see what I can find out. Is putting a volt meter on things an option? Assuming I have or can find one that will cope with up to 300V DC?

Cheers.

So because the THCAD-10 can withstand 500 volts over voltage indefinitely, sometimes we scale them to 200 volts to get a more accurate A-D conversion in the normal range. Cut volts will not exceed 130 volts and its not really necessary to see the voltage peaks on piercing. Having said that I have mine set up with a 300 volt full scale reading and I still get accurate results. (But I do have 120 amps)

This thread I mentioned shows the voltage getting to my 300 volt full scale and flatlining during piercing but we don't care about that until we get an ArcOK.

forum.linuxcnc.org/plasma-laser/41420-arc-voltage-on-piercing

I have been looking into how to deal with touching off and based on the reading I rather like the advantages of ohmic sensing and the hyper sensing system using a THCAD-5. So I have been testing my Cutskill 60 to see if the tip is always in open circuit to the work or not. Based on the schematic attached and tests done with a meter I think I might be in luck.

You should see from the attached circuit the the path between the torch tip and the work is controlled by a combo transistor and diode (Q7). When the pilot arch is not active Q7 is open so no circuit possible between Tip and Work. When it closes a pilot arch would be possible. Assuming I understand the ideas behind hyper sensing and using relays to control a 24v supply, with the +24V on the Work, the 24v-GND on the Torch Tip, then when the torch is off the only way a circuit can be achieved will be when the torch tip touches the work on a probe.

So looking better all the time for enabling this little machine to operate under a CNC setup.

Cheers - J.

Hypersensing negates the need for any relays becasue it can handle the torch voltage without any adverse affects.
When I had my Everlast I was all set to implement ohmic sensing, then it ocurred to me it would take the full brunt of the torch voltage so was worried it could jump over relay contacts if you implemented relays for ohmic sensing. I later read you need high current relays to implement it safely.

Then PCW encouraged me to develop hypersensing and there was a lot of help from him in the background. If I still had the Everlast, I would have tried it out by now.

On my machine the ohmic voltage jumps straight to full scale once cutting starts. This was expected as the full torch voltage is on the circuit.

So if you can jury up a simple probe by taking +24V from field power, attaching it to the torch tip and then connecting an input to the material and you can probe WITHOUT cutting, then you will be fine to wire in the hypersensing circuit properly per the diagrams. You don't even need any relays, just power up the isolated power supply in startup.