Long and slim plasma primo

Hello there! I am at the stage where I’m deciding what the exact dimensions for my setup will be. I’ll start with my first of two questions…

On the calculations page to determine lengths, is the belt length a finalized number, in other words does it account for the loop around the tensioning point on both sides with a little bit extra for safety, what have your results been, I’m trying to cut it as close as possible with the uncut belt length i possess.

Second question…

I want to have my working area 20" by 50" so that i can cut out long airfoil wing profiles with plasma. These can have a 1/5 - 1/9 height/length ratio. A long length like that would certainly have problems when using a router. I was going to wire the steppers in series off a single driver and make sure the wire run lengths are identical. Also I’m assuming that the shorter nonflexing 20" side in addition to adding more supports on the 50" lengths would be enough to keep the cuts precise. So with the series wiring and supports… am i going to have any foreseeable problems. Am i missing anything?

Hoping to overcome warp with a liquid feed.

I look forward to reading your replies… thank you in advance for your input!! :slight_smile:


Welcome to the board.

Can I just check that you are looking to make a CNC printer 20 inches by 50 inches, that’s 50cm x 127cm? That’s almost twice the size of my MPCNC and I think mine was a large build for a first one. You will almost certainly need some support for your long rails, no idea what the final size will be, but I’d assume another 25-30cm each way. What do plasma cutters weigh? That might have an impact as I would gave thought you will have some sag at that size. Might be only a mm but that could make difference to you.

I’d have though the belt length and the costs of that are immaterial to the costs of the rest of the piping. I paid circa $6/£5 for 5m of belt for the older Burly. You might need two rolls with a monster sized CNC bed, however the costs of the piping, especially if stainless steel, will make the belts a rounding error :slight_smile: Allow yourself some slack so you can adjust things.

I wouldn’t make the wire lengths identical, in fact I would advise you not to do so. You will end up with an enormous amount of wire left over that needs to be curled and put somewhere. I did exactly what you are planning to do and ended up going back and making everything the correct length. The electrical distance will make no difference unless you’re running close the speed of light AND you have exceptionally long runs, which I suspect not.

I have no idea what liquid feed is unless it’s alcohol, in which case I think that’s always a good idea.

Keep us informed as to how you go.


1 Like

Regarding the long length. Your Z will be supported nearby, because of the shorter axis. But the 60" long moving tube has two jobs, and it will be hard for it to do them at that length.

  1. It keeps the router upright. As ot gets floppier, the router can tip towards the long ends.
  2. It holds the position along the short axis. If you were to imagine the router in the middle, amd you pushed it along the short axis. The motors that resist that force are 30" away.

It might be ok with a plasma cutter. I don’t have one. Have you considered a low rider though? Those are designed for these kinds of sizes.

1 Like

Hi there Rob!
Yes those sizes are correct. Regarding the belt sizing i was mostly trying to get away with nothing left over vs achieving a certain size. I realize now i can just pull the lengths of pipe through the fittings and trim the rest off…
A plasma torch weighs about 3/4 lb since the guts are remote in a box. There is up to 60 psi air shooting out so that would potentially wiggle the z some i suppose.

Thanks for that info about the wiring lengths. I was remembering deep cycle battery wiring trying to keep everything balanced, i surmise that it doesn’t apply here.

And finally the liquid feed is just a steam of liquid being splooted constantly onto the cutting area to keep the temps down. I would agree that it’s similar to a beer hat however…

I really appreciate your feedback! Good stuff. Thank you

Hi there Jeff! Thanks for your reply!

I will take into consideration these issues. I haven’t cut anything yet.

Originally i had thought… do i want the lowrider…? Baww, plasma wont have any backlash.
It’s hard to charge my mind now that printing is almost done! :slight_smile: i think im going to go ahead, if anything i can post the results for science’s sake and then trim it down to a better length.
Well, thanks again for getting back to me, I’m really excited about this project!


That’s a great attitude. The hardest part about giving size advice is that some people expect perfection and aren’t looking to invest time in a project, so they should build it in the sizes we know work. But other people are tinkerers and willing to see some errors and in that case, why stop them?

It is your machine, and I hope you know the smaller machine works, so why not try to make it the way you want?

1 Like

Are you going to implement torch height control? Any tubing sag, or slight variation in z elevation shouldn’t really matter to much with good thc.

The torch is light but you’re still dragging the heavy lead around the table, so keep an eye out for a wobbling torch.

I’m excited to see your build, good luck! What electronics are you planning on using?

1 Like

Hi there Kyle, thanks for posting here!

I was planning to implement a THC of some type, admittedly i will try to avoid doing so if possible, due to the fact of driving this project with an arduino uno, v3 shield and the stronger of the two driver choices for that setup.
I realize that this is the gimp and cheapo setup… for $20 i got all those electronics off Ebay shipped from here in the us.
I went this route because in my mind it will allow a fair amount of modular design, although i will have a little more effort to implant such aspects, and to be fair im wary of frying a nicer brain via error.
I anticipate that packing a thc onto an arduino uno running GRBL + v3 shield may not work. I haven’t even done my due diligence in that aspect. However, i always enjoy trying to make modular design work, even when i have no idea if it’s possible.

While i know this approach is somewhat hairbrained, creating and solving chaos in a true old fashioned maker style is very fulfilling. The biggest downside is my overflowing parts bin(s). I learn quite a lot this way.

Anyway, i would appreciate any relevant advice you may have about implementing THC in this fashion. No worries if not, i will succeed eventually, learning about and building this project has been very enjoyable and that’s not going to end anytime soon.

Thanks for reading all this, hope all is well with you, have a great day there!

Theres been a few mentions of implementing thc diy with an arduino around the forums. I think there’s a project or two on GitHub too. But I’m not aware of anyone using it successfully.

The cheapest store bought options I’m aware of are Mesa THCAD and Proma. Proma would be easiest to implement in your current setup but is a bit pricey, around $300 USD I think. The Mesa is cheaper at $70 and actually would perform better BUT you’d need to move your system over to linuxcnc and change out (upgrade?) some of your electronics which of course adds more expense.

1 Like

I think it would be a great project to have completely separate from Marlin. Marlin could send a signal with an M3/M5 and then the THC would just measure (something) and have its own servo and mini Z to go up and down. It could also handle the on/off of the torch. Everything seems fine in my head, except for how to measure the height. And I don’t have a plasma or any metal working equipment.

1 Like

The height from the material is inversely proportional to the voltage of the arc. So all you really gotta do is measure the voltage and keep it steady. Safely :electric_plug: :skull: :ghost:

1 Like

Yeah. How high is the voltage though? How do you measure voltage that high safely.

1 Like

On the order of 120 V when cutting, but you have to survive any turn on arc voltages. I don’t know how high those get. Most setups seem to have some sort of opto-isolator between the nasty bits and your low voltage arduinoey logic bits. The mesa THCAD uses frequency modulation of the arc voltage to communicate that.

1 Like

My machine cuts around 130 volts depending on the thickness.

How to design a circuit safely? lol, I was hoping you would know! My plasma has an onboard voltage divider for this purpose that is adjustable from 50:1 to 20:1 so my Mesa card is actually only reading voltages under 10v and the software scales the reading back up. I honestly have no idea how common onboard voltage dividers are on plasma cutters from different manufacturers.

A relevant, admittedly biased, thread started the other day: https://forum.linuxcnc.org/plasma-laser/40186-thc-torch-hight-controllers
Interesting that the Mesa engineer chimes in on page 2 that a thc a-d could be made for $2…

@Claystorage I hope we’re not derailing your thread.


I make enough mistakes with my circuits that I don’t make any that can kill me, or need to work well enough to save my life. It looks like there is some esp32 software already, but I didn’t see what hardware that was using. I know I see funny gaps in PCBs that are meant to handle even AC 120V power. I am not very interested in learning the lessons that taught them to make those gaps.

1 Like

My primo is 25x50 (inches) , carrying around the 611 (for now). Only issue I’ve got so far is when cutting in the short direction, the long tube will flex a little and give me some chatter.
As long as you can keep the lead from getting snagged on anything and you build carefully, i think the size won’t give you any trouble to speak of.
FYI, I used 1 inch 0.1 wall DOM and I have mid length supports to reduce sag.


Thank you all for continuing the discussion here, very happy making for me to read the ideas flowing…
I did some researching various ideas regarding thc; ultrasonic, capacitive, and direct voltage sensing are all notable choices for the poor man in me, with the last choice being potentially the most capable and also the possibly most complicated choice. Like the engineer in the previously mentioned linked post says, a 555 and optoisolator could potentially be enough. I was thinking about using an arduino NANO as a standalone to process voltage sampling and free up the primary UNO for GRBL, with the UNO only recieving a simpler code from the NANO regarding the z axis. That premis seems sound enough, however the issue i have come across is that my lotos 3200 is a high voltage inverter type cutter.

Im guessing that it would be simpler to build a faraday enclosure for the cutter than all the separate electronic bits, although i don’t know if that would be enough due to the cutting wand being in the middle of the work area.

I am going to take a hint from Kyle mentioning that his cutter has a built in bit, i am guessing that somewhere in my cheapo unit there are some circuit guts that i can wire into for sampling, preferably at a reference voltage point, that would be great! This is all supposition at this point of course. Safety first.

Im kind of surprised there isn’t a mechanical or partially mechanical thc, i was imagining a floating z end, looks similar to a vaccum attachment circle around the tool, has some b.b. trapped underside, can provide a liquid cooling point easily. Uses opposed zstop switches to find the measurement and rolls across the cutting face as the tool head moves. That would be great.

Tony, good to hear that your dimensions work well enough! Thanks for that input. I sat down to cut up my belts and had a moment of trepidation. Glad to know that you’re out there!

1 Like

Just wanted to pop by and make a noise. Youtuber “Minh’s man cave” with this “THC.” Just pushing down on the thin steel, would be effective, if not crude… anyways, I switched to and am waiting for my ramps 1.4 card to arrive, i wanted to be able to run from a microsd card so there will be no risk of damaging a laptop with the HF going on. Will be doing the shielding on wires with tinned copper mesh from Jameco electronics very similar to DUI, forcerogue, with his great setup. Once everything is in order ill post a few images and the specs. Ok have a great day!

I don’t think you should worry much about the lenght, in my opinion it would even work fine if you were to build it twice this size. Plasma torches are lightweight, and plasma arc is quite forgiving, it will still cut fine even if you have one or two millimeters of height variation (though a constant height is always better, obviously).

Main concern while building a plasma machine is really the electromagnetic interferences, so I’d recommend you to get a plasma torch that is well known for being CNC friendly in the first place.

Also, a THC seems like a really good investment, though I never had the opportunity to try it myself unfortunately. I suppose it is not even optional when it comes to cutting thin sheets.

I’m a bit curious about what you plan to cut, you are mentionning airfoil wings, is that for some kind of airplane project? In which case, do you plan on cutting aluminum with it?

Hi Dui! Thank you for commenting here! I read through that 380 comment thread you have discussing your build… took a while but there was a lot to learn there.
To answer your question: to begin with this build i have two projects in mind, first one is a plastic shredder similar to precious plastics community built models, about 8mm stainless pieces less than a foot in size. The other project would be cutting out the internal shapes for airfoil wings. Some would be for “sail wing” for use on small sailboats, and perhaps i would try to build an autonomous solar panel “powered glider” drone that i would like to be able to go long distances with a 10-20 lb payload. I will persevere to these ends. The foil shapes would be aluminum. Do you have any tips for cutting aluminum? Thanks again for stopping by, have a great day there @forcerouge

1 Like