Plasma Cutter Success!

Post Reply
msg558
Posts: 23
Joined: Sun Jan 27, 2013 5:41 pm
Location: Fort Worth, TX

Plasma Cutter Success!

Post by msg558 » Tue Apr 23, 2019 5:04 pm

Hello all. I've been working on converting my Shapeoko 1 to a plasma cutter off and on for about a year now. I've finally gotten some success and was able to cut out a tiny sign (150mm x 75mm) as shown below.
IMG_0717.jpg
IMG_0717.jpg (799.35 KiB) Viewed 385 times
Improvements to the cut quality are still in progress, but I'm pretty happy with this result.

I thought I might share a few details in case anyone else is considering doing the same. The plasma cutter that I chose for this is the Lotos LTP5000D which is about 400 USD on amazon. The main reason for choosing this model is that it has a pilot arc function which allows for the plasma arc to start in the nozzle, and does not require initial contact with the work piece.

The main and most challenging issue that I came across is the high levels of electromagnetic interference (EMI) that occur when the signal for the pilot arc travels along the length of the torch hose, acting as a giant antenna. For me, the interference would cause the USB connection to the stepper controller to reset before any meaningful operations could be completed. I originally thought that it was from the arc itself, but by removing the hose and allowing the arc to jump across the terminals on the plasma cutter box with a small wire the interference was mostly eliminated. Obviously that's not a practical solution, but it proved that the signal travelling the length of the hose was critical to the interference. I was able to measure and plot the interference signal vs. time using a stripped audio cable and the microphone jack on a laptop. More details on that if people are interested.

As some have suggested on this forum, shielding and grounding was the key to attenuating the currents induced by the pilot arc EMI. You'll notice from my photos below that my approach is quite crude in its construction, but I wanted to prove that it would work first. Maybe someday I'll make it fancier.
IMG_0712.jpg
IMG_0712.jpg (689.65 KiB) Viewed 385 times
IMG_0713.jpg
IMG_0713.jpg (757.55 KiB) Viewed 385 times
The plasma torch hose was wrapped with heavy duty aluminum foil (about 2 layers) and then wrapped with duct tape to keep it all together. The yellow wire coming off of the aluminum foil runs outside to a grounding rod. I replaced all of the original stepper motor wires with shielded 4 conductor wire, and wrapped more foil around the unshielded portions of the wire that went into the steppers. I went ahead and wrapped the steppers too because why the hell not. The shielding is connected to ground via the stainless steel table that the laptop and controller box are sitting on, which is connected to ground via the ground connector on the laptop power plug. This means that the stepper wire shielding and the torch shielding are grounded by two different ground rods. This was done arbitrarily, and I honestly don't know yet if it has any effect. I also wrapped a plastic tub in aluminum foil and placed it on it's side to house the laptop; it's grounded by making contact with the table. Shielding the laptop may also have been overly redundant, but I don't know that yet either.

Grounding the shielding is critical. With the shielding ungrounded, I would have a problem about 50% of the time. I haven't had an EMI problem yet when I've made sure that my shielding connections to ground are solid. Since I have two different ground rods (one from the house electrical system and one that drove myself that just goes to that yellow wire), I am able to measure the resistance between the two by putting one multimeter probe on the foil on one of the steppers, and the other on the foil on the torch. When I wet the ground beneath the rods with soapy water, I measure about 1000 Ohms. This is how I test for good grounding. I believe the grounding allows a path for induced currents to be dissipated, whereas when the shielding is ungrounded, the induced currents just reflect in the wire and re-emit EMI.

I use a relay module that I got off of amazon to control the plasma with the grbl shield. I noticed that the wire that connects the relay module to the plasma cutter must also be shielded and grounded, so I just used more length of the shielded 4 conductor wire and connected the shielding to the table, which is grounded. I made a poor man's water table by suspending a cooking sheet across some angle iron and then clamping the angle iron so that the pan won't slip out under vibration.

Anyway, just wanted to share that I was able to get this working. I'll be glad to answer any questions or give more specific details.

Post Reply