CNC PlasmaCAM Cutter: Working, but a few caveats

Metal Shop
1

Greetings DMS Tribe,

“A lot of credit should also go to Joel Gray for donating a PC that
actually works and Michael Lass for figuring a lot of the machine out.” - Chris Wischkowsky

Chris, I greatly appreciate the mention and the offer to teach the tool!

But there are a few things to address about the PC before relying on it fully for your upcoming class and future users.

  1. I built up and installed the PC and software in mid-October, and then had to add a NIC card in mid-November so that I could activate that installation of the Windows XP SP1 OS before the 30 day activation period ran out. Michael Lass was kind enough to meet with me and walk me through the operation of the machine and I took a number of notes.

We also did a number of test cuts at different arc voltage settings and speeds, documented them with a white paint marker, and Michael concluded that the machine worked to his satisfaction. (last I checked the 0.25" steel test plate is still in the metal shop as a helpful reference and I took photos as well)

  1. And while the PlasmaCAM has already been technically operational for several weeks, during testing / using the machine early this December, Chris discovered that the software started locking up in the middle of a job and contacted me regarding the issue. I also had noticed some artifacts in the software graphic of the table where some grey areas were showing up on the lower left borders which I had not seen before.

Upon listing the symptoms including noise from the hard drive, we concluded that the drive is probably on it’s way out and I offered to install another one which I originally planned to do this past Sunday, but that was delayed due to an unrelated event.

However as is sometimes the case with hardware, problems can temporarily get better due to temperature changes, subtle movements, etc. and Chris made the announcement that the PlasmaCAM is ready based on the apparent disappearance of the above symptoms and my verbal commitment to look into it. But the truth is that I haven’t touched the machine since we discussed the latest issue, so there is a very real possibility that the hard drive could suddenly take a nose dive at any time.

  1. Because of the location of the PC in the metal shop, metal dust intrusion is a real issue, I had to seal up all of the open PCI slots in the back and in doing so, I have already noticed some contaminant beginning to collect on the bottom of the case. If this is metal contamination, then that could either be responsible for the behavior described above or at least shorten the lifespan of the motherboard, etc.

I purchased some filters for the power supply and case fan this past week, and while it is open, I will test the dust to see if it is metal, and then blow out the case before installing the filters. After a week or so, I will reopen the case and see if infiltrates are still getting in. If so, then I am looking into some inexpensive dust cover enclosures to help shield the computer.

I am bringing this up because I want to caution anyone using the machine that I need to look at it again (probably this coming Sunday) and install the filters to help prevent any sudden lockups / work stoppages. Also, I don’t want any fingers pointed at Chris if something fails when I am the one who should have already taken care of the above line items.

  1. Finally, I will feel better about contacting Ops regarding the indefinite loan / donation after determining that the PC will work reliably.

Best Regards,

JAG

1 Like
2

I’m guessing that a solid-state drive would perform more reliably in a dusty atmosphere with a lot of electromagnetic interference possibilities than a conventional hard drive.

3

Mr. Carson,

I’m guessing that a solid-state drive would perform more reliably in a
dusty atmosphere with a lot of electromagnetic interference
possibilities than a conventional hard drive.

I have come to greatly respect your knowledge and critical thinking skills, and in theory, that makes sense. However in this particular instance, there are some other considerations:

  1. If the drive is failing, it is likely due to age and usage (circa 2007 or earlier I think) not because of EMI (it would have to be at least 10 Gauss to even begin to affect a drive and it is unlikely a field that strong exists where the PC is positioned) and while there is a built in breather hole for all conventional hard drives to help equalize pressure at high altitude operation, they have internal filters to prevent contamination.

  2. This is an old computer (PII 350 MHz) and it would not likely support a solid state drive because virtually all of those are SATA interfaced.

  3. The motherboard is another potential candidate for metal particle shorts.

  4. Solid state drives can be just as unreliable as hard drives or even worse, depending on the manufacturer, design and manufacturing run. I have spoken with several sales reps in computer equipment and read reviews of SSDs that died within a few months or were even DOA.

I believe filtering the contaminants before they even get into the case at the air inlets is the best course. Positive pressure (like in a semiconductor fab cleanroom) should repel anything else, but if the PC is off part of the time, then the dust cover may be a good addition.

JAG

4

If it won’t accept a SATA replacement drive, you’re right, the point is probably moot.
I was thinking in terms of EMI as a problem for the power supply, actually – line spikes in particular. I would expect a solid-state drive to be less in danger of damage from brief undervoltages (or the sudden total failure of the PSU) than something with a motor in it. And, certainly, SSDs can be unreliable, but there is presumably some advantage to isolating the most likely sources of unreliability.
I definitely agree that excluding metal particles from the case is of great importance! I recall reading about a stuffing the cleaning air inlets with glass fiber, or something like that, to serve as an air filter.

5

I have a PC that use to be in the 3D fab room without ram (DDR2 I think) or a hard drive but it does have SATA and is pretty modern. You can have it if you want to use it if you think it will work better.

Also using a paper towel or mesh to make a fan filter might help.

6

I used the PC today with no issues for a few hours today and a week ago. It performs fine when only running PlasmaCAM.
I think ‘being old and crusty’ is a requirement for driving that machine. :slight_smile:

7

Used it PlasmaCam successfully yesterday as well. I do fit the old & crusty requirement.

8

Heavy Metal Enthusiasts,

I was up at DMS Sunday and I spoke with Mr. Joshua Masseo who has offered to teach a class as he was trained on the PlasmaCAM from one of the initial classes from a while back.

He related, (as has Mr. Ken Purcell on this thread) that the computer ran flawlessly through a considerable number of cuts.

I definitely agree that excluding metal particles from the case is of
great importance! I recall reading about a stuffing the cleaning air
inlets with glass fiber, or something like that, to serve as an air
filter.

(from Mr. Carson)

Also using a paper towel or mesh to make a fan filter might help.

(from Mitch)

I installed an 80 mm KingWin filter on the power supply fan - it is a bit ghetto (taped on) and I need to get another of the same size for the case fan. (they are inexpensive at $3 each) Also, I may look into using appropriate length screws to attach them a little more professionally.

However, I went ahead and opened the case for inspection and I definitely found metal dust intrusion (verified by running a small magnet over the dark residue on the inside which could have been responsible for the erratic behavior reported a few weeks ago, and over time will certainly become an issue.

I then blew out the case completely while unplugged by using a cord operated electric duster. (man, that thing is surprisingly powerful (4.0 Amps) for the size and better than the disposable compressed air cans)

An IT friend of mine referred me to an inexpensive solution for protecting the computer known as a Dirt Bag Cover:

http://dirtbag.biz/covers/tower.html

I measured the size of the PC as being 17.5" high and deep by 7.5" wide. The 18" bag at around $40+ should work.

I think ‘being old and crusty’ is a requirement for driving that machine.

Agreed.

In my experience with PC driven instruments, it is a good idea to stick with the original OS and PC hardware that existed during that era as legacy software will very often have trouble with very new systems. When I worked in a research lab with computers that controlled $1 million dollar electron microscopes, you NEVER even let that computer see the Internet as the Windows updates alone would often “break” the machine.

Also Joshua related that the PII 350 works plenty fast for PlasmaCAM.

I was thinking in terms of EMI as a problem for the power supply,
actually – line spikes in particular. I would expect a solid-state
drive to be less in danger of damage from brief undervoltages (or the
sudden total failure of the PSU) than something with a motor in it.
And, certainly, SSDs can be unreliable, but there is presumably some
advantage to isolating the most likely sources of unreliability.

Okay, Mr. Carson, so you are talking about conducted EMI as opposed to radiated EMI. That is a very good point and my solution would be to put the PC on a UPS or at least a good surge suppressor.

The advantage of a UPS is being able to keep the computer up during a rolling brownout or temporary power outage (which has happened at the space) while in the middle of a job. My concern then would be what happens to a plasma cutter if the PC dies in the middle of a cut.

At any rate, due to the reports of reliable operation of the software, I will hold off on putting in another drive because I don’t want to “fix” something if it is working well now.

Best Regards,

Joel-Anthony Gray

1 Like