10 kVA Kjellberg Plasma Cutter CUTi 31 Teardown

[Mads Barnkob]

/>
A teardown of a Kjellberg CUTi 31 Plasma Cutter with a power rating of 10 kVA. Made for cutting in 25-30 mm steel.

Connecting voltage (V): 3x 400VAC
Connecting power (kVA) 10.4
Mains fuse, slow(A) 16
Open circuit voltage (V) 240
Cutting current (A) 20...70
Duty cycle 70A/35% or 50A/100%

[rikkitikkitavi]

Hard to tell if all the dust is from the fan taking it in from teh workshop (plasma cutting is dusty buisness) or just that good old varporized epoxy housing from the thoroughly blasted semis are responsible.

In the end I wonder how these machines survive with all that dust, much of it is probably half conducting anyhow. Sooner or later somehting is going to discharge through the dust and of the magic smoke goes...

[petespaco]

I agree that many of our shop tools (as well as their owners) suck in a lot of bad dust.  We should all probably have  good exhaust systems, but yet, I don't.  I have much better dust collection in my wood shop than I do in my metal shop.
Not so worried about the old stick welder and big mig welder because they don't have any electronics to speak of.  But I think there's a big opportunity for the makers of newer inverter based equipment to do a better job of filtering.
I have blown up enough Mosfets over the last few years with my induction heaters to understand how some conductivity between gate and drain/source could lead to disaster.
  Several months ago, when we were holding a blacksmith meeting at a local chain Welding supply company, I noticed a large pile of trade-in welding power  supplies that was being scrapped.  Maybe for those same (conductive dirt) reasons?

Pete Stanaitis
---------------

[shrad]

don't forget to wear a high efficiency dust mask when metalworking, this kills...

[Mads Barnkob]

Hard to tell if all the dust is from the fan taking it in from teh workshop (plasma cutting is dusty buisness) or just that good old varporized epoxy housing from the thoroughly blasted semis are responsible.

In the end I wonder how these machines survive with all that dust, much of it is probably half conducting anyhow. Sooner or later somehting is going to discharge through the dust and of the magic smoke goes...

All the boards are coated. Some seems like they are just dipped in varnish or epoxy, while others undergo "conformal coating" like shown in this video

[klugesmith]

Good points about harsh environments.  Yay for conformal coating!  It's not always applied as selectively as in Mads video.  For decades it has resisted dust and salt spray, and been a nuisance at PCB rework and repair benches.

Years ago, my employer made a power supply vendor add conformal coating because of unexpected dust problem.
In an IC factory, of all places.  Here's a stock picture of one final-test IC handler (foreground) mated with electronic test equipment (background).

In a DRAM or flash memory mill, you need to put hundreds of devices in test sockets, and replace any that have shorts or opens, before the time-consuming "full speed functional test and repair algorithm" step.

Long story short: power supplies were failing in 10000-channel DDR memory test machines.  Traced to metal dust from the punched-out leadframes of devices under test.  When periodically cleaning the inside of robotic handler machines, techs used compressed air blowguns.  That scattered dusty material, mostly metal, which got sucked in to the test electronic machines in sections that didn't have closed circuit air cooling.

Naturally the semi mfg co. didn't want to change their floor procedures.  Point finger at the machines which were failing, and make that vendor fix the problem.

[MRMILSTAR]

One of my hobbies is as a self-taught amateur machinist. One of the first things that you learn is not to use compressed air to blow metal chips away from your work area. It just scatters the metal fragments all over. I always use a shop vac to remove the chips. This is especially valuable if you have electronics close by.