Plasma Cutter 7i92M stoped working

I have a plasma settup.
I have hybride steppers with a 7i76 connected to a 7i92M
I was just about finishing it.
I did cut about 20 test holes and some lines to adjust feed rate and current.
Suddendly after the arc firing it stoped (no response from the card).
I restarted the setup and it worked again, I did some more testing and it happend some time.
I thaught it is because of EMI and started to change some grounding.
But id did it again and now I can't start it any more.
All the Led's on the 7i92 just blink breathli at startup and then just power stays.
I can't ping and I cant start the machine.
I checked the cards visualy and found nothing.
Then I checked themperature and I found out that the ethernet chip (little chip next to the ethernet port) is very hot.
I measured about 80 degree celsius (after about 1 min).
I have another Mesa board and there the chip gets only like 35 degree celsius.

Is it dead? Or what can I do?

cnc4ever wrote: Or what can I do?

Check and recheck the wiring, redo the grounding.
DO NOT use a new 7i92 until you find and fix the issue, it will end up dead.
For plasma, as a general rule:
-Proper grounding, www.google.com/search?source=hp&ei=j07GX...ccDz4Q4dUDCAY&uact=5
-shielded cables, shield connected ONLY on one side, usually the control box side. NEVER both sides, except where required, see above.
-signal cables as far as possible from the high voltage/high frequency cables
-torch cable going parallel with motor/sensor cables ONLY if cables are shielded.

Yeah, its probably dead. The Ethernet chip (about $3.50) could be replaced but
that's fairly tricky.

As Tommylight said, its possible you have a grounding or signal coupling issue that
induced a large EMI spike into the 7I92 somehow.

Hi,
a short circuit is sometimes evident in the control circuit. I built a voltmeter to keep an eye on this. maybe this will help.


add in custom_gcvp.hal
net fieldpower pyvcp.fieldpower hm2_7i76e.0.7i76.0.0.fieldvoltage

add in pyvcp-pannel.xml
<!-- Fieldpower -->
<hbox>
<relief>RAISED</relief>
<bd>3</bd>
<meter>
<halpin>"fieldpower"</halpin>
<text>"Spannung"</text>
<subtext>"Volt"</subtext>
<size>200</size>
<min_>8</min_>
<max_>16</max_>
<majorscale>1</majorscale>
<minorscale>0.5</minorscale>
<region1>0,9,"red"</region1>
<region2>11,13,"green"</region2>
<region3>15,16,"red"</region3>
</meter>
</hbox>

Thanks for the hints.
I redid all the grounding as in the hypertherm document.

I replaced the 7i92 and tested everything.
Was working flewless.
But as soon as i started the plasma it happend again. Actualy even faster. the pasma didn't even fired.
Now the "new" is broken to, but the ethernet chip is cool. Don't know what is broken.

I added some pictures.

I know my plasma cutter is probably not the most commen one.
Chinese LGK-200. Did anybody ever worked with this one (or a similar?).
I could get it preaty cheep this is why he is here.

I don't know what I should do. Will probably try to get a osciloscope and do some measurements....
- 5v supply
- pe to 5v
.... ?

Its not very easy to test if the replacment costs 100 each time -.-
Will also order a replacment ethernet chip for the old one.
I do have some solder skills will see if its enought

It looks like the LGK-200 is a HF start plasma source. These are notorious
for generating huge amounts of RFI when starting and require very careful
attention to any possible coupling of this noise into low level circuitry in the
control box.

On the picture with the torch you have plenty of cables rolled, that is strictly not good on any HF machine.
All cables as short as possible, HF/HV cables never in parallel with low signal cables, the torch cable on that plasma is most probably not shielded at all, so shielding everything else from it is imperative.
You are using servo drives, those will also produce big amounts of interference from the moment they are enabled, so putting ferrite shielding on motor and power wires is a must.

it has a reason why your network card burned and not be any other part. it could be possible that short circuit go over pc ground or opposite. Check Ground of your Maschinen or pc. may one power cable or unit have a short. second step in test run disconnect ground of your pc.