Near Miss with a Class E Rig?

[WD5JKO]

  Yesterday at the shop I was looking at a Industrial class E rig when something bizarre happened.

OK, this was an Advanced Energy 300X RF Amp, which runs 3 KW RF out @ 13.56 Mhz from a 208v 3 phase power source. The efficiency was only 60% when 65% is more typical. I measure efficiency by AC RMS input power/measured RF output power X 100. Since there are three energy conversions ( 1 AC to DC, 2 DC to DC variable, 3 DC to RF) in this unit, 65% overall is not too bad.

I opened up the unit, and using appropriate scope probes (100X to look at the FET Drains since they peak at 800V when the B+ goes to 200V), I looked at the FET drains and gates simultaneously. Had to use these little spring ground clips on each probe to minimize ground lead inductance.

So with AMP at 3 KW I was holding a probe in each hand. I could see that the gate drive (RF Drive) rose above 5V when the FET drain was on the way down, but still at 100V. It is better for the drain to get closer to zero before the drive comes on, so this was likely why the efficiency was a bit low.

I wanted to see the scope images better, so I held both probes in one hand trying to make the 4 connections while reaching over to adjust the scope. As I reached over, suddenly the room went dark as the entire building lost power.

This is a fairly large building with 480 3 phase feed, and a very large 480/208 stepdown transformer. No breakers were tripped, and in about 5 minutes the power came on. The RF amp, scope, and probes were all OK.

So did I slip with the probe, possibly putting high level RF on the AC ground system, or was the power outage just a coincidence? I recall only one other outage like that in 8 years working in that building?

Jim
WD5JKO

[w4bfs]

Jim ... sounds to me like YOU were lucky to keep from going dark and losing power ... honestly, some of the acrobatics that checking out equipment requires ... interesting gear

[KA0HCP]

The darkout was one of your lives fleeting for trying to hold probes on a live circuit!

[W3RSW]

So you have eight left? 

[WD5JKO]

The darkout was one of your lives fleeting for trying to hold probes on a live circuit!

So you have eight left? 

  Well possibly a wake-up call for sure. There I was with 3KW RF, 200VDC to the FETS, Three phase power 208V pulling 14 amps / phase, Bird 5KW dummy load blowing hot air directed back to my lap, water flowing at 2 Gallons / minute to cool the unit, and there I was doing a hot measurement with one hand trying to maintain 4 connection points...

  I might need to just solder in a capacitive voltage divider terminated with 50 ohms, and have that go to the scope with 50ohm coax. It would have to be surface mount, and have to deal with soldering a heatsunk PCB, and conformal coating. Not sure I need DC accuracy, and maybe a 100:1 capacitive divider would work. I need a hands off way to make these measurements!

Jim
WD5JKO

 

[KA2DZT]

Now I know why my lights dimmed down yesterday.

Jim, let us know in advance if you plan any more testing.

Fred

[Opcom]

The darkout was one of your lives fleeting for trying to hold probes on a live circuit!

So you have eight left?  

  Well possibly a wake-up call for sure. There I was with 3KW RF, 200VDC to the FETS, Three phase power 208V pulling 14 amps / phase, Bird 5KW dummy load blowing hot air directed back to my lap, water flowing at 2 Gallons / minute to cool the unit, and there I was doing a hot measurement with one hand trying to maintain 4 connection points...

  I might need to just solder in a capacitive voltage divider terminated with 50 ohms, and have that go to the scope with 50ohm coax. It would have to be surface mount, and have to deal with soldering a heatsunk PCB, and conformal coating. Not sure I need DC accuracy, and maybe a 100:1 capacitive divider would work. I need a hands off way to make these measurements!

Jim
WD5JKO

 

Knocking out power? Are you a descendant of Nikola Tesla?
I've experienced a similar thing, in the middle of using a probe and the lights went out, but I didn't work on high power stuff and we frequently had outages at that place. I'm putting my $ on coincidence. Be careful.


(there i fixed it)

[Steve - K4HX]

Why are you guys quoting yourselves?   

[Pete, WA2CWA]

For every "quote" with brackets, you need a "/quote" with brackets. Without the right number, the dialogue becomes a jungle of confusion. Who said what 

The darkout was one of your lives fleeting for trying to hold probes on a live circuit!

So you have eight left? 

  Well possibly a wake-up call for sure. There I was with 3KW RF, 200VDC to the FETS, Three phase power 208V pulling 14 amps / phase, Bird 5KW dummy load blowing hot air directed back to my lap, water flowing at 2 Gallons / minute to cool the unit, and there I was doing a hot measurement with one hand trying to maintain 4 connection points...

  I might need to just solder in a capacitive voltage divider terminated with 50 ohms, and have that go to the scope with 50ohm coax. It would have to be surface mount, and have to deal with soldering a heatsunk PCB, and conformal coating. Not sure I need DC accuracy, and maybe a 100:1 capacitive divider would work. I need a hands off way to make these measurements!

Jim
WD5JKO

 

Knocking out power? Are you a descendant of Nikola Tesla?
I've experienced a similar thing, in the middle of using a probe and the lights went out, but I didn't work on high power stuff and we frequently had outages at that place. I'm putting my $ on coincidence. Be careful.

For every "quote" with brackets, you need a "/quote" with brackets. Without the right number, the dialogue becomes a jungle of confusion.

[WA1GFZ]

You were in the right place at the right time if all the FETs have their lids