.99,

TK's warning about a Q2 failure damaging your pulse generator's output is a valid one.  As well, a stray cliplead inadvertantly contacting the drain or +rail could do similar damage.  You might want to at least slip a piece of heat shrink over that 'gator clip on the FG out lead.

A couple of back to back 10V zeners directly across the PG output and a 100-150ma inline fuse between the PG output and circuit would offer some degree of protection.  At the very least, even without the zeners, the use of the inline fuse at the PG output might be better than nothing!

Using that very nice PG in this application is kind of like using a sports car to haul wood...

It appears your Q1 is turning on and passing current similar to FIG5.  Hopefully it will be OK with TK to just remove Q1 or swap its gate and source leads to duplicate FIG3, as opposed to actually having to blow a MOSFET!

However, I am assuming your primary investigation is with regard to the neg pwr measurement.

PW

Quote from: poynt99 on June 19, 2013, 11:58:02 AM
With a 38V supply, I am not too concerned, as the output devices should tolerate a dead short of Q2 in such case. However, I am concerned to go higher, and therefore plan on building an external drive to buffer the 8160A.

.99,

I would not count on the 8160A being able to handle a Q2 short at 38V. 

The buffer is a good idea but sounds like a bit more work.  No old FG's laying around?

PW

Quote from: picowatt on June 19, 2013, 12:27:57 PM
.99,

I would not count on the 8160A being able to handle a Q2 short at 38V. 

The buffer is a good idea but sounds like a bit more work.  No old FG's laying around?

PW

I don't have anything cheap and/or old that has duty cycle adjust.

I'll be looking at a whole lot more than just the negative battery power.

@.99:

Ainslie said, on the 3rd of June, sixteen days ago:

QuoteAnd while you're at it PLEASE, make that waveform for our 'ickle pickle'.  I'll post it hereunder.  It's no good my demonstrating this - even in a live broadcast.  He'll somehow find cause to contradict it.  And it's very easy to replicate.  If you struggle - then get hold of me.  I'll guide you through the required settings.  But they're simple enough.

As I've demonstrated, it is indeed very easy to replicate.

But not in the way she has claimed. That's a bit more of a struggle. Has she guided you through the required settings yet?

After all, they are simple enough. The only settings available are on the Function Generator (once you've got your 72 volt source hooked up.) The period of 160 seconds with a 10 percent duty cycle is very plain to see. I don't know if the 8160A can match that but I've got the manual here so I'll check.
All that's left is the FG amplitude and offset settings. It should be possible to specify those in a sentence or two, or a scopetrace of the FG alone with nothing else hooked up.

On second thought, I think a 250mA fuse will be a lot easier.

TK, I'm not particularly interested in performing the long period tests, so I won't be pursuing that, at least not right now.