DVM goes crazy when checking plate voltage

[pgussler]

This has happened to me several times over the past 20 or so years and I don't think I have ever truly gotten to the bottom of the cause.  When checking power tube plate voltages with a DVM a high frequency oscillation is produced and the meter display goes haywire when the probe is touched to the plate.  The several times I have encountered this has always been on P-P circuits with a LTP phase inverter and it is always on the tube (or set) closest to the preamp.  If I pull the PI tube I can get normal readings.  This particular circuit is a 12AX7 LTP pi and 2 x EL34 output.  I can only remember it happening on EL84's or EL34's but I could be wrong. Has anyone else encounter this and resolved it?

[pdf64]

Yes, it’s a common thing, to be expected, as the probe leads and meter form a transmitter antenna that delivers the plate signal back to earlier stages, creating feedback loops. I use the HT voltage supplied to the OT instead; unless the OT has failed, the (idle) plate voltage will only be a couple of volts lower, which is negligible.
If the plate voltage really must be known, as you’ve found, the simplest way to avoid the issue is to remove the tube from the socket of the preceding stage’s circuit, eg the phase splitter.

[tubenit]

Paul,

Are you still building your fine Oldfield Amps?   Isn't this your company?   http://oldfieldamplifiers.com/

Nice to see you on the forum!  It's been awhile.

With respect, Tubenit

[pgussler]

Yep, that's me.  Still whacking at it.

[pgussler]

PDF64 - Much thanks for that explanation and it makes perfect sense.

[pdf64]

PDF64 - Much thanks for that explanation and it makes perfect sense.

You're welcome 
Maybe think of probing a power tube plate momentarily creates something close to a worst case lead dress scenario.
Lovely looking amps, kudos!

[tubenit]

Paul,  I live in Charlotte also.  I will say your amps are well thought of by some of the Charlotte guitarists & a well known luthier that I've spoken with.   

The amps look fantastic!

Best regards,  Tubenit

[PRR]

The "right way", used in nearly all VTVMs, is to put 1Meg IN the probe, before the long wire back to the meter. The 10/11 error is accounted in calibration. Now the AC signal at plate only has an inch of "antenna", rather than 3 feet of antenna.

Obviously you could twist a 1Meg onto your DMM probe, and probe with the far end of the resistor, but the 10/11 correction is awkward.