Fluke SV225 LoZ adapter energy dissipation?

[Per Hansson]

Hi, last week I was troubleshooting a DC circuit at a customer's machine with my Fluke 28-II.
I saw a voltage of 15v where I expected 24v, I took a while before I opened the schematic and saw that it was not an energized circuit at all:
Just a (very long) jumper wire between two machines, to start one of them up: thus leading to capacitive coupling... (aka Stray Voltage reading)
So I realized I really need a LoZ function on my meter, which the 28-II lacks.

I found the Fluke SV225 adapter that looks promising: it has a 3kΩ input impedance and just plugs inline with the probe leads on the meter.
But this brought me finally to my question: another use I've envisioned for LoZ mode is to discharge capacitors.
Now of course a meter can't dissipate a large load, so it uses PTC's to limit the energy.
I have a Fluke 113 that really was a wrong purchase, well I bought it for it's capacitance mode and that it does ok so I can't blame it.
But it's an electricians meter in that it's always in LoZ mode, and also has some fully auto mode that does not work at all on lower DC voltages.
Anyway I charged up a 390µF capacitor I had on the bench to 325v and then used the Fluke 113 to discharge it.
I was surprised by how quickly it did it! Just a few seconds to get down to safe voltage levels.
My question here is how much energy can the Fluke SV225 dissipate? In my example if my math is not off that was 20J of energy.
Actually at 325v x 325v / 3000Ω = 35w, that's a very large load indeed, but for how long?

[HKJ]

The adapter is rated for up to 1000V, I would expect it contains a PTC

[Per Hansson]

Yea there are PTC's in it, just curious if anyone could give an estimate of how much energy it could dissipate.
Someone made a schematic of it in this post: https://www.eevblog.com/forum/testgear/brymen-bm867-bm869/msg1141765/#msg1141765

[HKJ]

With multimeters it is in the 1W to 2W range, probably the same here.

[Per Hansson]

I tested the Fluke 113 on a large servo drive, I had 350VDC on it. (unknown capacitance)
Measured with a second high impedance meter the servos voltage dropped very slowly after I turned it off.
I then connected the Fluke 113 and it dropped quite quickly down to 300-ish volts.
Then it slowed down allot, so the PTC's did their job...

I next tried another servo, but where I knew it's capacitance was in the 1000µF range, charged to 325VDC.
On it's own it discharged to 10v in a little over 5 minutes, with the Fluke 113 that took only 12 seconds.

It's totally off topic but still one thing confuses me with this:
How can the meter read the voltage correctly when the PTC does it's job?
Does it not mean that it changes the input resistance from 3kΩ to something allot higher?
EEVblog #373 goes into this, the link goes to 27:23 where Dave discusses the 1M hybrid resistor.
Is this how the meter can measure the voltage even if the PTC's go up in value? (As they are not in the same path).
If that is so how big a reading error can I expect when the Fluke SV225 adapters PTC's have become highly resistant, or will it not matter? (If so then why?)

[HKJ]

How can the meter read the voltage correctly when the PTC does it's job?

The PTC is parallel with the meter input terminals, this means the meter always sees the full voltage.

[Per Hansson]

Yea I think I get it now, in a meter the PTC is in series, but in the SV225 LoZ addon it's in parallel, that's what you meant by your message above right?
And following on with that logic it really is the 1M hybrid resistor in the meter itself that allows it to display the correct voltage when its PTC in series has changed resistance?