Transformer shot! (was scope SMPS/ capacitor venting)

[legg]

On Thu, 25 Feb 2016 12:53:04 -0000 (UTC), Cursitor Doom
<curd@notformail.com> wrote:

<snip>

I didn't get the chance to find out! Began this morning trying to get
some voltage readings off the psu outputs and there was nothing there to
read. To cut a long story short, further investigation reveals something
has gone short-circuit on one of the signal boards. When the psu is
removed and run from my make-shift dummy load, it's still 'fine' with its
new diode (not quite right, but functioning to high degree). So clearly I
jumped the gun slotting it back in the scope when it still wasn't 100%
and now it's damaged something - typical!

Sounds more like you're getting closer to root cause.

Troubleshoot the (unidentified?) signal board.

RL

 

[MJC]

In article <nalek2$k2e$1@dont-email.me>, nospam@no-address.com says...

It looks like the thing may have drifted a little bit out of resonance
over the years. That can happen, electronic parts age and tolerances
slowly increase. Running out-of-resonance, the power factor of the
resonant circuit was probably no longer close to one, but instead the
resonant circuit began to pull reactive power. If you try to drive an LC
circuit with a frequency that is slightly wrong, the driving source will
still force the LC into its frequency, but there will be a phase shift
between voltage and current. The further off the frequency, the larger
the phase shift will become. A phase shift means that a load is no
longer purely resistive, but also reactive (either capacitive or
inductive depending on direction) and so the power factor gets lower. As
the power factor gets lower, the total current draw increases (imagine a
constant current due to resistive load plus an additional current due to
the reactive part of the load, which increases).

Your wonderful description of resonant circuits reminds me of an
experience I had as an apprentice. I was given the job of trying to lay
down a silicone insulating film generated by polymerising a silicone
vapour in a high voltage AC plasma. (I don't remember why it had to be
AC.)

We didn't have any special HV AC supplies but stores did have a signal
generator and powerful audio amplifier. I managed to scrounge a large
open-centred coil (meant to generate a magnetic field around a bell-jar)
and a collection of high voltage capacitors, waxed paper in steel cans
with ceramic terminals on top.

With these I built a series LC circuit which generated a satisfactory
plasma. I'm afraid I don't recollect any measurements. Health-and-safety
consisted of large hand-written warning notices!

However despite being distracted by my plasma I also noticed that the
poor capacitors, excellent with DC no doubt, didn't like the AC current
they were subjected to, and bulged, leaked and fizzed. The experiment
was terminated abruptly!

Mike.

 

[legg]

On Thu, 25 Feb 2016 14:46:50 -0000 (UTC), Cursitor Doom
<curd@notformail.com> wrote:

On Thu, 25 Feb 2016 09:00:07 -0500, legg wrote:

Sounds more like you're getting closer to root cause.

I'm afraid not. It didn't happen yesterday when I first hooked the psu
back into the scope so this is a fresh fault - and probably my fault for
not testing the psu's output voltages properly before plugging it back
in.

Troubleshoot the (unidentified?) signal board.

May have to wait til the latter part of next week; I have to leave later
today for a 5-6 days due to family-in-law commitments.

Missing voltages don't necessarily indicate shorts. They can also
indicate open circuit to the source. Review solder joints and
connections around the transformer pins. Possible damage in recent
removal activity.

RL

 

Are the chopper transistors getting hot ?

Did you actually check the resistance of that resistor that is getting hot ?

 

[Cursitor Doom]

On Thu, 25 Feb 2016 23:48:16 -0800, jurb6006 wrote:

> Are the chopper transistors getting hot ?

The main chopper is a TO-3 cased BJT with a closely-finned heatsink
bolted to the top of it. By the time the resistor starts to emit a
scorching smell, the chopper hasn't even had the chance to get barely
warm.

Did you actually check the resistance of that resistor that is getting
hot ?

Yes, it's exactly 20 ohms as specified. But please don't ask me to do any
other checks for the next few days as I'm staying over 300 miles away at
present.

 

OK, when you get back to it, put together a bulb tester. Take the 20 ohm straight out and put a 100 watt lightbulb (incandescent) there.

Then you start disconnecting things.

That is probably the only way to troubleshoot this. You ain't finding anything with the ohmmeter, but it isn't shutting down. Something is not showing up unless under voltage. Ohmmeters can't detect that.

 

[Cursitor Doom]

On Fri, 26 Feb 2016 00:47:32 -0800, jurb6006 wrote:

OK, when you get back to it, put together a bulb tester. Take the 20 ohm
straight out and put a 100 watt lightbulb (incandescent) there.

Then you start disconnecting things.

That is probably the only way to troubleshoot this. You ain't finding
anything with the ohmmeter, but it isn't shutting down. Something is not
showing up unless under voltage. Ohmmeters can't detect that.

Hhmmm. As I've said before, I'm reluctant to replace that power resistor
with anything higher rated. At the moment it's acting as a robust
detector that something isn't right. I don't want to replace it and then
find the excess energy has burned out the transformer primary instead!

Dimitrij has already given me some steps to follow for resonance checks
when I get back and I'm trying to keep the suggestions made here in an
orderly queue, so thanks for your input which is appreciated, but no
further test ideas from anyone for the time being, please!!!

 

[Mark Zacharias]

"legg" <legg@nospam.magma.ca> wrote in message
news:k32ucb9jaaa5t2ndnkmoruegf3116ib12p@4ax.com...

On Thu, 25 Feb 2016 12:53:04 -0000 (UTC), Cursitor Doom
curd@notformail.com> wrote:

snip

Sounds more like you're getting closer to root cause.

Troubleshoot the (unidentified?) signal board.

RL

Yes! Once you get it fired up again, start looking for parts getting HOT on
those other boards!

mz

 

[Cursitor Doom]

On Thu, 25 Feb 2016 00:36:16 +0100, Dimitrij Klingbeil wrote:

[...]

Dimitrij, I think you may have missed this I posted elsewhere so I'm re-
posting it here now for you personally:

"Final update for the time being as I have to leave soon now:

That short turned out to be intermittent. I hope it was just due to
something shorting out on the bench that won't happen when the casing is
back on because you all know what a bitch it can be to trace intermittent
faults. Anyway, that fault has now disappeared, so I took some voltage
measurements before the 20W resistor got to hot (from 19'C to 60'C takes
about 1.50s now) and I have:

61.7 12.7 5.8 0
-5.8 -12.7 -62.4

This is with the psu board plugged into the scope and all power
connections made except for the VHT stuff.

The correct figures according to the manual should be:

60 12.7 6
0
-6 -12.7 -60

So very close! Looks like the main transformer may be ok after all."

Making progress!

 

[Dimitrij Klingbeil]

On 25.02.2016 16:51, Cursitor Doom wrote:

On Thu, 25 Feb 2016 00:36:16 +0100, Dimitrij Klingbeil wrote:

[...]

Dimitrij, I will have to run these latest checks you suggest next
week now as I have to leave on family matters and have no choice
other than divorce. The PSU is now putting out near-enough the
correct voltages in the 6-60VDC range as required when connected up
to the scope and the transformer is virtually silent. It's just the
power resistor heating that's causing concern. If you think of
anything else, please leave your thoughts here. If not, I'll proceed
with your checks on my return. many thanks again.

Hi

Since you were planning to be away for a while, I was in no hurry to
reply right away. I've seen your other post too, and obviously the
transformer must be ok.

I think that, from the major power-carrying components point of view,
your power supply is now "almost ok". The "power train" clearly works,
otherwise you couldn't get correct output voltages under load.

But the fact that the power resistor still overheats, hints to some
timing being slightly wrong. It can no longer be "completely" wrong, as
was the case with the slow diode, but it's not yet "right" either.


1.

There is still the question with V1808. You said it looks ok, and it
tested ok with a multimeter, but that's not really indicative of its
true behavior under full load at high frequencies. If it has degraded
for any reason ("lost its switching speed") then the resistor R1814
would be running at a higher load than normal. Not many times higher,
but about double or triple. That would be somewhat consistent with your
observation of it running too hot after a few minutes. You should now
have (hopefully) a few spare UF4007s, so if in doubt, replace V1808.

If you find out that the replacement of V1808 makes a (little) change
for the better (slightly lower load on R1814), then replace V1809 too.
It would in this case be likely that those BY208-1000s have all degraded
and became out-of-spec. They all have the same type and age.

Actually it's possible to test the condition of V1808 in circuit,
without replacing it, but the test is tricky. You would need to see, on
an oscilloscope, the voltage waveform across R1814. It should be
basically a flat line, with short surge-like spikes at some 20 kHz
intervals. All the pulses must be polarized in one direction only. The
left-hand pin (on the schematic) of R1814 must be positive. There must
be no spikes in the reverse direction. If there are any (the polarity
would be alternating), then V1808 is degraded and no longer operable at
full speed and needs replacement(, and so does V1809 likely as well).

Unfortunately this test is difficult, because you can't connect a scope
ground to R1814! This is a very fast switching signal that runs at high
power and reaches voltages of some 800 V in normal operation! Even if
you disconnect both mains grounds and "float" both the scope AND the
power supply, and even if you power both the scope AND the power supply
from two SEPARATE isolation transformers in order to increase isolation
and minimize the stray capacitance via mains, this test would still be
very dangerous and I would definitely not advise trying. Using two scope
channels in "subtract" mode might work, but only if you have two high
voltage probes rated for 1 kV, and only if both probes are exactly
identical and the compensation of both channels is precisely matched to
each other (a rather unlikely condition that requires some effort to
achieve). To be honest, to do this test properly, you would need an
isolated high-voltage differential probe. Unless you have one, don't
even bother trying, to replace the diode is easier and much safer.

Ok, so much for the other BY208s in snubber circuits. Replace and see.


2.

The other open question is that of the resonance capacitors (C1807 and
C1808). As I noted in another post, they may be degraded and it may be
difficult to test for this condition properly (LCR meter won't likely
show the problem). Again, if you can get known good spares, they can
easily be replaced, but the spares must be rated for resonant operation.
"Typical" film capacitors are not designed for this use.

Foil capacitors with Polypropylene isolation rated for continuous
resonant duty like the "FKP 1" type should work well here, and so may
"MKP 4C" type too, to some extent, but only the 630 V DC rated ones, and
only if two are used in series like in the original schematic ("MKP 4C"
with lower ratings would hit its high frequency AC limits).

So, "FKP 1" rated at 400 V or 630 V DC (two 33 nF in series) or rated at
1000 V DC (one single 15 nF) or "MKP 4C" rated at 630 V DC (two 33 nF in
series), would be feasible replacement candidates, but not many others
due to the high loading requirement in resonant operation.

If yours turn out to be degraded, and you replace the 30 nF originals
(now probably unobtainable) with 33 nF, you may need to re-adjust the
resonance frequency somewhat.


3.

Also, the frequency adjustment may be slightly out of resonance (maybe
the previous repairer has misadjusted it and component parameters can
also drift over the years). Again, a misadjusted frequency, especially
if it has been set too high rather than too low (compared to the true
resonance frequency of the LC circuit) can cause the dissipation
resistor to overheat (so a little low is better than a little high).

A resonant circuit driven too slow (below resonance), will pull reacive
power (will have a power factor below unity), but the direction of the
phase shift will be inductive. If driven a too fast (above resonance),
it will appear capacitive instead. Please note that the square-to-sine
conversion circuitry, especially the snubbers, will have lower stress
from peak currents when driving an inductive load than when driving a
capacitive load, so an inductive load is "easier" on them.

Please read the instructions in the service manual (I've also copied the
relevant part in my other post), and also note that the service manual
clearly advises to always adjust the frequency "from below" and never
"from above" ("use 170 V mains, then set the trimmer to lowest possible
frequency, and slowly raise it until the output voltage regulation can
just be obtained, but no more than this"). So the designers from Philips
must have preferred this design to run rather slightly below resonance
than slightly above it, and they must have had good reasons to write the
adjustment instructions in such a way, as to prevent an accidental "too
high" frequency setting.

Note that any frequency adjustment should be done with the correct dummy
load connected in order to avoid entering a "light-load" mode.

Regards
Dimitrij

 

[Cursitor Doom]

On Sun, 28 Feb 2016 03:01:02 +0100, Dimitrij Klingbeil wrote:

1.

There is still the question with V1808. You said it looks ok, and it
tested ok with a multimeter, but that's not really indicative of its
true behavior under full load at high frequencies. If it has degraded
for any reason ("lost its switching speed") then the resistor R1814
would be running at a higher load than normal. Not many times higher,
but about double or triple. That would be somewhat consistent with your
observation of it running too hot after a few minutes. You should now
have (hopefully) a few spare UF4007s, so if in doubt, replace V1808.

Yes, I bought 20 of those faster diodes to be on the safe side.

If you find out that the replacement of V1808 makes a (little) change
for the better (slightly lower load on R1814), then replace V1809 too.
It would in this case be likely that those BY208-1000s have all degraded
and became out-of-spec. They all have the same type and age.

Actually it's possible to test the condition of V1808 in circuit,
without replacing it, but the test is tricky. You would need to see, on
an oscilloscope, the voltage waveform across R1814.

[live power resistor procedure testing snipped]

Actually I did do this a while back without knowing the risks! As you can
see, I survived to tell the tale. All I was seeing was about 30V of noise
across that resistor but that was before I was informed of the importance
of hooking the supply up to a load, so the test was probably invalid.

> Ok, so much for the other BY208s in snubber circuits. Replace and see.

Certainly can do that, yes.

2.

The other open question is that of the resonance capacitors (C1807 and
C1808). As I noted in another post, they may be degraded and it may be
difficult to test for this condition properly (LCR meter won't likely
show the problem).

Is there any way of *definitively* testing such a capacitor against all
its possible failure modes? And I'd be interested to know where you get
this figure of 800V you mention from?

A resonant circuit driven too slow (below resonance), will pull reacive
power (will have a power factor below unity), but the direction of the
phase shift will be inductive.

Fortunately this is one aspect I pretty much totally understand. As an
old-style radio ham of more decades than I care to recall, the concepts
of resonance, reactance, impedance, power factor and phase shift are like
second nature so please don't go to any trouble explaining the finer
points in extreme detail; there's absolutely no need. BTW, your
explanations are unusually clear and thorough, I've noticed. If you don't
already, you really should edit or author technical manuals. It's an all-
too rare talent nowadays.

 

[Cursitor Doom]

On Sun, 28 Feb 2016 15:33:02 +0100, Dimitrij Klingbeil wrote:

But could you please make a complete list of found faults and your
replacements, and post it here:

I mean, you posted at the very beginning (long before finding the slow
diode) that you've found and replaced some obviously defective parts,
but I can't remember if you ever posted, exactly which ones they were.

I think you may possibly be getting mixed up with a different repair
here, Dimitrij. I do have some flaky capacitors to replace when I return
and I'll note which ones I change for your information. As for what
previous technicians may have done, I have no idea what if anything has
been replaced - apart from that one obvious diode. I got absolutely no
background information on this scope, it was given to me for nothing by
some guy who was emigrating so its past will now always remain a mystery.
It's a pity, because this obviously adds another set of unknowns into
troubleshooting the thing, but it's just something I'll have to live with
I guess. In all honesty, this repair is proving to be a 'baptism of fire'
for me in the world of SMPSs of which I admit I know very little (yet a
lot more than I did 3 months ago!)

 

[Dimitrij Klingbeil]

On 27.02.2016 01:42, Cursitor Doom wrote:

On Thu, 25 Feb 2016 00:36:16 +0100, Dimitrij Klingbeil wrote:

[...]

Dimitrij, I think you may have missed this I posted elsewhere so I'm
re- posting it here now for you personally:

"Final update for the time being as I have to leave soon now:

That short turned out to be intermittent. I hope it was just due to
something shorting out on the bench that won't happen when the
casing is back on because you all know what a bitch it can be to
trace intermittent faults. Anyway, that fault has now disappeared, so
I took some voltage measurements before the 20W resistor got to hot
(from 19'C to 60'C takes about 1.50s now) and I have:

61.7 12.7 5.8 0 -5.8 -12.7 -62.4

This is with the psu board plugged into the scope and all power
connections made except for the VHT stuff.

The correct figures according to the manual should be:

60 12.7 6 0 -6 -12.7 -60

So very close! Looks like the main transformer may be ok after all."

Making progress!

Hi

Noted your progress

But could you please make a complete list of found faults and your
replacements, and post it here:

I mean, you posted at the very beginning (long before finding the slow
diode) that you've found and replaced some obviously defective parts,
but I can't remember if you ever posted, exactly which ones they were.

Also, you have indicated other things that may impair reliability (like
capacitors with pieces of film isolation flaking off), and again, you
didn't seem to indicate the exact schematic part numbers.

As you may well know, to troubleshoot anything properly and reliably,
and to be able to assess the likely chains of cause and effect, one
needs to know the history of the repairs, as completely as possible, and
also anything obviously (visually or otherwise) suspicious too.

Therefore please make some lists, and take particular care to make them
complete, to leave nothing out, and to indicate each and every listed
part's schematic part number (important, since others can't see your
board and need the exact numbers to identify the parts in question).

- one list with all previous repairs that you have found: which parts
were replaced in the past, as visible from manual solder joints, and
where the replacements were of different type from the original, clearly
indicate the exact types of replacements.

- one list with all of your repairs: which parts you found defective and
what exact parts (exact type and manufacturer) you have replaced them with.

- one list with all parts that currently look suspicious or for whatever
reason seem to be of questionable integrity.

It would be nice if you could make a printout of the schematic, and mark
all those items in color (like for example yellow for previous repairs,
circled twice if the repair was inexact, red for those you replaced, and
blue for the suspicious ones), and then scan and post the color-
annotated schematic somewhere for us to see.

To avoid "... and what else was there?" or "... and what about part
XYZ?", please make sure that this annotation is really complete. Trying
to get such information one question at a time can be frustrating.

Regards
Dimitrij

 

[Cursitor Doom]

On Sun, 28 Feb 2016 18:55:47 +0100, Dimitrij Klingbeil wrote:

[...]

I think that the most realistic test would be to sweep the resonant
circuit with a signal generator and watch the waveform. If the resonance
frequency looks right (in the 20 kHz ballpark) and a signal generator is
able to drive it from a high 600 Ohm source impedance to a significant
amplitude without much "sagging" (that is, the resonant circuit presents
little load to the generator), it's probably OK.

Thanks again, Dimitrij. You're obviously an expert on the little
understood world of resonant converters so when you say try this or that,
I make a point of paying extra attention. I liked your theory on the
resistor heating due to this supply running out of resonance as a result
of component values changing over time; in fact I'm currently pinning my
hopes on it. It's a pity I'm stuck here for a few more days with my
revolting in-laws but it'll be the first thing I do on my return!

Somewhere I have a big old valve/tube capacitor tester capable of
simulating realistic high voltage working conditions. It'd be interesting
to know what kind of checks it's capable of performing if it's still in
working order and if I can find it among the towering piles of obsolete
test equipment I have here (a couple of million pounds worth of gear at
new prices adjusted for inflation) I may possibly hook it up and give it
a shot.

How about those 'Octopus' component testers? They subject the part under
examination to sweeping test voltages over the expected working range and
you look for any signs of breakdown on an oscilloscope in X=Y mode. I
guess this method is about as good as it gets?

 

[Cursitor Doom]

On Sun, 28 Feb 2016 19:06:38 +0100, Dimitrij Klingbeil wrote:

Also, even with a dummy load connected, the stray capacitance of an
oscilloscope, when hanging off the loose end of a power circuit with
some 800 to 900 V worth of HF on it, would probably cause so much undue
capacitive loading that the power supply circuitry would hardly handle
it.

Isn't this just another example of the unsatisfactory nature of this
resonant converter design? If the thing is *that* fussy that a little bit
of stray capacitance can catastrophically destabilise it, then AFAICS
it's a fundamentally unreliable topology and it would be better to have
used one of the non-resonant forms of converter. Unless there's some
compelling reason I may be unaware of not to for oscilloscope power
supplies, of course.

 

[Cursitor Doom]

On Sun, 28 Feb 2016 19:23:20 +0100, Dimitrij Klingbeil wrote:

P.S. That voltage estimate has probably surprised you. Unless one looks
at the circuit schematic and adds all the voltages from all the storage
elements (inductors / capacitors), considering timing and phase, it may
not be obvious that the thing was intended to run at such high voltage
levels. But there's a reason why they used a 1500 V transistor in it.

And yet C1804 is rated at 'only' 630V. Weird!

 

"Hhmmm. As I've said before, I'm reluctant to replace that power >resistor
with anything higher rated. "

It is NOT higher rated. A 100 watt incandescent in that spot will limit the current even lower and there is less chance of blowing anything else.

I did not mean to suggest that you change your plan of troubleshooting, just that next time when it comes to a hot test, use the bulb. If something is still shorted you have a hell of alot more time to figure out what, rather than having overheat in seconds. When the light dims, you probably found the problem. the light goes down in resistance as all the filters charge and get almost to full voltage, once it does that it works without a net. Regular fuse and all that.

I consider a dim bulb tester a must for this type of work.

 

[Dimitrij Klingbeil]

On 28.02.2016 16:18, Cursitor Doom wrote:

On Sun, 28 Feb 2016 03:01:02 +0100, Dimitrij Klingbeil wrote:

1.

There is still the question with V1808. You said it looks ok, and
it tested ok with a multimeter, but that's not really indicative
of its true behavior under full load at high frequencies. If it
has degraded for any reason ("lost its switching speed") then the
resistor R1814 would be running at a higher load than normal. Not
many times higher, but about double or triple. That would be
somewhat consistent with your observation of it running too hot
after a few minutes. You should now have (hopefully) a few spare
UF4007s, so if in doubt, replace V1808.

Yes, I bought 20 of those faster diodes to be on the safe side.

If you find out that the replacement of V1808 makes a (little)
change for the better (slightly lower load on R1814), then replace
V1809 too. It would in this case be likely that those BY208-1000s
have all degraded and became out-of-spec. They all have the same
type and age.

Actually it's possible to test the condition of V1808 in circuit,
without replacing it, but the test is tricky. You would need to
see, on an oscilloscope, the voltage waveform across R1814.

[live power resistor procedure testing snipped]

Actually I did do this a while back without knowing the risks! As
you can see, I survived to tell the tale. All I was seeing was about
30V of noise across that resistor but that was before I was informed
of the importance of hooking the supply up to a load, so the test
was probably invalid.

Ok, so much for the other BY208s in snubber circuits. Replace and
see.

Certainly can do that, yes.


2.

The other open question is that of the resonance capacitors (C1807
and C1808). As I noted in another post, they may be degraded and it
may be difficult to test for this condition properly (LCR meter
won't likely show the problem).

Is there any way of *definitively* testing such a capacitor against
all its possible failure modes? And I'd be interested to know where
you get this figure of 800V you mention from?

A resonant circuit driven too slow (below resonance), will pull
reacive power (will have a power factor below unity), but the
direction of the phase shift will be inductive.

Fortunately this is one aspect I pretty much totally understand. As
an old-style radio ham of more decades than I care to recall, the
concepts of resonance, reactance, impedance, power factor and phase
shift are like second nature so please don't go to any trouble
explaining the finer points in extreme detail; there's absolutely no
need. BTW, your explanations are unusually clear and thorough, I've
noticed. If you don't already, you really should edit or author
technical manuals. It's an all- too rare talent nowadays.

Ok. Usually most people who ask here understand DC parameters well
enough, but rarely get to consider impedance, phase angles and such.

As for the 800 V, that was mostly a guess. Basically I've taken 320 V of
the storage capacitor, added to that another 300 V of the resonant
circuit (when the power transistor is off and it's being swung in the
other direction) plus the voltage rise from the winding reset from the
primary of L1806 (which is actually unknown since I don't know the ratio
between primary and secondary, the secondary being at 320 V), which I
guessed to be somewhere in the 200 V ballpark.

That's 320 V + 300 V + 200 V = 820 V, likely even to be more because the
300 V may reach up to 320 and the 200 is only a guess and may likely end
up higher than that, plus there may be some 50 V from L1804 adding up in
the same polarity, so even a 900 V total won't be out of the question.

That would be consistent with the rating of the BU208 power transistor,
which has a 1500 V absolute maximum collector rating when driven from a
low-impedance base drive signal.

As for definitely testing the resonance caps: I'm somewhat at a loss.

First thing, you can measure the capacitance, that an obvious test. If
the capacitance is wrong, they're can't be working properly.

But reduced current handling ability comes from an increase in ESR and
in the dissipation factor. To measure them, you would need to run the
cap at the intended target frequency (and preferably at a realistic
voltage too).

LCR+ESR meters can measure the dissipation factor and ESR, but those
intended for electrolytics will often measure only ESR and also may have
trouble testing such small foil capacitors like 33 or 15 nF.

Also, I don't know the target numbers for ESR and dissipation here, so
one would need to compare them against a known good pair somehow.

An other way I can think of, would be to run them at resonance with the
transformer, and measure both frequency and "Q". But that's also not
meaningful unless one has a known good reference value for Q.

I think that the most realistic test would be to sweep the resonant
circuit with a signal generator and watch the waveform. If the resonance
frequency looks right (in the 20 kHz ballpark) and a signal generator is
able to drive it from a high 600 Ohm source impedance to a significant
amplitude without much "sagging" (that is, the resonant circuit presents
little load to the generator), it's probably OK.

Dimitrij

 

[Dimitrij Klingbeil]

On 28.02.2016 16:18, Cursitor Doom wrote:

[live power resistor procedure testing snipped]

Actually I did do this a while back without knowing the risks! As you
can see, I survived to tell the tale. All I was seeing was about 30V
of noise across that resistor but that was before I was informed of
the importance of hooking the supply up to a load, so the test was
probably invalid.

Also, even with a dummy load connected, the stray capacitance of an
oscilloscope, when hanging off the loose end of a power circuit with
some 800 to 900 V worth of HF on it, would probably cause so much undue
capacitive loading that the power supply circuitry would hardly handle
it. That may have been the reason why you just got noise (the overload
from the hanging scope may have affected the over-current shutdown of
the power supply controller). As I said, the proper way would be with an
isolated high voltage differential probe (such a probe would present
very little stray parasitics) or maybe with a well matched pair of
(identically compensated) HV probes in subtract mode.

Dimitrij

 

[Dimitrij Klingbeil]

On 28.02.2016 19:06, Dimitrij Klingbeil wrote:

On 28.02.2016 16:18, Cursitor Doom wrote:

[live power resistor procedure testing snipped]

Actually I did do this a while back without knowing the risks! As
you can see, I survived to tell the tale. All I was seeing was
about 30V of noise across that resistor but that was before I was
informed of the importance of hooking the supply up to a load, so
the test was probably invalid.

Also, even with a dummy load connected, the stray capacitance of an
oscilloscope, when hanging off the loose end of a power circuit with
some 800 to 900 V worth of HF on it, would probably cause so much
undue capacitive loading that the power supply circuitry would hardly
handle it.

P.S. That voltage estimate has probably surprised you. Unless one looks
at the circuit schematic and adds all the voltages from all the storage
elements (inductors / capacitors), considering timing and phase, it may
not be obvious that the thing was intended to run at such high voltage
levels. But there's a reason why they used a 1500 V transistor in it.