Safely testing 22 kV capacitors

[Rich256]

Thanks, now I understand a little better. I will soon make a stack of,
say, 30 1N4007 diodes. That would let me test the capacitors with 13
kV.

i

Did something like that about 40 years ago. In our case we had to
bridge each diode with a small capacitor and large resistance resistor
because when seeing the reverse voltage, they did not all turn off fast
enough.

 

[Ignoramus27098]

Did something like that about 40 years ago. In our case we had to
bridge each diode with a small capacitor and large resistance resistor
because when seeing the reverse voltage, they did not all turn off fast
enough.

Hm... Were you trying to rectify a 60 Hz sinewave?

i

 

[Dave Hinz]

Hm... Were you trying to rectify a 60 Hz sinewave?

Well, first thing you need is a big capacitor...

 

[Rich Grise]

I am aware that I cannot built a divider with regular resistors, and
high voltage resistors might be hard to come by.

Of course you can!! Do you have a piece of plexiglass about 15"-18" long,
maybe 1"-1.5" wide? Just put 100X 1M resistors in series, and pick off
1/100 of the voltage at the bottom resistor. Silicone them down in sort
of a zig-zag, solder the leads one to the next, trim them - leave a round
solder glob at each junction to cut down on corona, and slather silicone
all over the top and around the assembly.

Use a very high-impedance, input-protected voltmeter, like a VTVM ;-)

Good Luck!
Rich

 

[Rich256]

Hm... Were you trying to rectify a 60 Hz sinewave?

i

Both rectifying 400 Hz and use as a series diode in a resonant charging
circuit. Actually I was using 1 ampere silicon rectifiers which were
not fast enough. All it takes is one not turning off fast enough.

 

[Rich256]

Both rectifying 400 Hz and use as a series diode in a resonant charging
circuit. Actually I was using 1 ampere silicon rectifiers which were
not fast enough. All it takes is one not turning off fast enough.

I don't remember just what we ended up using. I think it was about
..001u and 10 Meg across each diode.

Another problem was that a few diodes had bad junctions and couldn't
take the current. When you put 30 of them in series the probability of
getting a bad one is pretty high. Had to weed them out first.

 

[Jeff Wisnia]

If you were observant enough of the thread, you would find that even
the original poster mentioned the HV probe loading.

I guess my powers of observation are failing me Roy, but I always like
to check out where I missed something and try and figure out why it
escaped me.

Would you kindly point out where the OP revealed that he realized a HV
probe would serve as a safe discharging means, albeit a rather slow one?
(Before I mentioned it that is.)

Thanks,

Jeff

--
Jeffry Wisnia

(W1BSV + Brass Rat '57 EE)

"Truth exists; only falsehood has to be invented."

 

[Jeff Wisnia]

Big Van DeGraf machines produce microamps! Charging a uF to 20,000 V
at one uA will take 20,000 seconds. And, that's one of those 6 foot
tall jobs with a 6" wide belt.

Jon

Live and learn. I was remembering the ones I played with years ago which
IIRC put out more like 20 ua, but I do agree that even that much current
would make for an uncomfortably long wait. <G>

I remember having to test the radiation hardness of some satellite
equipment I designed. We went to a GE facility in Pennsylvania where
they had a "flash x-ray" generator.

It used a BMF horizontally oriented Van De Graf IIRC about 25 feet long
and two feet in diameter which took many minutes of running to build up
the requisite high voltage in a storage capacitor. The charge was
flashed through an x-ray tube to produce a humungous blast of x-ray
photons into my poor little circuits.

Thanks for the mammaries,

Jeff

--
Jeffry Wisnia

(W1BSV + Brass Rat '57 EE)

"Truth exists; only falsehood has to be invented."

 

[Cydrome Leader]

I doubt that the shock from a 1u cap at 20KV is going to kill unless
still hooked to the power supply. However, the reaction (falling
backwards and hitting your head) sure can do damage.

People are regularly killed off the filter caps in microwave ovens.

 

[Mark Zenier]

Ian, would you or anyone have any idea if what I have is HV resistors?


They are blue, about 3 inches long, and are marked VICTOREEN MOX-3 12
MEG.

You need to get a collection of EEM catalogs. Back in the 1993 edition,
the Victoreen pages in the Resistor section says that it's most
likely a MAXI-MOX rated at 22.5 kV. Rated at 4.5 watts if it's a
MOX-3-13, epoxy coated or 7.5 watts for a MOX-3-12 silicone coated.

Mark Zenier [email protected]
Googleproofaddress(account:mzenier provider:eskimo domain:com)

 

[RoyJ]

Why not keep them and start trying to shink coins?
http://205.243.100.155/frames/shrinkergallery.html

This site has been recently edited to take out most of the technical
details. I suspect it was a bit too dangerous for the average user.

 

[DoN. Nichols]

According to Ignoramus27098 <[email protected]>:

[ ... ]

Thanks, now I understand a little better. I will soon make a stack of,
say, 30 1N4007 diodes. That would let me test the capacitors with 13
kV.

You can make 30KV from a single 1KV 60 HZ transformer, with
enough diodes and capacitors. The basic circuit of a voltage multiplier
is:

(AC)---+--)|-+------+--)|--+-----(DC)
| | | |
\---/ --- \---/ ---
\ / / \ \ / / \
--- /---\ --- /---\
| | | |
(G)-)|--+-----+-)|---+------+

That one is enough for a 4X multiplication of the voltage. Keep
adding pairs of diodes and capacitors and you keep adding voltage
multiplications -- and losing current capacity.

1 KV AC is 1.414 KV DC peak, times 4 gives you 5.656 KV just
with this simple circuit. At work, we used fairly small assemblies
potted in epoxy to get 45 KV (three 15KV taps from a 1KV P-P input.

But -- it might take nearly forever to charge your caps.

Enjoy,
DoN.

 

[DoN. Nichols]

Of course you can!! Do you have a piece of plexiglass about 15"-18" long,
maybe 1"-1.5" wide? Just put 100X 1M resistors in series, and pick off
1/100 of the voltage at the bottom resistor. Silicone them down in sort
of a zig-zag, solder the leads one to the next, trim them - leave a round
solder glob at each junction to cut down on corona, and slather silicone
all over the top and around the assembly.

Use a very high-impedance, input-protected voltmeter, like a VTVM ;-)

You can't get more high-impedance than the electrostatic
voltmeter which I posted a link to last night. (Probably the auction is
closed by now.) But that kind of voltmeter would allow monitoring the
voltage real-time until you decide that it is now time to discharge the
capacitor.

Enjoy,
DoN.

 

[Mike Henry]

[alt.marketing.online.ebay -- I am talking about an item that I won
on ebay recently, see

http://cgi.ebay.com/ws/eBayISAPI.dll?ViewItem&item=7603382621
]

It's more likely that lab "junk" is sold off for salvage and that the
salvagers or someone they've sold to are auctioning off the stuff. It
seems
really unlikely to me that anyone at Fermi Lab, or DOE for that matter,
would bother to check Ebay seller names for usurpers of their "name". I
haven't checked, but would guess that Fermi Lab's budget is at least a
few
hundred million dollars per year and technical staff probably bill out at
around $125 and up per hour. They'd have to get pretty good prices for
those caps at that rate to break even - did they?

Mike, I am going to Fermilab to pick up these capacitors on Tuesday
and they want payment in form of checks made to Fermilab. They take my
check on the spot and do not require it to clear. I would be greatly
surprised if that was not a legitimate Fermilab operation.

I think that what we are observing is typical government/public funded
surplus bullshit. I see the same stuff all over in military surplus.
Everyone is going through the motions ("we are auctioning surplus to
get highest competitive value"), but no one really cares to even
understand what they are selling or what it costs to sell that
stuff. Some things sell well for them (like high dollar electronic
test equipment sold one piece at a time), some do not.

Now if this item was adequately described with love and care, then, I
would suspect that it was not Fermilab but some private person who
actually cares to get money for his stuff.

Well, it's pretty hard to argue with that scenario.

I work at a similar facility and will have to see if they are doing the same
sort of thing. Just the other day a Hardinge bed was sitting outside
waiting for disposition of some sort. The stuff that gets thrown out there
would make a die-hard scrounger cry, though most of it is probably entirely
used up.

Mike

 

[Roy L. Fuchs]

It is, doubtless, definitely Fermilab. I will go to him on tuesday
into Fermilab itself. I find it difficult to imagine a private
individual sellings his own things in this manner.

Yeah. Those caps have a very limited number of pulse cycles where
they are guaranteed to maintain their rated specs. After that they
are typically swapped out of a system for new components. The old
ones get used in the lab or get sold as a lot. That is very likely
what those are.

 

[Mac]

News flash: the person doing the writing gets to decide how, or if, his
work is distributed and stored. Demanding that he do it differently is
no different than the people who ask a question and demand an email
response. (equally stupid and arrogant, is the point I'm making).

Maybe Ignoramus is within his rights.

But Win has a very good point. In any event, if Ignoramus alienates Win
and just a few other knowledgeable regulars here, his (Ignoramus's)
experience will be badly degraded.

So he may want to think seriously about Win's protest.

--Mac

 

[Ignoramus26172]

According to Ignoramus27098 <[email protected]>:

[ ... ]

Thanks, now I understand a little better. I will soon make a stack of,
say, 30 1N4007 diodes. That would let me test the capacitors with 13
kV.

You can make 30KV from a single 1KV 60 HZ transformer, with
enough diodes and capacitors. The basic circuit of a voltage multiplier
is:

(AC)---+--)|-+------+--)|--+-----(DC)
| | | |
\---/ --- \---/ ---
\ / / \ \ / / \
| | | |
(G)-)|--+-----+-)|---+------+

That one is enough for a 4X multiplication of the voltage. Keep
adding pairs of diodes and capacitors and you keep adding voltage
multiplications -- and losing current capacity.

1 KV AC is 1.414 KV DC peak, times 4 gives you 5.656 KV just
with this simple circuit. At work, we used fairly small assemblies
potted in epoxy to get 45 KV (three 15KV taps from a 1KV P-P input.

Thanks DoN. I think that I will just make a 13 kV charger with my 9 kV
AC source and a diode.

i

 

[Ignoramus26172]

"Ignoramus27098" <[email protected]> wrote in message

I work at a similar facility and will have to see if they are doing
the same sort of thing. Just the other day a Hardinge bed was
sitting outside waiting for disposition of some sort. The stuff
that gets thrown out there would make a die-hard scrounger cry,
though most of it is probably entirely used up.

What is worn out for some people, is treasure to other people...

i

 

[Ignoramus26172]

Yeah. Those caps have a very limited number of pulse cycles where
they are guaranteed to maintain their rated specs. After that they
are typically swapped out of a system for new components. The old
ones get used in the lab or get sold as a lot. That is very likely
what those are.

Hm... So... DO you think that they are now useless, or simply a little
out of spec?

i

 

[Rob]

Given that they were made by Maxwell and sold to
Fermilab, they are probably spec'ed for pulse
discharge. That said, I'd not want to be on the
same city block when iggy crowbars them.

Putting a hard metalec short across them will no
doubt blow half the terminal away, create a deafening
blast and generate a nice EMP pulse. There's no
way I'd do it.

I worked in a medical laser research group. We had 2u odd caps @ 10kv+ as
part of a home brew laser supply. Before working on it we'd double check the
caps were disharged in case the discharge resistors had failed.

A couple of PVC pipes 2m long with a 12" length of heavy neon sign cable
linking them, a couple of M4 screws were stuck through the ends.

When the caps were charged it was like a gun going off - very sharp bang, it
usually bought people out of offices all down the hallway. You never stood
behind whoever had the rods in case you got a reacting elbow in the face.

r.