Using "Energy Storage" capacitors as filter caps?

[W9LBB]

I feel sort of dumb having to ask this one, mainly because I used to work for a
company that MADE oil filled capacitors (Plastic Capacitors Inc.).      The problem
is that I never encountered this issue before; I thought I'd pick the brains here
instead of possibly making a mess of this.

I have two beuatiful energy storage capacitors; 32 MFD @ 4500 VDC. Obviously,
something THIS big in a supply will make it VERY ripple free! However, I'm hearing
sparse & conflicting information about the suitability of energy storage caps as filters.

Rather than trying to use these caps at full voltage rating, I'm planning to run them in
a supply delivering about 2500 VDC @ 500 MADC.

Just what's the difference between a filter cap and an energy storage cap? Using a
storage cap as a filter, what is the critical parameter that becomes a limiting factor
in thier use?

Inquiring minds wanna know!


73's,

Mr. T., W9LBB

[W1RKW]

I could be wrong here but I don't see the difference myself.  I always thought that a capacitor was an energy storage device.

[W3SLK]

Bob said:

Insert Quote
I could be wrong here but I don't see the difference myself.  I always thought that a capacitor was an energy storage device.

Just like your flux capacitor!

[K1JJ]

These commonly called "photoflash" capacitors that are about the size of a small loaf of bread and non-polarized, are very nice compact caps for power supply filtering.  It is common to see, "51uf at 5KV" in this small package.

I've run them for years at close to their rated HV ratings with no failures to date.

73,
T

[Tom WA3KLR]

I believe that "energy storage capacitors" are the ones that are engineered for a very high peak current discharge pulse.  Essentially all of the energy in the capacitor is removed in one fast pulse, unlike the normal filter capacitor application where the majority of the applied voltage remains in the capacitor.  The ripple current in the filter capacitor application is therefore much less than in the energy storage capacitor.  

I did review a bonifide energy storage capacitor data sheet many years ago.  The engineer I was working with at the time was designing a transient generator that would put a transient out on the power line equal to a lightning strike surge.  This is IEEE 487 surge specification as I recall.

Trying to use a normal filter capacitor in this application would not cut it.  If you really dig into the peak current spec. for the aliminum electrolytic capacitors, there is a max. spec. of approximately  a few amps.  Their data sheet usually recommends X ohms per volts when discharging a capacitor, contrary to the hammy hambone practice of discharging the fiiter cap. into a dead short - screwdriver, chassis, etc.  Bad for the filter cap.

Energy storage capacitors are engineered to deliver the peak current - perhaps 100's or thousands of amps peak and in repeated service with good long-term reliability, probably oil-filled metal-paper technology.

So an energy storage cap. can be used as a filter capacitor in my opinion, the service is no stress whatsoever.

But to put a normal filter capacitor in energy storage service is very bad for that capacitor.

[W9LBB]

Greetings, Gentlemen.

Thanks for the replies on this; they make a lot of sense to ME. 

The reasons for using these caps are simple. First, they're available for the project.

Second, a choke input filter with 32 MFD in the output cap is GREAT for an AM rig
that uses a shunt reactor in the modulator; 32 MFD has VERY low audio impedance!

The caps were obtained from a cap bank used to fire a HUGE flash tube in a laser
power supply (some very strange parts surface in a university town that has a big
engineering school). The guy I got them from was talking about reduced ratings on
the caps in filter service because there they're dealing with "dynamic" energy
storage (removing ripple), while in energy storage they're dealing with "static" energy
storage, charged up for one huge shot.

It sounds like these cans are a "go" for the plate supply here.    And at 2500 VDC
@ 500 MADC I doubt that there's gonna be ANY problems.

The only thing I'll have to keep in mind; filter caps this big are packin' a LOT of Joules,
and an accidental flashover in a tube will probably destroy the jug unless there's a
current limiting resistor in the B+ line. I figure 100 ohms at a hundred watts or so
can take care of things, and won't drop B+ unduly.


Thanks for the input, Gentlemen! 


Tom, W9LBB

[WBear2GCR]

Afaik, don't use Electrolytic "photoflash" caps as the FIRST stage in a PI filter on a capacitor input filter. The reason being is that they are not designed to handle the ripple current... they're super peachy keen for the second stage. I presume they'll be fine after a choke input supply...

    _-_-WBear2GCR

PS. Good find on the big earl filled caps!!

[W9LBB]

Afaik, don't use Electrolytic "photoflash" caps as the FIRST stage in a PI filter on a capacitor input filter. The reason being is that they are not designed to handle the ripple current... they're super peachy keen for the second stage. I presume they'll be fine after a choke input supply...

    _-_-WBear2GCR

PS. Good find on the big earl filled caps!!

Howdy, Bear.

Nope, I don't believe in capacitor input filters here; the voltage regulation sucks, not to
mention the fact that they're HARD on rectifiers. For my designs, it's choke input or
nothing, and the FCC regs won't let me run NOTHING. 

Re electrolytics; something else I don't believe in for anything except VERY SMALL power
supplies. The caps in question are about a foot high and maybe 5" square, and filled with
nice, non-PCB oil.

When I design a big rig, halfway measures are NOT employed. Ever.


73's,

Tom, W9LBB

[WBear2GCR]

Aye Matey! Real men - yar - test 'em with their tongues! 

K1JJ mentioned the photoflash caps...

        _-_-WBear2GCR

[WA1GFZ]

I thought they had a problem with frequency since they are designed to charge slowly and discharge fast. I have a few and notice the insulators are a bit small. fc

[K1JJ]

I also dislike electrolytic caps for supplies over 800 volts or so.

My previous post refers to non-polarized photo flash caps for HV use. They work great and I usually buy all I can get my hands on at the flea markets...  That's all I use for HV supplies anymore.  I notice the Russian vendor has his own version.

Yes, Franz, the insulators ARE small, I think too. But never had one flash over yet.


T

[WA3VJB]

The only difference I see, and it usually doesn't matter, is that the construction of an "energy storage type" capacitor has the case and one terminal grounded. I imagine some applications require the capacitor to float above ground at their given point in the circuit, but I can't recall one. Regular HV capacitors, like the oil type, have both terminals up on ceramic standoffs.

[k4kyv]

I have seen them both ways, with one terminal grounded and with both insulated.  I have heard over the years that the voltage specs of photoflash caps should be derated  about 25% for continuous duty.  I have had a couple in use for over 30 years with no mishap.  I, too, use only choke input filters.

One thing I have noticed about the physical construction is that with oil-filled filter caps with rectangular footprints, the terminals are placed at the far ends of the rectangle, so that they are as far apart as possible.  With photoflash caps, the terminals are placed midway along the sides of the rectangle, leaving the terminals closer together.