The AM Forum
March 29, 2024, 09:39:45 AM *
Welcome, Guest. Please login or register.

Login with username, password and session length
 
   Home   Help Calendar Links Staff List Gallery Login Register  
Pages: [1]   Go Down
  Print  
Author Topic: high voltage oil caps and their volts-to-GND ratings  (Read 15377 times)
0 Members and 1 Guest are viewing this topic.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« on: December 22, 2014, 11:45:47 AM »

I never see a spec for the volts-to GND rating on a HV oil capacitor unless it is implied, as with a 'Can Common' type where one terminal is the casing and the other the voltage terminal.
Here is a 6KV 1uF cap with 2" to 3" insulators. There must be a reason for that. Other 5-6KV caps have only 1" to 1.5" of insulator height.

I want to make up a series parallel bank of 8 of them to have 4uF/12KV for use with a 6400V transformer, bridge rectifier, and a choke input filter scheme.
I would prefer to bolt them all to a chassis. Trying to insulate them would be very inconvenient.

What are the general considerations for putting two metal-cased caps with big insulators like this in series for higher voltages?


* 1u6000v.jpg (96.2 KB, 599x900 - viewed 615 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
WBear2GCR
Member

Offline Offline

Posts: 4135


Brrrr- it's cold in the shack! Fire up the BIG RIG


WWW
« Reply #1 on: December 22, 2014, 04:41:53 PM »

I be no expert, butt, I expect that the caps  themselves will only see half the voltage when they are in series. But the voltage to ground on the first cap will be the max B+, looks like it will be how much with your proposed power supply??

That might be an issue.

I think that you may be able to increase the effective breakdown voltage (if needed) by placing some 1/8" acrylic or polycarbonate rings over the ceramic insulator and affixing them with judicious use of silicone rubber. There is a "flowable" silicone rubber that night fit the bill for a neat adhesive job. Epoxy might work too, but I think it has maybe lower breakdown voltage.

The insulator height plays a role WRT the breakdown of air vs. voltage - They make the HV insulators on power lines have shape to increase the surface area and effective length.

Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #2 on: December 23, 2014, 12:04:21 AM »

Sorry about the capacitor picture missing. It's posted now in the previous post.

The peak voltage from the transformer is 9050V, so the max I can see with LC filter and 100mA bleeder is about 8KV during transients etc., or less if a soft start is used (as shown).
I plan to use two 866 sticks in series for each "3B28". The choke is probably going to be a swinging choke, will be in the negative lead.

The basic capacitor is going to be a stack of two 1uF/6KV caps in series for a 12KV rating. - multiply this by four in parallel, for a series parallel 2uF/12KV stack.

My question is to whether the terminal-to-case insulation on the shown capacitor is likely to be substantially more than 6KV.

The thing that makes me say yes, is the quite tall insulators as opposed to shorter ones on other 5-6KV caps. So that is what I am asking about if it makes sense.

If I could find a nice 4-6uF 12KV or better cap then I would not have this issue, but we use what we got! 8-)


* dennis-psud.png (42.87 KB, 945x732 - viewed 629 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #3 on: December 23, 2014, 12:25:03 AM »

I don't think the insulators are an issue.  If the cap in going to break down it will be where the insulator passes through the casing.  Or the cap may short internally to the case.  You probably need to insulate the first cap in the string from ground.

Fred
Logged
WQ9E
Member

Offline Offline

Posts: 3282



« Reply #4 on: December 23, 2014, 05:18:49 AM »

Patrick,

The voltage to ground (actually to case which normally is grounded) rating is the same as the capacitor voltage rating.  The design voltage rating is for conditions seen during normal operation so if you want to use oil caps in series to increase the voltage rating then insulation from ground is needed and pay attention to the usual safety issues of having normally grounded exposed surfaces now floating above ground. Of course there is some safety factor built into the caps but I wouldn't push it.

You should be able to build an insulating envelope for the floating case caps using insulation made from Teflon sheet.  I have used this for tube chimneys and it is easy and neat to work.  I think I got mine from McMaster Carr.   You could also use plexiglas or one of the variants if you want to build a solid mount rather than a compression mount using Teflon sheet material inside metal.

The unusually long insulator length is probably due to anticipated environmental conditions for that particular cap specification (humidity/dust/altitude).

Logged

Rodger WQ9E
W7TFO
WTF-OVER in 7 land Dennis
Contributing
Member
*
Offline Offline

Posts: 2521


IN A TRIODE NO ONE CAN HEAR YOUR SCREEN


WWW
« Reply #5 on: December 23, 2014, 11:36:02 AM »

Pat, don't get too worked up, I can send along a 4uF @ 12kV cap in your box (crate).

73DG
Logged

Just pacing the Farady cage...
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #6 on: December 23, 2014, 11:42:32 AM »

Pat, don't get too worked up, I can send along a 4uF @ 12kV cap in your box (crate).

73DG

Dennis to the rescue.  Send him the 12KV cap so he doesn't electroplate himself.

Fred
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #7 on: December 23, 2014, 11:57:28 AM »

Thank you Dennis. Soon as Christmas day is done I can ship your stuff out. I need to get some bubble wrap. am totally out of it. Could you please PM or e-mail me your receiving address again?
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #8 on: December 23, 2014, 01:36:02 PM »

here's the project, as it is coming together. Just the schematic, trying to finalize it, although the RF section is partly rebuilt. The power supply is still a pile of parts.

* 4-1000A_amp_DR_gg.pdf (33.9 KB - downloaded 308 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
John K5PRO
Member

Offline Offline

Posts: 1033



« Reply #9 on: December 23, 2014, 02:50:00 PM »

Did you use LTSPICE IV to generate the schematic? The meter symbols are neat.

Are you OK with floating the CT of the HV transformer? You are running it in FWB instead of FWCT operation, and that post/wire was probably designed near ground in the winding.

If you can, a little more C would be worthwhile, no? But then again, gotta use what you can get. There are lots of 4 - 5 kV caps hanging around, but the 12 kV rating is more rare round these parts. I wish the Black Hole was still in business, Ed would have had them.

Happy Holidays
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #10 on: December 23, 2014, 03:17:28 PM »

Thank you, I did use LTSpice. I made up some components to my taste. No electrical models for them.
"zip" files can't be uploaded here so I put it at http://www.bunkerofdoom.com/00_misc/ltspice_kd5oei.zip and anyone can take and use it for whatever.

The transformer looks like one of those rare ones with a CT that is fit to float. I say it because the CT has also a huge insulator same as the winding ends. I hope so!! Unfortunately there are no markings on it of any use and but one way to find out.

It's a zero budget project, totally junkbox, so buying a different transformer is not economical, unless there's one for very cheap somewhere. I thought about voltage doublers, one on each side so the load would be as a bridge, but that is very big due to many caps maybe more expensive.

The transformer is maybe 1.5-2KVA, 3KVA ICAS. The amp should be OK for AM at reasonable levels.

The cabinet should have enough room to add the relays and everything. I think the mostly empty backside of the front panel would be a good place to mount many of the smaller things, and if there is a wire harness rather than a bunch of wires everywhere then it could 'hinge' down for easy access while rebuilding this amp.

The higher voltage stuff and tall stuff could be put toward the back. I have not planned all the mechanical stuff yet. I want to mount the blower on rubber shock mounts to isolate vibration. It is not too hard and I have them. also need to repaint the cabinet etc etc. but it has been wet and cold.


* 100_0903.JPG (666.62 KB, 1711x1436 - viewed 641 times.)

* 100_0905.JPG (837.07 KB, 1453x2285 - viewed 645 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #11 on: December 23, 2014, 05:53:02 PM »

Any suggestions for the resistor R3 in the diagram?

It seems to have to be determined experimentally based on the discovered pull-in DC voltage of the AC-coil contactor.

Maybe the heating element 300W will have to also be changed higher, in case the volts never come up enough.


* softstart_.png (18.76 KB, 448x314 - viewed 641 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
KD6VXI
Contributing
Member
*
Online Online

Posts: 2648


Making AM GREAT Again!


« Reply #12 on: December 23, 2014, 07:22:01 PM »

Leave the Henry Heater (tm)  in,  and just use another resistor (proper value)  for the rf circuit (read pull in).

That resistor can be a half Watt job.   No reason to change the heating element out lol.

--Shane
KD6VXI
Logged
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #13 on: December 23, 2014, 09:02:21 PM »

Patrick,

That xfmr may just be 1KVA if that.  Most likely OK to run with a FWB judging from the center insulator.  If it's not OK for a FWB, you'll be the first one to know. Grin   I don't like potted xfmrs, I try to avoid using them.  I think open frame xfmrs are less likely to fail.

Fred
Logged
kc2we
Member

Offline Offline

Posts: 138


« Reply #14 on: December 24, 2014, 03:39:43 PM »

All capacitors like those used in the electric power industry have a nominal operating voltage, i.e. 7.2 or 7.6 KV terminal to terminal, but they have another rating called the BIL (Basic Insulation Level) rating. At JCP&L we have 7.2 KV rated line capacitors that are used for power factor correction on 12.47 KV primary lines (7.2 KV "Y" - 3 phase) that have a BIL of 110 KV. That 110KV BIL is for the insulators and capacitor internals to case ground as a "withstanding" value. Normally, the arcing distance for 60 HZ is about 10 KV r.m.s  per inch gap. Any longer insulator leakage paths are for designs that require insulation under less than ideal conditions, i.e wet conditions etc.. In most indoor amateur capacitor applications, I would think any pillar insulators longer than 1" are over kill. You don't need insulators that have a 110 KV breakdown for that use.

Seth KC2WE - Engineering: Energy delivery, Jersey Central Power & Light Co.
Logged

Seth Taylor
N2DTS
Contributing
Member
*
Offline Offline

Posts: 2307


« Reply #15 on: December 24, 2014, 03:52:03 PM »

Why the crazy high voltages?
Logged
kc2we
Member

Offline Offline

Posts: 138


« Reply #16 on: December 24, 2014, 05:31:21 PM »

Typically the BIL is 110 KV on 12.47 KV systems to allow for impulse (line surges) and lightning excursions. Also, when available fault currents are in the order of 4000 to 6000 amperes, lots of destructive forces are at work that can cause destructive impulses. Faults can be caused by a downed wire after a tree falling or a car-pole accident. Sub station circuit breakers can take as long as (3) cycles @ 60 Hz (48 milliseconds or 0.048 seconds) to trip all three phases during a line fault. We use MOV arresters that fire at around 10 KV to ground to limit the voltage levels, but even they blow up at times during a nearby lightning strike event when the ground voltage gradient rises significantly for a few micro-seconds.

Seth KC2WE
Logged

Seth Taylor
N2DTS
Contributing
Member
*
Offline Offline

Posts: 2307


« Reply #17 on: December 24, 2014, 05:59:51 PM »

I was thinking about the amplifier.
I can see why crazy high voltages happen in power transmission.

But running 4000 or more volts at high power (currents) at home would not be something I would want to do.


Logged
kc2we
Member

Offline Offline

Posts: 138


« Reply #18 on: December 24, 2014, 09:07:21 PM »

I guess my point about the voltage ratings of caps is that there are three values. One is the terminal to terminal working voltage, the terminal to terminal peak voltage and the third and not often stated BIL. The BIL is usually much higher than would ever be encountered in home construction. Seth KC2WE
Logged

Seth Taylor
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #19 on: December 24, 2014, 11:49:51 PM »

The 4-1000 in grounded grid performs best at 5KV. There is an eimac sheet on this.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #20 on: December 25, 2014, 01:15:00 AM »

The 4-1000 in grounded grid performs best at 5KV. There is an eimac sheet on this.

If you're running the tube at 5KV, why the need for FWB with the xfmr you're using?  And the need for the 12KV cap??

Fred

OK, I re-read some of your comments.  Using a FWB, you get about 8KV and only half that with a FWct rectifier which is a KV less than the 5KV you need.

You can bring up the voltage another 1KV if you have another xfmr that can get you the extra 1000 volts you need.  Using another xfmr with its own set of rectifiers, connect the rectifier output of this xfmr to the CT of the first xfmr and get the 5KV you need.  This second xfmr can use either FWB or FWct rectifiers, doesn't matter.  They can be SS rectifiers.  Just connect the rectifier output directly to CT of the first xfmr.  You can step-start both xfmrs with one step start relay.

Also, the VA load is shared by both xfmrs.  This method only brings up the CT voltage by the 1000 volts as compared to the 4KV if you use a FWB on the main xfmr.  I've used this method many times and it works FB.  If you want to lower the voltage you can just shut off the boost xfmr without disconnecting anything, just shut it off.  You could also use a variac on the boost xfmr for adjustment.

With 5KV total you could probably just start the boost xfmr without any step-start and step-start the main xfmr.

One last thing,  the current rating for the boost xfmr has to the same as the main xfmr as all the current flows through both xfmrs equally.

A side note,  this method of running xfmrs in series also works if the main xfmr is using a FWB rectifier.  Instead of connecting the boost xfmr rectifier output to the CT of the main xfmr, connect it directly to the negative terminal of the FWB rectifiers of the main xfmr.  Everything else is the same.

The secondary winding of the main xfmr has to be able to withstand the higher voltage put on it by the boost xfmr.  The boost voltage adds to the winding voltage of the main xfmr.
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #21 on: December 25, 2014, 05:33:11 PM »

FWCT under load and LC filter shows about 2.5KV at 600-700mA and not enough to get 1500W PEP.
FWB would be more like 5KV under load, 8K with no load. I'll use  bleeder of some kind to lower it.
C input on FWCT is not a good idea because the transformer has a higher impedance than desirable.
I've done the whole thing stacking supplies, it works. I added 1000V @700mA to a transmitter, but much more than that and I would be concerned about shorting the CT to GND on other transformers.
The transformer is on the small side but I am going to try it.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
KA2DZT
Member

Offline Offline

Posts: 2192


« Reply #22 on: December 26, 2014, 04:06:00 AM »

Patrick,

OK FB,  I better see what you have.  I thought the 8KV was with the heavier load.  Seems, based on the 2.5KV with FWct and a heavy load, you would get the 5KV you need with a FWB.

Good luck with the project.

Fred
Logged
w4bfs
W4 Beans For Supper
Contributing
Member
*
Offline Offline

Posts: 1433


more inpoot often yields more outpoot


« Reply #23 on: December 26, 2014, 08:52:49 AM »

It is risky to assume that even with an insulator as large as the winding ends (hv) that the ct will have the same voltage rating to frame as the ends .... a hipot tester with small current detection (1 uA) dc will allow this to be done non-destructively and can settle the issue .... 
Logged

Beefus

O would some power the gift give us
to see ourselves as others see us.
It would from many blunders free us.         Robert Burns
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #24 on: December 26, 2014, 12:35:08 PM »

I may be able to do that hi-pot. There are schematics posted here for it.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
Pages: [1]   Go Up
  Print  
 
Jump to:  

AMfone - Dedicated to Amplitude Modulation on the Amateur Radio Bands
 AMfone © 2001-2015
Powered by SMF 1.1.21 | SMF © 2015, Simple Machines
Page created in 0.077 seconds with 19 queries.