The AM Forum
March 28, 2024, 06:14:00 PM *
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: RF capacitor substitution  (Read 7561 times)
0 Members and 1 Guest are viewing this topic.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« on: April 30, 2018, 08:31:21 PM »

If the capacitors C930 and C911 are originally fixed vacuum, is there any reason that suitably rated high current high voltage mica caps can not be used?


* New Bitmap Image.png (120.42 KB, 738x559 - viewed 407 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
DMOD
AC0OB - A Place where Thermionic Emitters Rule!
Contributing
Member
*
Offline Offline

Posts: 1768


« Reply #1 on: May 01, 2018, 07:50:16 AM »

If the capacitors C930 and C911 are originally fixed vacuum, is there any reason that suitably rated high current high voltage mica caps can not be used?

No problem.

I have replaced vacuum types with Broadcast Micas.


Phil - AC0OB
 
Logged

Charlie Eppes: Dad would be so happy if we married a doctor.
Don Eppes: Yeah, well, Dad would be happy if I married someone with a pulse.NUMB3RS   Smiley
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #2 on: May 02, 2018, 09:20:18 PM »

That is great news. I have not worked with this high of power levels so, better to ask rather than risk a fire. Vacuums cost much more as well.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
WA1QHQ
Member

Offline Offline

Posts: 111



« Reply #3 on: May 03, 2018, 07:30:59 AM »

Rather unconventional way of doing a Pi network, floating the tank circuit at the PA B+ potential, possibly dangerous but also puts a lot of circulating tank current thru C930 and C911 that would not be the case if the tank was not floated at the PA B+ and C911 was moved to the high impedance side of the tank circuit. I'm not sure why it was done this way.

Mark WA1QHQ
Logged
W1ITT
Member

Offline Offline

Posts: 573


« Reply #4 on: May 03, 2018, 08:24:23 AM »

The reason for "floating" the output network, and running B+ through the inductor revolves around plate RF choke design and performance.  In the conventional situation, we connect the choke at the plate end of the inductor, where the impedance is high, perhaps 2500 ohms or so, and the B+ is blocked by the plate blocking capacitor which keeps it out of the inductor and out into the antenna system.
In this TMC method, used by others as well, the plate choke is connected at the low impedance end of the network, where the Z is around 50 ohms.  It's much easier to design and operate a choke that will work properly, without series resonances, yet with enough inductance to keep RF out of the HV supply.  This requires running HV through the inductor, which requires some attention to detail, often in the way of a better switch changing the taps of the coil for different bands.  The plate DC blocking capacitor just moves down the road a bit.  And we don't stick out hands into the plate circuit when things are running anyway.  The added complication is that both ends of the plate tuning and loading capacitors have to be above ground, which requires insulated mounting and shaft couplings.  But if we take all that trouble, we often end up with a more stable amplifier without no-go frequencies.  Note that many common amateur amplifiers are not recommended for use in the 12 meter band, because of a resonance in the plate choke designs at that frequency.  This method used by TMC makes that problem go away. 
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #5 on: May 03, 2018, 11:01:29 AM »

To expand on W1ITT comments.

It's a LOT easier to design a choke with 10 to 20 times Xl at 50 ohms than at 1200 ohms!

With less wire, less chance for resonance.

--Shane
KD6VXI
Logged
w4bfs
W4 Beans For Supper
Contributing
Member
*
Offline Offline

Posts: 1433


more inpoot often yields more outpoot


« Reply #6 on: May 03, 2018, 12:37:08 PM »

bravo guys .... well stated
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
WA1QHQ
Member

Offline Offline

Posts: 111



« Reply #7 on: May 04, 2018, 11:27:35 AM »

OOOPs missed that the inductor was connected to the low impedance end, of course you guys are right, much easier to do the RFC on the low impedance end, same thing was done on the T-368. This is a common practice on tube amps that have to have continuous coverage through the entire HF range.

WA1QHQ
Logged
W1ITT
Member

Offline Offline

Posts: 573


« Reply #8 on: May 04, 2018, 12:13:20 PM »

I believe that the concept was popularized by Pappenfus, Bruene and Shoenike of Collins Radio in their SSB book that came out in the Sixties.  I'm not sure if it was a Collins invention, or just a description of work that had been done. But feeding HV up through the inductor in a link-coupled parallel tuned plate circuit in the old days did essentially the same trick.    By the way, if you look at the RF choke that is commonly placed from the output connector to ground in a Pi-net for HV safety, you'll notice that it's ordinarily a garden variety pi-wound choke, nothing critical, and rarely a problem unless it gets blown to smithereens by errant HV breaking down the plate blocking capacitor.  RF chokes at low impedance are relatively easy.  (Relatively)
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #9 on: May 04, 2018, 11:15:25 PM »

Well that circuit is what I hope to do, but with a 3CX3000 instead, the usual I suppose "CCS AM" amp. I thought a lot about a ham bands only vs continuous tuning circuit and the cost was much less with the GPT-10 pi coil that already includes an otherwise very expensive bandswitch.

I don't have those expensive decoupling/blocking caps yet. Just about all I'm missing for major parts.
For high safety on tuning/loading mistakes on HV up to 4800V I would like to find 20KV units, but 15KV should be enough.
Logged

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

Posts: 8308



WWW
« Reply #10 on: May 05, 2018, 04:03:29 PM »

Is there any reason a higher capacitance unit would not drop in?
Is there any reason both of them would have to be the same capacitance?
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
PA0NVD
Member

Offline Offline

Posts: 615


Nico and Chappie (Chappie is the dog...)


« Reply #11 on: May 05, 2018, 11:30:59 PM »

We used normally ceramic disc caps for DC blocking and decoupling upto 50 kWatts without any problems. Were much more readily available and low priced But preferably a higher value than 1000 pF to prevent RF voltage across the cap.
For 10 kW at 27 MHz ceramic disc caps in the loading also performed without any problems
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #12 on: May 08, 2018, 02:52:44 AM »

That's interesting - been looking on a few alibaba /ali express like sites - there are some big flat plate-like ceramics as you describe.

Many don't have a stated current rating, and some say it is 60 or 120KVA and 20 up to 40KVDC. I'm not sure what to think of the KVA rating.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
PA0NVD
Member

Offline Offline

Posts: 615


Nico and Chappie (Chappie is the dog...)


« Reply #13 on: May 08, 2018, 10:24:14 AM »

In order to find the max current, divide KVA rating by voltage rating.  Than you have all info at the frequency you use it
For reliability reasons, I always stayed a factor two away from that limit, especially for the Chinese caps. So when used for decoupling or RF bypass, a bigger cap means less RF voltage so more margin.
We used them in 27 MHz RF generators for plastic welding, I even used them in the PI filter output side of a 10 kW generator with a 3CX6000A7 in GG amp circuit and as a decoupling is a balanced 25 kW generator for a dehydration system
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #14 on: May 08, 2018, 01:20:38 PM »

I've used the 25kva chicom caps as plate block caps on the high impedance side of a 3cx3000 running 3amps at 6.5 kv before.

However, keep in mind.....  Your plate current demands are a LOT higher on the low impedance side for CBlock.

Maybe a couple chicom caps in parallel.  At the price you find them in ebay, use a quad and be a real gangster.

--Shane
KD6VXI
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #15 on: May 20, 2018, 12:32:43 AM »

How about this type?

The ones I found (free) are Plessey Type 06, 1250pF +/-20%, 10KVDC DC working, and except for being gray in body color, look like the image.

Anyone got an old Plessey capacitor catalog? I don't think they make these any more.


* ec_rf-tubular-shell-capacitor.key-visual.jpg (20.84 KB, 700x370 - viewed 258 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
PA0NVD
Member

Offline Offline

Posts: 615


Nico and Chappie (Chappie is the dog...)


« Reply #16 on: May 20, 2018, 10:27:29 AM »

These come in various formats. With a lug at the ground side as in the pic, or with a plate around the cap for panel feed-through of HV. Than the center connection is at both sides. We used them as a standard in panel feed through for 27 MHz generators up to 50 kW for leading the grid or the HV into the RF compartment (the flange type cap)
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #17 on: May 21, 2018, 09:39:48 PM »

They may make good RF bypass caps then. Cool.
I still prefer to find the Micas if I can for those first mentioned uses.
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 0.1 seconds with 19 queries.