The AM Forum
March 28, 2024, 07:31:33 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 [2] 3   Go Down
  Print  
Author Topic: Homebrew a plate choke for 4-1000 GG with warc  (Read 17910 times)
0 Members and 1 Guest are viewing this topic.
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #25 on: October 27, 2020, 03:20:45 PM »

He should have used one of the big plunger HV contactors.

Plate choke design is an art itself.  I've only begun to understand that statement after building amps and then measuring the chokes in place with the VNA.  Never had a GDO, so this was just one of those "I always used what worked" scenarios.


--Shane
KD6VXI
Logged
VE7RF
Member

Offline Offline

Posts: 56


« Reply #26 on: October 28, 2020, 11:52:03 AM »

He should have used one of the big plunger HV contactors.

Plate choke design is an art itself.  I've only begun to understand that statement after building amps and then measuring the chokes in place with the VNA.  Never had a GDO, so this was just one of those "I always used what worked" scenarios.


--Shane
KD6VXI

I'm seriously  thinking of a  simple   HB  horizontal cam operated plunger setup.   Sorta like a spdt  mini knife  switch.  What else that just might work is a spare  single wafer  model  88  I have packed away.   They are rated for   13 kv  @  60 hz...but that would be marginal..and real  bulky...stray C  etc.

2 x  hb  delrin  plungers, one for each choke would work.   Banana plug on the end of each plunger.   One banana  jack for each choke.   Select  choke  #1...or  choke  #2.

Jim   VE7RF
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #27 on: October 28, 2020, 12:27:02 PM »

50 uh for testing at 5kv on 17M


* 16039024802281261658875730628839.jpg (3918.61 KB, 4128x3096 - viewed 468 times.)
Logged
VE7RF
Member

Offline Offline

Posts: 56


« Reply #28 on: October 28, 2020, 03:25:43 PM »

50 uh for testing at 5kv on 17M


On paper,  you should experience zero issues.  50 uh should have it's  1st  series  resonance at  40 mhz.
On  3.5 mhz,  50 uh  =    1100  ohms  xl.    5 kv  x .6  =  3 kv       3000/1100 ohms =  2.72  amps  flowing through  the choke..and also the  bypass cap at the base of the  choke.   The bypass cap(s)  should be something substantial to handle  that much current.   Like a pair of  4700 pf  @  10 kv  discs, or what other combo  that will handle 2.72 Amps  @  3.5 mhz.

Even a HEC   HT-57  in  200pf will handle  15A  > 5.2 mhz.....and  12 A  @  3.5 mhz.   And 9 A    at 1.8 mhz.

3.5 mhz  would be the  worse case.  As freq increases, bypass cap  current drops like a rock.... like  1.363 A on 40m...etc.

What gauge magnet wire did you use ?

The  C1  tune cap value will have to be increased  a bit  on  80m.   1100  ohms of  xl  requires   1100 ohms  of  xc  to cancel it out.  IE:  the 50 uh plate choke  will... 'eat'    42 pf from the C1  tune cap.

Jim   VE7RF
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #29 on: October 28, 2020, 03:33:21 PM »

Bypass caps are 200pf and 500pf at 7.5 kv...choke is 22 ga.wound on delrin?  Mounted 2 inch above chassis on porch insulator..haven,t tried it yet
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #30 on: November 02, 2020, 08:40:42 PM »

Haven't tested the 50 uh choke yet...got distracted by other stuff...the choke measures 51 uh on the wooden bench and 61 uh installed...Ill post the results...stay tuned....
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #31 on: November 28, 2020, 04:23:04 PM »

I got around to testing the 50uH plate choke......It measured 50.5 uH on the bench ...61uH installed..   It won't work on 75 meters.  Amp draws lots of plate current with just enough output to tune...  40 Meters is full output, normal operation...  Break for lunch..
Logged
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #32 on: November 28, 2020, 06:48:43 PM »


Been hammering out some CQ's on 14332/3, a  few takers, but only me on AM.

Reading through this thread, I have a really dumb question.

Why not put the choke on the 50 ohm side of the pi-network?

Sure, that eliminates the plate coupling capacitor on the input side, but that requires the B+ to be on the entire pi-net. A output series cap would be required, but here a quad of .005uf door knobs would do. The tuning cap shells could get an AC ground through a series capacitor to minimize DC arc damage through the capacitor plates. This is done at a low level on the Central Electronics 20A Mixer tuning capacitor, i.e a .005uf cap.

That 50uh choke would be fine on the output side, from 160-10m. 

Yea, I know, a dumb question.

Jim
Wd5JKO


Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #33 on: November 29, 2020, 10:57:40 AM »


Been hammering out some CQ's on 14332/3, a  few takers, but only me on AM.

Reading through this thread, I have a really dumb question.

Why not put the choke on the 50 ohm side of the pi-network?

Sure, that eliminates the plate coupling capacitor on the input side, but that requires the B+ to be on the entire pi-net. A output series cap would be required, but here a quad of .005uf door knobs would do. The tuning cap shells could get an AC ground through a series capacitor to minimize DC arc damage through the capacitor plates. This is done at a low level on the Central Electronics 20A Mixer tuning capacitor, i.e a .005uf cap.

That 50uh choke would be fine on the output side, from 160-10m. 

Yea, I know, a dumb question.

Jim
Wd5JKO



Hi Jim....I have been calling around 14340 mid day.....Your question is not really that dumb....It's making me think....I just don't know what to think.    That 50uH choke made the tune cap run out of C on 75 M...time to experiment some more..
Logged
W2JBL
Member

Offline Offline

Posts: 676


« Reply #34 on: December 08, 2020, 12:37:16 AM »

Working on a hombrew (not by me) GG 4-1000 amp here. The Plate Rf choke was a ham made clone of the AL-82. It fried on 160. Replaced it with another ham made choke that measured 90 microhenries. Works perfcetly on 160, 80, 40 and 20. Six inches of #24 on a 3/4" form close wound.
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #35 on: December 10, 2020, 11:28:09 PM »

Aside from high voltage hazards and the rest, isn't there a choke advantage to leaving the pi coil hot, putting the DC blocker at the cold end, and getting by with a much smaller 'plate' choke to feed HV through the pi coil via the 'cold' end? TMC used to do that. I want to try it.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
VE7RF
Member

Offline Offline

Posts: 56


« Reply #36 on: December 11, 2020, 02:24:49 AM »

Aside from high voltage hazards and the rest, isn't there a choke advantage to leaving the pi coil hot, putting the DC blocker at the cold end, and getting by with a much smaller 'plate' choke to feed HV through the pi coil via the 'cold' end? TMC used to do that. I want to try it.

A  couple of ways to do this..and one is to feed  HV wire through cu tubing,  BEFORE it's  wound into a tank coil.   Blocking cap  still at anode end....and  exiting HV wire at anode end of tubing is  routed  around block cap to  anode.  That only works on tubing coils.   

If hv wire can't be  routed through tubing, like  edge wound coils, airdux  etc,  and  50 uh choke down at 50 ohm end, the  tune + load  caps + tank coil,  Plus band switch  are all hot with  B+.    Jennings  et  all sez u gotta  severely de-rate vac caps if  B+ is on them...and that's on top of the  existing de-rating  fro  RF.   IE:  15 kv rated cap is the hi -pot rating.  RF rating is  60% of that or  9 kv.  With  B+  present,  it's reduced  down to  5 kv.

Good news is....  W8JI  has come up with a 'big boy'  version of his ameritron choke.   It's  designed for these bigger tubes, and higher voltages.   It's  1.3"  diam  x 10"  long.   Rated  for 10 kv loaded  @ 5-6 amps  ccs.    400 uh.....and does not resonate on  any of the 9 ham bands  from  160-10m.  Does not resonate on 60m +  11m  either.   It's gapped in 2 places.   Took him a week, and several lbs of wire to get it  right.   He just got  50 more (unglazed) ceramic forms delivered a couple of weeks  ago.  He uses a semi lathe device, that also counts the turns....so he knows when to stop, and insert a precise length gap.    Tapped for  1/4-20  at each end.  Transverse brass screw  at each (inboard) end of  choke top/bottom winding, to terminate windings.  It can be mounted to the  chassis..and stoof vertically.   It can  also be mounted parallel to a side or rear wall....as long as it's at least  2" away.

Dunno when they will become available, but they will be a godsend for us  HB folks.  400 uh =  4523 ohms of  XL   @  1.8 mhz.    He sez the RMS  RF  voltage  component is 60% of the loaded  B+ value.   IE: 5 kv loaded B+  =  3 kv  rms.   3000 / 4523 ohms = .663 amps  of  RF flowing down the choke...which requires a big value bypass cap at base of plate choke...like a pair of  4700 pf  @  10/15 kv  disc ceramics.

IF the  bypass cap at the cold end of plate choke is not big enough, like  say a 1000 pf  doorknob used,  ( even if the  1000 pf  doorknob will handle lotsa current on 160m) what happens is.... the  too small a value bypass cap will  only pass  1/3  to  1/2 of the  RF current  coming down the choke...and the rest  goes straight into the  HV filter caps.   

Most of us have tried every  version of a 2 x piece  (big and small) separate  plate choke schem, that's involves a relay.    Problem is, both chokes are used on 160m only..and the smaller  50 uh  choke still has to be used on  80-10m.   Both chokes will not work on  80+40m.... the peak V  at the junction of the
 2 x chokes is  sky high.

Hopefully he will get them available after xmas.   I need  2 of em  real bad.... one for the  3x3 amp...and the other for the  3x6 amp.

Later.....  Jim   VE7RF
Logged
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #37 on: December 12, 2020, 10:06:17 PM »

A  couple of ways to do this..and one is to feed  HV wire through cu tubing,  BEFORE it's  wound into a tank coil.   Blocking cap  still at anode end....and  exiting HV wire at anode end of tubing is  routed  around block cap to  anode.  That only works on tubing coils.   

   I like that idea about using tubing. For a multi-band rig, maybe a hybrid approach would work. Maybe use a single piece of copper tubing for 10m-15m-12m-17m, and then switch over to air-wound for 20m-40m-80m, and then possible a toroid for 160m. The idea is to thread the HV B+ wire through the copper coil, and from there use a regular plate choke to the B+ that only needs to deal with 160m to 20m. Any parasitic resonances of the plate choke between 14.5Mhz to 30Mhz would not matter as much because the circuit impedance would be 50 ohms at that point when operating on the upper bands.

Would that work?

Jim
Wd5JKO


Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #38 on: December 13, 2020, 09:39:19 AM »

Aside from high voltage hazards and the rest, isn't there a choke advantage to leaving the pi coil hot, putting the DC blocker at the cold end, and getting by with a much smaller 'plate' choke to feed HV through the pi coil via the 'cold' end? TMC used to do that. I want to try it.
I have a TMC transmitter with a coaxial plate choke..It tunes from 1.5 to 30 Mhz with no plate choke problems..
I have also ben playing with different chokes on the 4-1000A amp....55 uH worked except for 75 meters..I added another 22 uH and the choke worked fair on all bands...I wound a straight 300uH on a glazed ceramic 1 inch form and it works the best so far..full output on all bands except 10 .  I cant put more than 5 KV on it however because the bandswitch arcs and I need some bigger doorknobs for bypass...more later
Logged
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #39 on: December 15, 2020, 11:39:01 PM »

Sounds pretty good. My TMC pi coil is tubing designed to be @7500VDC but I will run no more than 4800.
Impractical to get any sort of HV wire through.

On the caps discussed, how are these older micas for HF with a 3CX3000 at 6-7KW? I'm not ashamed to ask. LOL was never ashamed of driving the speed limit in a fast car either.


* side-w-specs-small.jpg (121.96 KB, 750x604 - viewed 213 times.)
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
WD5JKO
Member

Offline Offline

Posts: 1996


WD5JKO


« Reply #40 on: December 16, 2020, 09:20:19 AM »

LOL was never ashamed of driving the speed limit in a fast car either.

Pat, you still have that /6 Dodge Dart in the driveway?  Tongue

Jim
Wd5JKO
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #41 on: December 16, 2020, 11:21:20 AM »

Sounds pretty good. My TMC pi coil is tubing designed to be @7500VDC but I will run no more than 4800.
Impractical to get any sort of HV wire through.

On the caps discussed, how are these older micas for HF with a 3CX3000 at 6-7KW? I'm not ashamed to ask. LOL was never ashamed of driving the speed limit in a fast car either.


I've seen them used, Pat.  But the TX I saw them in was for the lower portion of HV.  Guy had a 75/80 meter amp with a 4CX10k in them.  Dude had dedicated tens on 160 - 40!  Guess he wanted to have the world hear his CQs.


--Shane
KD6VXI
Logged
K1JJ
Contributing
Member
*
Offline Offline

Posts: 8893


"Let's go kayaking, Tommy!" - Yaz


« Reply #42 on: December 16, 2020, 12:29:58 PM »

Pat,

I have used those older mica caps and still do (QRO) for years up to 20M and never had one pop.  I was told that they become somewhat inductive above 40M, however.  (How wud we test this)  I have used them in PDM filters (100 KHz) and QRO on 20M.  I have faith in them from 160 to 40M.

They are very common from BC transmitters over the years.

As for 10, 15 meters, (and possibly 20M) I would go with the big Russian caps to be safe.

T
Logged

Use an "AM Courtesy Filter" to limit transmit audio bandwidth  +-4.5 KHz, +-6.0 KHz or +-8.0 KHz when needed.  Easily done in DSP.

Wise Words : "I'm as old as I've ever been... and I'm as young as I'll ever be."

There's nothing like an old dog.
Opcom
Patrick J. / KD5OEI
Contributing
Member
*
Offline Offline

Posts: 8308



WWW
« Reply #43 on: December 17, 2020, 11:24:23 PM »

LOL was never ashamed of driving the speed limit in a fast car either.

Pat, you still have that /6 Dodge Dart in the driveway?  Tongue

Jim
Wd5JKO

Yes I do still have that Dart. I have no time to mess with it because of other stuff and been thinking of letting it go.

-==

On the cap, well, you guys say it's iffy, so I better listen about the freq limitations.

My LCR measurement stuff  -specific purpose instruments -are far short of HF measurements. Maybe the MFJ259 would give some interesting results -if the cap is resonant at some frequency it could be calculated I am postulating.

Also looking at the cap's nameplate -the current rating decreases after a point. Well they make good parts for other gear and otherwise always nice paperweights. For me the offer an additional advantage as a standoff so I was hoping it might work out.
Logged

Radio Candelstein - Flagship Station of the NRK Radio Network.
VE7RF
Member

Offline Offline

Posts: 56


« Reply #44 on: December 26, 2020, 04:16:44 PM »

Sounds pretty good. My TMC pi coil is tubing designed to be @7500VDC but I will run no more than 4800.
Impractical to get any sort of HV wire through.

On the caps discussed, how are these older micas for HF with a 3CX3000 at 6-7KW? I'm not ashamed to ask. LOL was never ashamed of driving the speed limit in a fast car either.


W8JI uses one on his  3CX-15,000A7  GG  AM SW broadcast amps.(3-22 mhz), 6 kw cxr, 24+ kw pep out.   Look at  CDE's  site.   Although touted as  100 khz  to 3.3 mhz   caps, the  detailed graphs   show another story,  some will   peak for max current at 10 mhz..then drop in current rating  above / below  10 mhz.... and still handle bucket loads of current   up to 22 mhz.    Some will handle a lot oc  current up to lower vhf. 

Folks were saying the mica caps don't like  B+.     CDE engineers  told me  3 months ago, there is  zero issues  running B+ on em, and using em as plate blockers. ...even up to almost rated voltage.   There is no...'chemical migration'  with  B+  applied. 

But u have to use the right cap..  rated to handle the caculated rf current through the cap.... which can be substantial on upper  HF  +  6m.   Very easy to calculate the  rf current through the block cap...on any freq.

Myself,  I use 4 x  200 pf  @  15 kv  HEC-HT-57's for plate blocking on 80-10m.   The  200 pf HT-57 is the..'go to' cap..rated for  15A..each.   I will use  8 of em on a 160-10m amp.   On my own  160-15m  3x3 amp, I used  6 x  500 pf  HT-57's.

Jim   VE7RF
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #45 on: December 26, 2020, 07:08:47 PM »

In the past I would use some monster Draloric caps.

Now, with the commie stuff so cheap, caps from China are my go to for 6000 and lower.  They bolt in with M8 terminals and I've passed some pretty big peak and avg current through them at 29 mhz.

Anything bigger I go vac cap from Russia.

--Shane
KD6VXI
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #46 on: January 05, 2021, 05:59:10 PM »

Aside from high voltage hazards and the rest, isn't there a choke advantage to leaving the pi coil hot, putting the DC blocker at the cold end, and getting by with a much smaller 'plate' choke to feed HV through the pi coil via the 'cold' end? TMC used to do that. I want to try it.

A  couple of ways to do this..and one is to feed  HV wire through cu tubing,  BEFORE it's  wound into a tank coil.   Blocking cap  still at anode end....and  exiting HV wire at anode end of tubing is  routed  around block cap to  anode.  That only works on tubing coils.   

If hv wire can't be  routed through tubing, like  edge wound coils, airdux  etc,  and  50 uh choke down at 50 ohm end, the  tune + load  caps + tank coil,  Plus band switch  are all hot with  B+.    Jennings  et  all sez u gotta  severely de-rate vac caps if  B+ is on them...and that's on top of the  existing de-rating  fro  RF.   IE:  15 kv rated cap is the hi -pot rating.  RF rating is  60% of that or  9 kv.  With  B+  present,  it's reduced  down to  5 kv.

Good news is....  W8JI  has come up with a 'big boy'  version of his ameritron choke.   It's  designed for these bigger tubes, and higher voltages.   It's  1.3"  diam  x 10"  long.   Rated  for 10 kv loaded  @ 5-6 amps  ccs.    400 uh.....and does not resonate on  any of the 9 ham bands  from  160-10m.  Does not resonate on 60m +  11m  either.   It's gapped in 2 places.   Took him a week, and several lbs of wire to get it  right.   He just got  50 more (unglazed) ceramic forms delivered a couple of weeks  ago.  He uses a semi lathe device, that also counts the turns....so he knows when to stop, and insert a precise length gap.    Tapped for  1/4-20  at each end.  Transverse brass screw  at each (inboard) end of  choke top/bottom winding, to terminate windings.  It can be mounted to the  chassis..and stoof vertically.   It can  also be mounted parallel to a side or rear wall....as long as it's at least  2" away.

Dunno when they will become available, but they will be a godsend for us  HB folks.  400 uh =  4523 ohms of  XL   @  1.8 mhz.    He sez the RMS  RF  voltage  component is 60% of the loaded  B+ value.   IE: 5 kv loaded B+  =  3 kv  rms.   3000 / 4523 ohms = .663 amps  of  RF flowing down the choke...which requires a big value bypass cap at base of plate choke...like a pair of  4700 pf  @  10/15 kv  disc ceramics.

IF the  bypass cap at the cold end of plate choke is not big enough, like  say a 1000 pf  doorknob used,  ( even if the  1000 pf  doorknob will handle lotsa current on 160m) what happens is.... the  too small a value bypass cap will  only pass  1/3  to  1/2 of the  RF current  coming down the choke...and the rest  goes straight into the  HV filter caps.   

Most of us have tried every  version of a 2 x piece  (big and small) separate  plate choke schem, that's involves a relay.    Problem is, both chokes are used on 160m only..and the smaller  50 uh  choke still has to be used on  80-10m.   Both chokes will not work on  80+40m.... the peak V  at the junction of the
 2 x chokes is  sky high.

Hopefully he will get them available after xmas.   I need  2 of em  real bad.... one for the  3x3 amp...and the other for the  3x6 amp.

Later.....  Jim   VE7RF
Have you been able to get one of hhe big plate chokes yet?
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #47 on: January 07, 2021, 10:50:20 AM »

Check out this 90 degree plate choke for 10 thru 160 meters...very clever


* 90 deg plate choke.jpg (170.8 KB, 1152x864 - viewed 256 times.)
Logged
KD6VXI
Contributing
Member
*
Offline Offline

Posts: 2648


Making AM GREAT Again!


« Reply #48 on: January 18, 2021, 04:46:38 PM »

https://eaglemade.net/collections/chokes/products/plate-choke-10-thru-160-meters?fbclid=IwAR3lpXM1vWkwtW_m4xqXfNiAvd38zAzDt--AYL0xbOHmID--z8kSdbbWJbA

10-160

As mentioned in another thread, Floyd makes some good stuff.  I don't know where the resonances are, but you could probably email him and find out.

--Shane
KD6VXI
Logged
KL7OF
Member

Offline Offline

Posts: 2313



« Reply #49 on: January 24, 2021, 09:32:36 AM »

I finalized the design on the 4-1000 GG. Running 6KV at present. Wound up with a 220Uh plate choke wound with #18 formavar on a ceramic core...bypassed with (2) 3300pF  @12KV caps.  The bandswitch problem was solved by switching the coil taps with Russian Vac caps..so simple and works very well.  The amp has great output on 15, 17, 20 40, and 75.  I ran this amp with only a 10 meter coil and was able to get 900 watts on 29mhz with only a little heating of the plate parasitic suppressor.  With a multiband tank installed, 10M performance is dismal..So no 10M band on this amp...


* 20210121_095958.jpg (1975.23 KB, 2160x2160 - viewed 267 times.)

* 20210121_100032.jpg (1813.97 KB, 2160x2160 - viewed 260 times.)
Logged
Pages: 1 [2] 3   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 0.097 seconds with 19 queries.