Warning: Creating default object from empty value in /homepages/11/d132647312/htdocs/Amfone/mkportal/include/SMF/smf_out.php on line 47
Battling a Collins 30K-1




 
The AM Forum
April 13, 2021, 06:31:31 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: Battling a Collins 30K-1  (Read 1096 times)
0 Members and 1 Guest are viewing this topic.
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4614



« on: March 13, 2021, 11:04:35 AM »

After working on KA1KAQ's 30K-5 I fell in love with 30K's the styling and bulletproof engineering, so when one came up for auction in Harrisburg PA I jumped on it.  Seemed to be in good physical shape but I noted in the auction pictures that someone had Roto-Rootered the final amp deck to convert it from balanced output to a pi-network.  I hoped for the best, won the auction, and dragged the beast home.

Below is the sequence of problems I found, and how I overcame them.

1:  Mod transformer disconnected, and sitting on the floor of the transmitter instead of on the mod deck. I hi-potted the power tranny and mod tranny and all looked good up to about 3500V, the limit of my Biddle tester.   I traced the wires out and connected according to the manual.  I also asked on the Collins forum if this was factory, since the mod tranny seemed to be original and there was no trace of it ever having been mounted to the mod deck.  I did receive a few pictures of other 30Ks having this configuration, and the wiring harness seemed original, so I proceeded to connect the 310A-3 exciter....

2:  Checked voltages at the input of the exciter.  Hmm, no negative bias.  Removed the PS deck and found a 1K wirewound resistor open.  Replaced that, and noted the proper grid bias.  But....

3:  No grid drive.  I noted the key jack wires for the exciter, which connect to two terminals on the back, were hanging in the air but ignored them.  That was a mistake, as the exciter needs a ground return thru the key jack to key.  Connected them up and got grid drive (but not the 12-15mA grid current needed).

4:  Not enough grid drive:  Exciter was mis-aligned.  In the process, of course two trimmer capacitors broke their collars and needed to be replaced.  There is also a mechanical spring-loaded linkage between the buffer plate tank variable capacitor and the rest of the mechanical tuning.  There are three variable caps that need to track as the exciter is tuned in frequency and the linkage was binding throwing everything out of sync.  A bit of cleaning and mechanical alignment correct that issue.  Alignment improved things, but due to the changes in the grid circuit on the amp deck (see #5) I still couldn't get over 12 mA of grid drive on 80.  I checked the tubes, voltages, etc and everything was spot-on.  I noted that there was a bank of series resistors between ground and the ground return on the exciter output coils.  I shorted across one of them and grid drive was now adequate.

5:  No plate current and HV (someone had mounted a non-stock, but useful, HV meter dead center on the front) was about 500V too high.   I pulled the final amp deck and noted a true mess of clipped capacitors and parasitic suppressor resistors on the screen and grid.  One of the screen terminals had one leg of a sad-looking (and shorted) disc capacitor soldered to it.  All of the grid circuit coils had been removed and a small, all-in-one coil/bandswitch assembly had been installed instead; it worked well enough so I left it be.  I cleaned it up, wired the screens directly to the bypass cap, and noted the 5K/50watt screen dropping resistor was open.  I replaced that and got plate current and some semblance of output except on 80 meters.  While I was testing all of this out, I reset the power transformer connections for 2KV vice the 2.5KV.

6:   No 80 meter output:  The bandswitch, which had been replaced with a big multi-deck ceramic job, had contacts for only 4 positions but it would rotate through 5 positions, repeating the fourth position.  So the "80 meter" position was actually wired to the 10 meter tap on the tank coil.  Removed that connection and got some output on 80.

7:  Loading cap all the way meshed on 80, could not reduce plate current below 150 mA with the loading control.  The loading capacitor, albeit 4-sections, was only about 800 pF fully meshed.  Using an unused terminal on the band switch, I connected a 500 pF doorknob cap to ground through that contact.  However, still not enough loading.

8:  Not enough inductance on the plate choke:  Whoever modified the deck apparently wanted to use it on the higher bands, and the plate choke was only 150 uH.  The plate resistance of the single 4-125 calculated to be about 8.6K, so 150 uH was not nearly enough at 40 and 80.  I found another 150 uH choke and connected it in series with the existing choke, but I knew that wasn't enough.  While I waited for a National 2 mH high voltage choke to arrive, I turned my attention to the tank circuit.

9:  Not enough inductance in the plate tank on 80.  The plate coil only had 16 uH or so and needed about 20 uH, according to some online pi network calculators.  I made a 4 turn coil but that only added 2 uH or so.  I robbed a Airdux tank coil out of an old commercial slopbucket carcass I had lying around and things were markedly better.  Now I could load to under 150 mA, with about 200W output (should be 250 at 150 mA) which I attributed to the plate choke still not having enough inductance.  Now, time to test the modulator.

10:  Not enough idle current on the mod tubes.  Someone had replaced the 75TH tubes with 100THs.  A valid mod, but the 100THs require much less bias voltage.  This problem was a combination of a bad bias pot and the 2KV plate voltage.  Cleaning and exercising the bias pot got some idle current on the 100THs.

11:  Audio oscillation and final plate current decrease when turning up the audio gain.  I puzzled over this one until I remembered that the mod tranny had a screen winding.  Although I had carefully traced out the connections and had connected the transformer according to the schematic, reversing the wires on the screen winding solved both of these problems.

12:  The new plate choke arrived and I installed it.  The 30K now made a good 250W at 2KV, and I figured once I reset the HV tranny taps to 2500V the mod idle current would be about right.  However, I noted that the plate current would drift, often fairly suddenly.

13:  Loading cap bearing did not have enough drag.  The drifting plate current was caused by the weight of the loading cap plates moving the loading cap position as the cap did not have enough drag.  A lot of messing with the bearing was fruitless; solution was to put a plastic bushing between the shaft coupler and the inner panel, and secure the shaft coupler as tightly against the bushing as I could to increase the drag.  Now the loading cap stayed put.

14:  Arcing, sparking, and fire.  I moved the transformer taps to 2500V and things started arcing.  I found that the parallel 500 pF/20KV doorknob blocking capacitors on the plate circuit both were breaking down.  Replaced them with a pair of new blocking capacitors.  Then, the National plate choke started arcing at the top and caught fire.  I installed the old 150 uH choke and IT started arcing.  I figured there were probably parasitics and possibly a neutralization problem.

15:  Neutralization. The Eimac 4-125 data claims no neutralization was generally needed, but a wire in close proximity to the plate and conencted to the cold end of the grid circuit would suffice if necessary.  However, I hooked a clip lead to the cold end of the grid circuit using it as a makeshift neutralizing cap, and feeding drive but no B+, there was an impressive amount of feedthrough as observed on the spectrum analyzer.  I pulled the RF deck again and found that the grid circuit bypass was .05, which obviously swamped any negative feedback from the JS-ed neutralzing cap.  I changed this cap to .001 uF, and fed a nice stiff wire up through a chassis hole, connected to the cold end of the grid circuit and placed close to the envelope of the 4-125.  The feedthrough was reduced by about 20 dB, indicating a large improvement.

16:  Cleaning up the plate circuit wiring.  While I had the deck out I fashioned a parasitic choke out of a 47-ohm resistor with a few turns of wire around it, and replaced the large, wide, silver straps in the plate circuit with smaller braid.  Finally I replaced the plate choke with a 1.6 mH choke out of an old broadcast transmitter.  Any/all of these fixes could have contributed to taming the amp, but all probably needed to be done.

I've tested the beast for a few hours and it seems to work just fine.  Audio is good and it modulates easily with a decent 50 mA of modulator idle current.  No burn marks or arcing on the plate choke, testing it on every band.  I still need to re-do the final tank so I can use 10 meters and delete 15 meters, since the switch only has four positions and I don't intend to use it on 15.  I'm a receiver guy and learned a lot about RF final stages during this saga.  Hope this will help someone out somewhere down the line.

Now I gotta build up a T/R relay box, wire it up for PTT, connect a receiver, and have some fun with it! 


Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
w4bfs
W4 Beans For Supper
Contributing
Member
*
Offline Offline

Posts: 1382


more inpoot often yields more outpoot


« Reply #1 on: March 13, 2021, 07:14:59 PM »

now that is a diagnostic journey !    some real nice work  Smiley
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
kb3ouk
Member

Offline Offline

Posts: 1621

The Voice of Fulton County


« Reply #2 on: March 14, 2021, 02:58:14 PM »

Replacing the silver strap with braid, I've always heard that being done the other way around? Does the braid vs strap really make a difference? My p-p 813 deck i got at a hamfest a few years ago has short inch long braid connections from the plates to the plate cap. Haven't finished that rig yet so don't know how it will play yet. Those 30Ks are a nice looking transmitter, i like the look of them a lot better compared to say a KW-1. Isn't the 30K-1 the amateur version and the 5 was the commercial version with the dual output tank circuits?
Logged

Clarke's Second Law: The only way of discovering the limits of the possible is by venturing a little past them into the impossible
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4614



« Reply #3 on: March 14, 2021, 06:23:59 PM »

The strap was about an inch wide and unnecessary for that power level.  Probably created a fair amount of parasitic capacitance and was unwieldy when removing the tube and re-arranging components like the plate choke.  Dunno whether it's better or worse, I had the braid (it's a bit smaller than the shield on RG-58).

Yep, the 30K-1 is the ham version.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
Steve - K4HX
Administrator
Member

Offline Offline

Posts: 2729



« Reply #4 on: March 14, 2021, 08:12:46 PM »

That's quite a message! Congratulations on your perseverance and success.
Logged
WBear2GCR
Member

Offline Offline

Posts: 4041


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


WWW
« Reply #5 on: March 15, 2021, 03:30:06 PM »

JN, pictures please?

Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4614



« Reply #6 on: March 15, 2021, 08:23:19 PM »

Here ya go Bear


* 414D6559-F495-435C-BADC-8BB6CD6E6A1A.jpeg (2491.82 KB, 4032x3024 - viewed 95 times.)

* 1759D2A9-05D6-466F-BB83-2989EAC960BF.jpeg (2361.82 KB, 4032x3024 - viewed 118 times.)
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
w4bfs
W4 Beans For Supper
Contributing
Member
*
Offline Offline

Posts: 1382


more inpoot often yields more outpoot


« Reply #7 on: March 15, 2021, 09:55:42 PM »

oh boy !   that rig doesn't have to live out its days in the basement or garage !  N I C E
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
Todd, KA1KAQ
Administrator
Member

Offline Offline

Posts: 4279


AMbassador


« Reply #8 on: March 16, 2021, 09:30:25 PM »

Congrats on driving out the gremlins, Johnny. You certainly had a lot of hacks to clean up! And in record time, too. What'd it take me - 27 years to get mine back on the air? And then only after you came to NC to help me with it.  And again in Manassas.

As much as I like the 'busy' look of the commercial models, those three large Daka-Ware knobs across the top of yours look badass. If you get a chance, please send me a photo of the tag on the back door. I'm guessing you have an early model as it doesn't have that small knob (coupling I think?) between the center & right knobs, unless it was removed and plugged when they re-did the outpoot network.

Really looking forward to working you on the air with it. And when we get the 30K site back up and running, we'll add you to the list. After being through two of them now - both amateur and commercial models - you get the 'offical' title of 30K Guru, with oak clusters for Gremlin Removal.  Grin
Logged

known as The Voice of Vermont in a previous life
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4614



« Reply #9 on: March 17, 2021, 05:37:37 AM »

Todd, your 30K-5 is much more complex than the K-1 since the osc/multiplier is all outboard in the 310A-3, and no dual-channel stuff.

The coupling control was removed and the hole filled.  It's S/N 136.

* 96627522_7_x.jpg (1050.26 KB - downloaded 31 times.)
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
Tom WA3KLR
Contributing
Member
*
Offline Offline

Posts: 2038



« Reply #10 on: March 17, 2021, 10:36:54 AM »

Hi John,

Great work! It's sad and amazing what Hammy Hambone can do to what was a nice piece of equipment to begin with.

HPE 2 WRK U SOON ON 75 OM.
Logged

73 de Tom WA3KLR  AMI # 77   Amplitude Modulation - a force Now and for the Future!
Todd, KA1KAQ
Administrator
Member

Offline Offline

Posts: 4279


AMbassador


« Reply #11 on: March 17, 2021, 05:06:04 PM »

Interesting to hear that your mod tranny is up on the mod deck. Mine is indeed on the bottom deck/floor of the cabinet along with the plate iron, though I think someone may have turned the plate tranny sideways when they reinstalled it.

The only iron on the mod deck looks to be for the 6B4G mod driver.

When I got mine, all of the balanced antenna insulators were missing and a SO-239 was kludged onto the back. Managed to score enough to return it to original, pretty sure I have some spares if you want them. Even if you don't hook them up, they look pretty cool and fill the holes nicely.

Hey, you got the hard part(s) done - now the fun of detailing as little or as much as you want is ahead.


* Inside_bottom.JPG (812.55 KB, 1330x2000 - viewed 75 times.)

* RF-out_4.JPG (502.23 KB, 997x1500 - viewed 65 times.)

* Top5.JPG (258.19 KB, 960x720 - viewed 62 times.)
Logged

known as The Voice of Vermont in a previous life
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4614



« Reply #12 on: March 17, 2021, 06:46:44 PM »

Todd, like yours, my mod tranny is sitting on the floor, but most of the pics I've seen of 30K-1s including the manual show it mounted on the mod deck.  A few guys on the Collins list sent me pics of their 30K-1s with the mod tranny on the bottom deck next to the plate tranny.  Who knows at what time during the length of production they did that, but it doesn't look like the mod tranny was ever mounted on the mod deck.

Yeah, all my insulators have been removed.  I did use two of the holes for SO-239s, I mounted a T/R relay on the ceiling of the cabinet.  Gives a good satisfying "clunk".
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
Todd, KA1KAQ
Administrator
Member

Offline Offline

Posts: 4279


AMbassador


« Reply #13 on: March 17, 2021, 07:52:00 PM »

I did a little research this afternoon, seems the mod iron was on the mod deck in the -1 and the first commercial version, the -2. It had insulators on the side instead of top like the later models, and also had a heat shield between it and the 75THs. Only ever heard of two 30K-2s, fortunately JP had one and I got to look it over when I lived down there. He posted photos on his site. I wonder if this is the one that the ol' N3JUH is running now?

I'll dig out those insulators, iff'n you want 'em you can get 'em when you come down. I just ran some RG-8X pigtails off mine as you know.

Logged

known as The Voice of Vermont in a previous life
kb9r
Member

Offline Offline

Posts: 8


« Reply #14 on: April 02, 2021, 11:53:52 PM »

John,
Thanks for the list of repairs to your 30K-1.  It was a fun read.  Not so fun for you maybe.
Anyway, what is the meter down below the Filament and Plate switches for?
Thanks for the post,
Joe - KB9R
Logged
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4614



« Reply #15 on: April 03, 2021, 03:00:30 PM »

It's a HV meter.  Non-stock, but useful.
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
WBear2GCR
Member

Offline Offline

Posts: 4041


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


WWW
« Reply #16 on: April 04, 2021, 12:31:16 PM »

Right perty!

Now you have to work Jim, in Beeth-le-hem Pennsylvania. What's his call?
My brain... oh my poor brain... anyhow he's on a 30K-1 all the time..

                  _-_-bear
Logged

_-_- bear WB2GCR                   http://www.bearlabs.com
w3jn
Johnny Novice
Administrator
Member

Offline Offline

Posts: 4614



« Reply #17 on: April 04, 2021, 09:38:19 PM »

The ol' N3JUH caw-mawn, cept I think his is a 30K-2

Jim's a hoot and always a blast to talk to   Grin
Logged

FCC:  "The record is devoid of a demonstrated nexus between Morse code proficiency and on-the-air conduct."
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 0.087 seconds with 18 queries.