Here's another fine mess I got myself into

Discussion in 'Technical' started by K4TQF, Dec 19, 2014.

  1. K4TQF

    K4TQF Member

    Some list members may remember my big plans to build an "old buzzard" rig from another thread on this forum. While accumulating parts for that rig, I ran across a "deal I couldn't refuse". It was cheap enough, close enough, and darn it, I liked it. It was purpose built as a 1000 watt plasma generator. It is currently set for operation on 13.56MHZ. It uses 2 X 4-400 in the final driven by a 6146 > 12BY7 osc. I picked it up Wednesday in Edmond,OK from W0BUW. Boyd, being a retired USAF fighter pilot, is currently net control for Airforcesflyersclub.net . They meet daily at 11:30 EASTERN, 08:30 PACIFIC on 14.290MHZ.

    The rig, as it stands, uses 3KV on the 4-400. They are operated in parallel. I suppose I'll leave well enough alone and use my 250THs as modulators if my RCA mod iron can stand it.
    Attached are some photos of the actual rig and a couple of others of a sister rig in the proper gray color.
    This rig was designed to operate into a 50 Ohm load, so I'll need to do some wire twisting to make it operate into my 600 ohm fed dipole. I'll need to wind some tank coils anyway, I guess I could make them link coupled. Too bad, as it has built - in power meters for forward & reflected, and alarm circuits for high VSWR.

    HFS 1000E.JPG
    HFS100 PS.JPG
    HFS1000E output.JPG
    HPF2000D controls.JPG
     
  2. K4TQF

    K4TQF Member

    As you can see, the 2KW model in the photo above has a crank style adjustment for the variable C in the tank. On the 1KW model, that I have, it's just a knob. It is quite hard to turn as the linkage is at a right angle to the variable. I'll probably swap this out.
    The meter to the right is switchable for reading all the critical voltages & current. The top two meters are the forward & reflected power meters. Looking at the schematic, I'll need to add a little more C to the B+.
     

    Attached Files:

  3. W5HRO

    W5HRO Administrator

    Oh, so you are starting to go the modern route now. What happened to sticking with the old vintage concept?

    Anyway, its looks good, but just don't turn into this guy or you will never be able to stop :lol:

    steve_in_shack_2009_small.jpg

    Are you planning on buying a home up there in Maine? I can see it now, you start going the class E route next, but sound just like HLR.
     
  4. K4TQF

    K4TQF Member

    Well... It started with converting the 808 rig to 250TH. But, that's too much of a butcher job. So, I've decided to restore it as built. As I was accumulating parts for the 250 th rig, I happened across the 4-400 rig. It's just a shortcut... I will probably wind up building another one to match the Norton "modulator" using the 8 X 211 OPT. Also, I could use that chassis and plug the 250THs into those same sockets, but use the HV from the 4-400 rig. Choices, choices...
    The house in Maine was built in 1900, totally re-done this year. New maple floors, new appliances & kitchen cabinets, etc... move -in ready.
     
  5. K4TQF

    K4TQF Member

    Here's a real puzzler. Looking at the HFS1000E schematic, they have 2 X 100K 200W bleeders paralled across the HV supply. The puzzler is... why would the designer put what is effectively 400 watts worth of bleeders in series with 13 ohm 2W & 1 ohm 25 watt resistor to gnd ? The only other connections to this power supply are for metering.
     
  6. W5HRO

    W5HRO Administrator

    Those are 50K of supply bleed which is about right. The 400W doesn't really mean anything, they just used two heavy 200W resistors in parallel for durability's sake. I didn't look a the screen protection circuit, but they may be higher wattage for protection in case the plates ever draw high current for any reason.
     
  7. K4TQF

    K4TQF Member

    Yes, 50K 400 watts in series with 13 ohm 2 watt doesn't make any sense to me in any case... :icon_crazy:
     
  8. W5HRO

    W5HRO Administrator

    Those are for HV supply bleed. All HV supplies around that voltage will have somewhere around 50K of bleed to help pad everything at initial start-up and mainly to discharge the filter caps after you stop transmitting. If not the caps would stay charged. Very dangerous. And if those resistors were not in series with the smaller meter shunt values to ground those shunts would burn up the second the HV came up :icon_crazy:

    I'm not actually sure where the confusion is :icon_wtf:
     
  9. K4TQF

    K4TQF Member

    Oh sure, that's one of the first things I learned about HV supplies 50 years ago... always have a bleeder. Keeps a little load on the output... a crude form of regulation and bleeds off the HV when powered down. But, the confusion is why have 400 watts worth of bleeders in series with a 13 ohm 2 Watt?
    It just doesn't make sense to me in a series circuit. the same current will flow thru those 100K 200 watt resistors as does thru the 13 ohm 2 watt... NO?
    @3KV you would have about 60ma flowing to ground. That's 180 watts, so they divide that by using 2 X 200 watt bleeders. Great, overbuilt, OK. But that same 60ma will flow thru the poor 2 watt resistor or .0468 watts...
    Nevermind....
     
  10. W5HRO

    W5HRO Administrator

    Yeah, but if the plates on the 4-400's ever draw excessive current for any reason everything would fry in a matter of seconds if it wasn't for those bigger value and wattage resistors in series with the smaller ones. The current in a series circuit will always be the same no matter if one resistor is a lot larger than the other. Simple Ohms law...

    The 50K of bleed is normal. They just placed those smaller values in series at the bottom of the string to ground for metering.
     
  11. W5HRO

    W5HRO Administrator

    Mike,

    It dawned on me what is probably throwing you off is that those tiny resistor values are never going to dissipate any real power because of the 50K in series with them. The voltage drop across them will always be very small. However, the voltage drop across the 50K will be very large and therefor dissipate a lot of power. Maybe that's where the confusion is.

    Anyway, Merry Christmas...
     
  12. K4TQF

    K4TQF Member

    Yeah, I worked that out in my previous post... #8
    TNX
     
  13. K4TQF

    K4TQF Member

    Actually, the shack layout, as originally planned, was to have the WW2 vintage rcvrs and the old PP808 rig to the left of the front door (as shown in the profile photo ). On the right side of the door I was planning on a really "old buzzard" corner that would have two tables with pre WW2 equipment. I suppose the 250TH rig qualifies to be moved over to that corner now. The schematic I'm following is from a Feb 1938 "Radio" magazine.
    Fixed C final, loop tuned, link neutralized PP250TH. http://www.americanradiohistory.com/Archive-Radio/30s/Radio-1938-02.pdf

    Just received 500LBS of denim insulation. My Christmas present to myself ! I guess I need to get away from this computer and go do some actual work !
     
  14. K4TQF

    K4TQF Member

    This old Collins looks like a good choice for a multi-band tank.

    Collins 185J-1 tank circuit.jpg
    big collins tuner.jpg
     
  15. W5HRO

    W5HRO Administrator

    Just find some B&W coil stock and make your own pi-network output to 50-ohms. All you need is a 150pF plate tuning cap then some switchable loading caps on the antenna side.

    I'd leave the 4-400A's in parallel and maybe configure the 6146 as a buffer for the T-386 exciter. Just tie its screen and plate together and use it as a wide-band cathode follower with no tuning to the 4-400A grids. You can probably run the exciter's output up to 7 watts by increasing the 6000 tube's screen voltage without any problem. Just keep it under 10 watts.

    Then use two 5933's in parallel for the 4-400A screen clamp or use one for each tube :icon_thumbup:
     
  16. K4TQF

    K4TQF Member

    Yeah, I may need to roll my own. I have a switchable tank coil out of a TMC-GPT-10K which was made for 2-28MHZ or thereabouts, with a little pig-tail to lower it to 1.5MHZ. Plasma-Therm chose not to use a clamp tube, but they have about 8 relays connected to protection circuits all over the amp.
     
  17. K4TQF

    K4TQF Member

    BTW:
    I don't have the original 6000 output tube in my T-3 exciter. I think i received it with a 6AU5 or 6AV5. Looking at the specs for the 6000, a 6146 looks to be a much closer match. Of course, it is not pin compatible, so I have made an adapter socket. I figured it may make a neat little stand alone QRP rig when not being used as an exciter for the "big rig". Still need to figure out the best way to key it for CW.
     
  18. W5HRO

    W5HRO Administrator

    No, don't even think of using 6146 in place of the 6000 tube or for its equivalent replacement tube. That exciter has built-in auto-tuned tanks circuits that won't be able to handle the power of a 6146. That's why I was also saying keep it's output under 10 watts.
     
  19. K4TQF

    K4TQF Member

    Well... looks like I'm back to the sub. (6AV5) As far as the plasma amp, it has provision for external excitation via BNC to S8 (ext). I was going to remove the 47 ohm resistor across that connection and jumper over to pin 3 on the 4-400, thereby by-passing the input tuning and other tuned circuits on the input side. Of course, I would pull the 6146 &12BY7 and everything in-between and including L17/R28. I would leave R23 in place.
     
  20. W5HRO

    W5HRO Administrator

    Not only that, but at +375V a 6146 is going to draw super high current. You would need to run it up to at least +500V to +600V to get it down and I doubt the exciter's output circuits can even handle it either way.
     
  21. W5HRO

    W5HRO Administrator

    Below is an idea of what I was referring to. You could use the 6146A as a wide-band buffer for the T-368 exciter at the input of the 4-400A grids.

    idea.png

    You could also use a 5933 or 807W as a clamp tube and eliminate the other fixed supplies. You would get better modulation doing so. You would only need one 5933 and not two. Or just substitute it with an 813 for super durability.

    Hope this helps and gives you some ideas..
     
  22. K4TQF

    K4TQF Member

    Looks good to me... My new years resolution is to build a fire every morning in the shack and do something... even if it's just for a few hours. I have to get it insulated & the inside walls up before I can seriously get to work on any radios. I have the T-3 exciter in my old shop and will try to get a decent supply built for it this winter.

    Thanks, MD
     
  23. W5HRO

    W5HRO Administrator

    Yeah, one of the reasons why I suggested using the 6146 to buffer the exciter is because the 6146 is already there. However, the main reason is because of what we already discussed. The exciter is high Z out so using two 4-400A's in parallel will just make it's load even lower Z. The T-368's RF deck has around 20K of grid leak and with two 4-400's it will be 10K. If you go by the data sheet then at -220V it is even slightly lower than 10K. I put 9.1K on the drawing because it's a standard resistor value, but using a 10K would probably work fine.

    If you decided to do it just don't use a resistor larger than 100K from the grid of the 6146 to ground. Since the T-368 exciter was setup for a 20K load then I'm not sure what it would do with a 100K load. It could be much better and it's output power might come way up. A 100K to 10K step-down could work great, but a 30K or 20K to 10K step-down may still work better. You would just need to try both and find out.
     
  24. K4TQF

    K4TQF Member

    In other words... a "build-out" type load. It's a real hindrance for me to have worked in audio most of my life. I automatically tend to think that way... but, they are all just waves, albeit at a different frequency. I'll need to do a little reconfiguring as the HFS-1000 was fixed at 13.56MHZ. There are some traps, suppressors, etc... that will need to be removed or otherwise changed surrounding that 6146.
    I may wind up padding the output of the T-3 exciter way down.
    IIRC, you were using a 6146 driver in your original setup, no?
    Tried to do some work in the shop this AM... blowing snow & 17 degrees outside. :icon_thumbdown: No snow and 45 degrees in the shop. :icon_eh: I didn't last long... now back in front of the fire pouring over old manuals and Army TMs on transmitters. :icon_thumbup:
     
  25. W5HRO

    W5HRO Administrator

    Well, sort of... What I was suggesting is using the 6146 they already have in there as a cathode follower buffer. The T-368 exciter would only see the load at the grid of the 6146 via whatever resistor to ground you use. It would just be a buffer stage. The only difference between an AF buffer circuit and an RF buffer circuit is that you must take into account that it is RF. That 250uH plate choke would be important. That's another thing, you probably need to change the plate choke on the 4-400A's as well. An example would be you need at least 90uH to 100uH for use on 75/80-meters. Closer to 200uH or more for 160-meters. I found a 250uH choke for my 4-400CG. Just use the same thing for a 6146 buffer.

    I never had any driver tube in my RF deck. I just used another transmitter attenuated for an RF driver with the link coupled input circuit I recently removed. The T-368 exciter will drive my 4-400CG tube just fine all by itself.