Gonna Build a 2 Meter Tube Amp

Discussion in 'Amateur Radio Amplifiers' started by K2WH, Jul 4, 2011.

Thread Status:
Not open for further replies.
ad: L-HROutlet
ad: l-assoc
ad: Subscribe
ad: L-rfparts
ad: l-innov
ad: l-Waters
ad: l-rl
ad: l-gcopper
  1. K2WH

    K2WH Premium Subscriber QRZ Page

    I have decided (after blowing out 2 solid state amps for 2 meters) to build a tried and true rugged single tube 2 meter amp. A single 4CX250 or 150 vacuum tube, can put out 300-400 watts easily and, very common to find at and; that power level is all I need or want.

    I used to build them in the 70's and 80's but as I got older, I sold most of them and never really got back into building. However, after the loss of these 2 amps, I decided to go back to what I know works reliably without constant monitoring for problems. Strangely, all of the stuff needed to build the amps in the 70's and 80's could be had at hamfests. No more, after perusing Ebay, just about everything and anything needed for the amp can be obtained there. Tubes, tube sockets, finger stock, copper stock, doorknob caps, feedthrough caps, variable caps, connectors, transformers, chassis's etc.

    Sorry state of affairs but very convenient.

    Last edited: Jul 4, 2011
  2. WB2WIK

    WB2WIK Premium Subscriber QRZ Page

    Actually, you can find most of these items "new" from RF Parts.

    Some other stuff you can get from Ameritron as "spare parts," they sell everything used in their amps for fairly reasonable prices.

    Fair Radio Sales in Ohio has a lot of stuff, too.

    My 2m kW amp uses a pair of 4CX250Bs in P-P. Built in 1979 and never serviced, still works fine with the same tubes. The only maintenance ever performed is to blow out the blower, sockets and tubes with an air hose every couple of years to get the dust out of them.

    If you want it for FM only, even easier since they'll be in cutoff and not draw any current at all until needed. For FM you really don't even need regulated bias or screen supplies, since linearity isn't required.

    GREAT project. A good circuit to emulate is the one used in the Gonset 903A amplifier, it's very efficient and easy to copy. Running Class C for FM, it hits about 70% efficiency and makes 250W output with a few watts drive.

    They silver plated the plate tank stripline, but I've made "copies" of this amp just using bare copper flashing and it didn't make any difference.
  3. K2WH

    K2WH Premium Subscriber QRZ Page

    Hey Steve. Actually I will be using the amp in SSB service so I will need to provide stiff regulation of the screen supply. As you say, many of the parts are available on line. When you think about it the most difficult part in building your own vacuum tube amp IMHO is the required power supply. Can't just hook it up to the 12vdc supply.

  4. KM1H

    KM1H Ham Member QRZ Page

    Using 70's circuitry you will be the scourge of the band on SSB in a crowded area as IMD is in the low 20's at best. Use a G3SEK tetrode board or a clone and upgrade with a Ruskie 4CX400A (600W per tube max vs 300 for the 250) which is a cleaner tube to start with or find the elusive 4CX350F. Many 150/250 pulls were from commercial FM and have likely developed screen emission which further destroys the IMD.

    The Europeans went on a 150/250 cleanup mission years ago due to population density and their wide availability which became a serious problem.

  5. W0BTU

    W0BTU Ham Member QRZ Page

    I didn't realize they were that dirty of a tube, Carl. I ran a single grid-driven 4X150A on 144 MHz SSB/CW many years ago with 275 volts regulated on the screen and 2 kV (IIRC) on the plate. Input and output circuits were stripline. I never thought to check how wide my signal was, and no one complained.
  6. K2WH

    K2WH Premium Subscriber QRZ Page

    Agreed. I built many 2 meter amps using 250's and 150's and never ever had a single complaint. Perhaps that was because of the stripline construction and the stripline filter on the output of the amp. Since the website references power and control schemes for tetrodes not amplifier design considerations, perhaps the use of these tubes on HF put out a bit more garbage than on VHF. I reviewed the kits and the theories etc. (thanks for the link), but I would rather roll my own. Its more fun and much more self satisfying.

  7. WB2WIK

    WB2WIK Premium Subscriber QRZ Page

    I think the issue with the 4X150/4CX250 family is just moderately acceptable IMD performance, which hasn't much to do with the tank circuits.

    My P-P dual 4CX250 amp has been on the air in hundreds of contests using big antennas and lots of strong local signals and never had a complaint, either, even operating 5 kHz away from other big guns in the contest. I run 2200V Ep, 350V Eg2 (regulated with gas tubes!), and no grid current -- careful loading will achieve that. It can develop 800W PEP output with no grid current indicated at all, if it's loaded well.

    These things probably get "sloppier" when overdriven or underloaded so grid current starts showing up.
  8. W0BTU

    W0BTU Ham Member QRZ Page

    I said what I did just to show it had a tuned input. Stripline vs. coil isn't going to make much difference in IMD, but tuned input vs. no tuned input might.

    I was just wondering what Carl meant when he said 70's circuitry. I think he (and you too) knows more than I do about amplifiers. :)
  9. WB2WIK

    WB2WIK Premium Subscriber QRZ Page

    My 4CX250 amps were all grid driven and required such little drive I normally had a 6-10 dB 50 Ohm resistive pad between the exciter and the amp, so IMD would likely be the same no matter what the input circuit looked like. However, since they're push-pull and required phase splitting, I did use a tuned input circuit in all of them, at least to accomplish that. If I used high-Q "tuned lines" (1/4" copper tubing to make loaded 1/4-wave line sections) they would drive easier than with lower-Q "lumped constant" input networks, like a small coil and link, but since there was always too much drive available anyway, it hardly mattered.

    My "plumber's delight" 2m P-P amp can be driven to 600W output on FM using the output of a 2W handie-talkie to drive it, if I omit the 6 dB pad. I've demonstrated that a lot of times to visitors, who, if they grew up in the "post vacuum tube" era, were usually astonished.

    I usually drive it with an FT-736R followed by a 6 or 10 dB pad. With proper loading, it runs real AB1 (no grid current at all) and is still about 50% efficient with no grid current. I can push it into AB2 and achieve a bit more output and efficiency, but who cares? It runs plenty of power.
  10. W0BTU

    W0BTU Ham Member QRZ Page

    Oh yeah! I did have a grid swamping resistor on my amp, which I almost forgot about. I drove it with a Hamtronics transverter which put out 2 watts, still way too much without the resistor.
  11. KM1H

    KM1H Ham Member QRZ Page

    In the 60's and 70's very few HB VHF amps used regulated anything, loading was by guesswork, used tubes were often used, and the magazine articles didnt help nor did hams trying to run more than about 600W out of a pair.

    Nobody complained because they were the norm.

    Several years ago I measured a customers Johnson 6n2 Thunderbolt at -23dB 3rd using NIB 4CX250R's. Then gutted the bias and screen supplies and used the G3SEK boards with a resultant -29dB. The listening test with it back at the customers was significantly improved.

    When I converted a NCL-2000 to Svetlana 4CX400A's for George Baxter, their VP of Mareting, the stock PS and 8122's were also -23dB, after going to the G3SEK boards it was also -29dB. With the 4CX400A's it was -33dB.

  12. W0BTU

    W0BTU Ham Member QRZ Page


    My own screen supply used electronic regulation (paralleled 5881's/12AU6/OB2 in my HB 32S-3 power supply). The +275V (kind of low, yes) didn't vary by even .01 volt. And the main indicator of loading was when the screen current was zero or VERY slightly negative.

    The grid bias was not regulated, but since we never drew grid current, it was almost as constant.

    The G3SEK triode board has been on my wish list for almost a year now.
  13. WB2WIK

    WB2WIK Premium Subscriber QRZ Page

    My screen regulation was usually a pair of 0B2s and a 0A2 in series, drawing about 30mA through the gas regulators.

    But regulation was pretty close to perfect, since in AB1 the amp never drew any screen current anyway.:) It would hit maybe 2-3 mA on voice peaks on SSB, but that's about all I'd ever see. Again, proper LOADING was crucial, as loading greatly effects screen current. In fact, my final loading adjustment would be key-down CW and adjusting the loading for zero screen current, which almost always coincided with maximum output power anyway.

    The T-bolt wasn't bad, I had one. It used octal tube gas regulators and if you loaded it properly, it was fine. The only "issue" I ever had with the T-bolt is it could make a pretty good tripler if you drove it on six meters and had a two meter antenna connected, since the plate tank circuits were always both in line. What I'd do to make sure I wasn't transmitting a signal on 150 MHz was use Dow Key relays that would only key on the band I was using, and terminate the unused port into a dummy load. They only sold about 350 of those or so during the product's life, so not that many are out there.

    I remember I sold my 6n2 T-bolt to W2ZND, and since I used it on AM sometimes (1969 or so) he thought it was an "AM rig." I did explain it's only an amplifier (I drove it with a TX-62 sometimes on AM), but I'm not sure he ever quite got that. When he came to pick it up (and pay me $1000 for it, which was $400 more than I paid!), he looked inside and said, "Where are the modulators?" Oofa.

    Well, he had lots of money and I didn't feel bad when I noted his chauffeur was the one picking up the amp to put it in the back of his Bentley.:p
  14. KM1H

    KM1H Ham Member QRZ Page

    When I was building single and 2 hole 4-1000 amps "to order" in the 80's I used the 5881's and a 6SJ7 (octal 6AU6). About 2 years ago I ran across a few dozen NIB Raytheon 5881's I had forgot about and got obscene money on Fleabay from audiophools as "matched pairs". I had bought about 100 at a surplus dealer for almost nothing in the late 70's.
  15. K2WH

    K2WH Premium Subscriber QRZ Page

    Steve, you got a schematic of the regulator system you used? I'm thinking of using the gas tubes also or zeners but I would rather stay vacuum all the way. What about the transformer or did you tap off the HV supply?

  16. WB2WIK

    WB2WIK Premium Subscriber QRZ Page

    Bill, these circuits were in the Handbook and the VHF Manual for many years, in the 60s and 70s (and some probably in the 50s). Not any more, but if you pick an old Handbook or VHF Manual from 1964-1974 or so I'm quite sure the same "plumbers delight" type dual 4CX250 amp circuits were in all of them.

    0A2 = 150V at 30mA
    0B2 = 105V at 30mA

    Two 0B2s plus one 0A2 in series = 360V.

    I take the screen voltage from the B+, usually, if for no other reason than to assure if the B+ fails, so will the screen voltage, and at exactly the same time! You'd never want B+ to go down while screen voltage is still present...the screens can't handle that.

    To drop 2200V to 360V at 30mA requires a 60K, 100W resistor (or so). I'd usually use a pair of 30K, 50W wirewound (ceramic tubular) resistors in series. They're "long" enough to handle the voltage just fine.

    You can use a separate screen supply with its own transformer or transformer winding but I'd employ a relay in series with the screen voltage to the tubes which would drop out if the B+ disappears for some reason, or some way to protect the screens in the event of a HV failure.

    Then again, if you have a lot of spare tubes, maybe it's not very important.:) At one time I probably had 30-40 4CX250Bs "in stock" and wasn't too worried about running out of them. Those were the days when you could pick them up at Dayton, brand new, in sealed olive drab JAN cans, for $20 each. I don't think they're available quite that cheaply anymore, but maybe.
  17. W0BTU

    W0BTU Ham Member QRZ Page

    I do. If you still need it, send me a PM too, or better yet, an e-mail. I'm liable to overlook it if you only post here.

    The circuit I built was for my Collins 32S-3. After replacing a couple of sets of 5881s, I added some resistance in parallel with the tubes to make them run cooler. I haven't changed those tubes since I did that.
  18. K2WH

    K2WH Premium Subscriber QRZ Page

    Thanks for the reply Steve. My first power supply I built had variable voltage levels for all voltages required including the plate supply. Variacs on the Plate, screen and grid. Later on I found an old regulated Lambda vacuum tube supply 0-400vdc. Worked perfectly. This time however I will be using hard wired plate transformer and as I suspected, a dropping resistor and regulator tubes.

    Yes I do have a very nice (almost new) VHF manual from the 70's and some real old Handbooks to research, but you provided what I needed. Incidentally, for screen protection I designed and installed a simple cutoff relay using optical coupling. I forget exactly how I used it but the IR emitter and IR detector along with a 2N2222 switching transistor shut down the entire supply if the screen current went too high. I think I put the IR emitter across a 1 ohm resistor in series with the screen supply and at a certain current draw, the IR emitter would "Lite Up" and the IR detector would shut the supply down. It was foolproof. Actually got the parts from good ol Radio Shack.

    Thanks again,
  19. KA5LQJ

    KA5LQJ Ham Member QRZ Page

    Hi Guys!
    I'm a noobie to the forum. :D I have some questions, if you don't mind.
    #1. What would be a good, inexpensive, tube to build a kW ssb/fm 2 meter amp. #2. Where can I get the aluminum chassis and cover for such a project? The folks I used to deal with all went out of business or were bought up.

    Thanks in advance,

  20. W8JI

    W8JI Ham Member QRZ Page

    As long as screen voltage source is 2X the required screen voltage or more, it is impossible to have more screen dissipation than you set as quiescent dissipation in the shunt regulator.The sole exception to this is very high values of secondary emission causing negative screen current. In this case dissipation can be higher, but that is a rare condition.

    To satisfy that condition I used either a center tapped transformer as the screen source, or a different winding on the plate transformer.

    When you get voltage that way, through a dropping resistor with at least 2X screen voltage from the source, it doesn't matter if you lose anode voltage or not. It is electrically impossible to exceed screen ratings. The system does not need to be complex.

    Tuned inputs do not help IMD in grid driven PA's. Padding the grid with a resistor sometimes does, as does adding an unbypassed cathode resistance for negative feedback. With a little unbypassed resistance in the cathodes and good screen regulation a 4CX250B can make around low -30's for third order.

    A 3CX800A7 can make around -45 dB on third order. :)

    The real problem is many two meter radios seem to be pretty crummy. I think people don't complain on two meters because they are used to the splatter.

    73 Tom
Thread Status:
Not open for further replies.

Share This Page

ad: NSIComm-1