Call Search
     

New to Ham Radio?
My Profile

Community
Articles
Forums
News
Reviews
Friends Remembered
Strays
Survey Question

Operating
Contesting
DX Cluster Spots
Propagation

Resources
Calendar
Classifieds
Ham Exams
Ham Links
List Archives
News Articles
Product Reviews
QSL Managers

Site Info
eHam Help (FAQ)
Support the site
The eHam Team
Advertising Info
Vision Statement
About eHam.net

donate to eham
   Home   Help Search  
Pages: [1]   Go Down
  Print  
Author Topic: Push Pull Transmitters with Tubes  (Read 5896 times)
AE7TE
Member

Posts: 57




Ignore
« on: October 18, 2017, 09:39:55 AM »

I've wanted to make a tube transmitter for a long time. People continue to point me to a pi-network transmitter such as is seen in almost every transmitter built by everyone from the 60s and 70s. There are a couple of things I don't want in my transmitter, such as slat board construction (I'm going to make it out of heavy metal plates) or crystal control (a digital VFO will drive the input tube, although I might design a tube VFO later) or single-tube operation. My extensive research on the topic points to a single tube oscillator/amplifier/transmitter asks so much of one tube that could more effectively be done by two, a driver/power amplifier. Were I to build a rig with a VFO, I'd have a third tube create the output frequency.

I tried a push-pull transmitter before. I used a 6CG7 in an attempt to make ~4w of RF. I made it in the traditional Jones 6A6 transmitter style, using small HC-49 crystals and 225v from the supply. It was disappointing, as I ended up getting differing power levels based on the crystal I chose, with the strongest crystal being a 6 MHz which gave me a full watt. Upon taking measurements of input current and voltage, I was putting 5.5 watts into the two triode plates to get that amount. I feel like none of the crystals were able to give me enough voltage to drive the transmitter, and had I used an oscillator tube to excite the grids properly, I'd have probably turned more of that DC power into RF.

So...I'm thinking of trying again. The code key would have to key both the driver tube and the two output tubes, where the driver tube would need a delayed circuit to ensure keying waveshape was good. I also don't think RF chokes in the grid circuits would be a good idea...a capacitor with a grid leak resistor would be essential to generating the right bias voltage for the tubes when they're on. I think that was part of the reason my last transmitter failed...the tubes didn't have enough drive power to turn most of the DC input to RF and they mostly ran class A. I also do not know if I want to make this transmitter with smaller tubes like two 6CL6 or 6V6GT, or go with something stouter such as 6L6, 6BQ6, or even 2E26. I have quite a few tubes that would do well for a medium-power transmitter. In practice, I'm not expecting to go higher than 10-25w for my first finished unit.

The last part would be making the output coil. I would (most certainly) have to make it a plug-in coil for each band. The literature (RCA TT-5) indicates the capacitor should be chosen for whatever band you want to tune, and the coil would simply be chosen to resonate to that capacitor. I'm obviously not going to change the capacitor out every time I switch bands. It's possible I could attach a fixed capacitor onto the coils for 160 and 80. I also have not operated on those bands...ever...so I may not even make coils for those bands. I'd definitely have to experiment with the number of taps on the output link.

Another thing I've wondered...if I wanted to plate-modulate the output of this transmitter, it would be easy if I could find a suitable mod iron. I could also use a standard single-ended OPT, and feed it from a modulator with its own output transformer. That would allow for a longer interconnect cable that didn't carry B+ voltage, and make things much simpler than trying to find a suitable modulation transformer. I'm also wondering how hard it would be to screen-modulate such a transmitter. I'm not quite sure whether I'd need to feed the screens of both output tubes in-phase or I'd have to split them into two.

Hopefully we can get a discussion started. It's annoying when someone tries to shut down the talk because they only ever learned one way to do things. I'd love to hear your ideas.

Ed AE7TE

Logged
G3RZP
Member

Posts: 8142




Ignore
« Reply #1 on: October 19, 2017, 07:04:43 AM »

You modulate the screens in phase. Of course, you take a hit on the carrier power, since ideally it's one quarter of the peak output power.

Tank coil value depends on plate volts and plate current and band. See any of the older 1950's ARRL Handbooks.
Logged
N3QE
Member

Posts: 4930




Ignore
« Reply #2 on: October 19, 2017, 09:30:46 AM »

In short: push-pull for tube transmitters can be a workable choice for a single-band amp. It gets really complicated to multi-band the amp although both pluggable coilsets and even turret band selection have been done (commercially available sub-assemblies even). Several commercial amps (notably Collins 30J) that did not have to be changed frequency often, were push-pull. The 30J RF deck is a thing of beauty.

Typically for link-coupled output network tuning both the inductor and capacitor have an impedance at operating frequency in the ballpark of the plate impedance. Plate impedance is doubled for push-pull which implies larger L and smaller C. Often the smaller C were in-phase-rotation butterfly capacitors which were effectively in series anyway.

Broadband neutralizing of triodes in push-pull can be done VERY NICELY with small-pF variable capacitors cross-coupling from the other tube. These small plate neutralizing capacitors look very distinctive.

Output networks for these push-pull amps almost always have exposed metal B+ plus RF voltages potential. Considering they have adjustable items like swinging links etc. safety is a real concern.

For ham construction articles:

April 1956 QST, VE2RZ, "Push-Pull 6146's in a Two-Stage Rig".

April 1948 QST, VE3ACL, "Notes on Push-Pull Triodes". I think this became a couple pages standard in the ARRL handbook for several decades.

Aug 1948 QST, W1TS, "807's in Push-Pull".

March 1941 QST, W1TS, "Push-Pull 809's in a Low-Frequency Transmitter".

And the article below which was a historical retrospective at the time it was written:

November 1955 QST, W6DTY, "A Pair of 45's in Push-Pull".

Logged
N2DTS
Member

Posts: 745




Ignore
« Reply #3 on: October 19, 2017, 10:52:02 AM »

Push pull is great, great with triodes:


Push pull 812a's modulated by 811a's.
Plug in coils are the way to go, and balance is VERY important.

That transmitter worked VERY well.
Triodes lend themselves to push pull and you do not have to worry about screens.

You can triode connect a lot of tetrodes as well.
For AM you need a mod transformer but some have used various other transformers with good results.
Logged
N3QE
Member

Posts: 4930




Ignore
« Reply #4 on: October 19, 2017, 11:00:51 AM »

Push pull is great, great with triodes:

Very nice! Is the white (ceramic?) tubular things standing up between the tubes, some sort of standoff for anode leads? Or are they also neutralizing capacitors? It looks like there might be top screws for adjusting the neutralizing capacitors. Are they vintage neutralizing capacitors or something made out of ceramic plumbing parts maybe?

I like the chain drive for the swinging link!
Logged
N2DTS
Member

Posts: 745




Ignore
« Reply #5 on: October 19, 2017, 11:46:49 AM »

They were some vacuum variables I got out of old paging transmitters, 10,000 volt 3 to 30 pf.
Neutralizing caps.
The chain drive allowed me to get the front panel layout symetrical.

The entire transmitter:



I never should have sold that transmitter.
Logged
G3RZP
Member

Posts: 8142




Ignore
« Reply #6 on: October 22, 2017, 03:29:52 AM »

For multi band use, you can use the 'multi band tank' concept. That's a coil and variable capacitor in series placed in parallel with another coil and variable capacitor. Needs bit of careful playing with so that it doesn't resonate at harmonic of the operating frequency. Have a look at the ARRL handbooks of the 1950s. National and possibly B & W made them as commercial units: they were superseded by pi tanks mainly because the pi tank offered better harmonic suppression when TVI was a problem, as well as being more flexible in terms of load impedance.
Logged
AE7TE
Member

Posts: 57




Ignore
« Reply #7 on: November 02, 2017, 11:45:28 PM »

All of the diagrams I have read show a split tank capacitor. That is, a capacitor with a rotor connected to B+ (RF ground) and two stators connected to either side of the coil. Is this critical or is it possible to use a single tuning capacitor? I'm sure that, were a builder to use a single tuning capacitor, the shaft would have to be coupled with a fiberglass or wood shaft.

I have a couple of dual-stator capacitors but the capacitance is small for both of them. Were I to use one of them, it would have to either use a larger coil, a higher frequency, or supplemental capacitance.

Ed AE7TE
Logged
G8HQP
Member

Posts: 611




Ignore
« Reply #8 on: November 04, 2017, 03:55:10 PM »

Push-pull transmitters were routinely used at VHF. If you use European style dual-tetrodes (e.g. QQV06/40, or smaller or larger cousins) then the neutralisation cross-coupling capacitors are adjusted for you at the factory.
Logged
KM1H
Member

Posts: 2633




Ignore
« Reply #9 on: November 05, 2017, 08:14:06 AM »

Derived from the US WW2 815, 829B, 832A and others. The later 5894, and several smaller and larger versions, was used in thousands of commercial and ham transmitters including the popular 50's Hallicrafters HA-2 and HA-6 and copied by several EU manufacturers.

I still use a modified HA-6 to drive a 1200W amp on 6.

Carl
Logged
G3RZP
Member

Posts: 8142




Ignore
« Reply #10 on: November 07, 2017, 01:22:21 PM »

I wonder how many working 832s and 829s there are these days? All the ones I've seen were WW2 manufacture and after 70+ years, the seals have leaked. I was told that the wartime tubes weren't expected to last more than a few months (or more precisely, the aircraft with the radios in weren't expected to last more than few months at the most, especially the B17s on daylight bombing raids in Europe.....so the tubes weren't pumped that well so they could get 'volume out the door'.

I suppose I should 'hi-pot' my pair of 832s......I might be agreeably surprised, but I doubt it...
Logged
N2DTS
Member

Posts: 745




Ignore
« Reply #11 on: November 08, 2017, 05:19:23 AM »

I have some really old 100TH tubes, they ALL still work great:



Pick up an old handbook, Bill Orr ones are very good for push pull info.
You can buy split caps cheap at fests and on ebay.
The plug in coils are the hard part.
Logged
KM1H
Member

Posts: 2633




Ignore
« Reply #12 on: November 08, 2017, 07:08:28 AM »

Quote
I wonder how many working 832s and 829s there are these days?

As with many TX tubes a lot depends upon who built them, did they improve over the war years or get worse?

I sold off my 826's and 829B's a few years ago on Fleabay and there were no complaints.
Logged
JS6TMW
Member

Posts: 1192




Ignore
« Reply #13 on: December 09, 2017, 05:07:35 AM »

I wonder how many working 832s and 829s there are these days? All the ones I've seen were WW2 manufacture and after 70+ years, the seals have leaked. I was told that the wartime tubes weren't expected to last more than a few months (or more precisely, the aircraft with the radios in weren't expected to last more than few months at the most, especially the B17s on daylight bombing raids in Europe.....so the tubes weren't pumped that well so they could get 'volume out the door'.

I suppose I should 'hi-pot' my pair of 832s......I might be agreeably surprised, but I doubt it...

You want short life? In the Atomic Bomb Museum in Nagasaki there is a battery-powered radiosonde using an 832A that was dropped a few days after the bomb to measure radiation.
Logged
KM1H
Member

Posts: 2633




Ignore
« Reply #14 on: December 09, 2017, 04:57:15 PM »

My last PP rig was a pair of VT-4/211's on 80 CW.

All voltages were regulated and the output went thru a balun to coax so I didnt have a wind blowing OWL causing frequency instability. Signal was quite stable with just a little Yoop when keyed.

I took it apart once I confirmed DXCC.
Logged
Pages: [1]   Go Up
  Print  
 
Jump to:  

Powered by MySQL Powered by PHP Powered by SMF 1.1.11 | SMF © 2006-2009, Simple Machines LLC Valid XHTML 1.0! Valid CSS!