some figuring on a pair of 4-65

[KC4VWU]

Here is what I came up with. I figure I can use a pair of 811A's class B and drive them with audio amp since I have quite a few guitar and small PA amps sitting around. Looks like I'll need at least 200w and close to 6750 ohms secondary for the mod iron. Does everthing look O.K. so far?

Phil

[N2DTS]

The books say a 4-65 at 1500 volts and 120ma will do 140 watts out.
180 watts in, 140 out.
For two tubes, 360 watts in, 280 watts out.

You would do better at 2000 volts, or 2500, but you can run 811a's at 1500 volts zero bias....

A 4-125 does the same.

Something like a 100TH might be a better choice, you could run them at 1500 to 3000 volts.
Its always good to have as much audio power output as DC input to the rf deck.

A 2X4-65 rig modulated by a pair of 100th's would be more interesting, 811's are so mainstream...
Plus, both tube types would look really cool when you strap things!


Brett

[KC4VWU]

That's a good idea Brett. I was thinking the 811's because I have a number of them. But I also have 5 or 6 4-125's, sockets, etc.  The proposed plate iron is from the GSB-101; 3600v C.T. @ 500 mils. I plan on using the p/supply as built in the Gonset; 866A's, filter choke in the CT. I'll have to take a look at the 4-125 specs.

Phil

[KC4VWU]

A pair of 4-125's in AB2, 1500v @ 400 mils max signal, gets 350 watts out. Now, all I need is suitable mod iron and reactor.

Would they work / better triode configured?

[N2DTS]

Running the 4-125/250/400 needs some attention to details if you want it clean.
I would run them in AB1 for less distortion, with regulated screen voltage.
 
A 4-125 modulated by a pair can work fine, plus, if you use a big enough filament supply, you can plug in 4-400's!

The problem with these tubes is they like high voltage, for the 4-125, 2000 to 2500 volts, for the 250 and 400, 2500 volts and up, otherwise, you don't get much power out.

Your power supply should do 1700 volts choke input.
Not a good voltage, to low for the 4- series tubes, to high for 811a's, although fine with bias.
A push pull 811a rig modulated by a pair would do real well, I used to run that setup, and it worked real well.
An 813 would do well, 300 watts for one, 600 watts for two.
A pair of 4-65's modulated by a pair might be good.

Brett

[N4LTA]

Running a pair at 2500 volts gets the modulating resistance above 11500 ohms.

That's why I was looking at 1500 volts. They are very efficient at 1500 volts on the plate. The modulation resistance drops to 6250 ohms which may make the modulating iron easier to find in a 200 watt transformer.

RCA's data sheet shows the max dissapation as 45 watts in am service.  They show 108 Ma at 2500 volts which is 270 wats DC input.

At 75% efficiency that leaves 67 watts of plate dissapation so it must be a misprint.

I'm going to run my pair at 1500 volts for a DC plate input of 360 watts. My modulator does an easy 250 watts so it will have plenty of power to modulate the pair. The Modulation transformer will give me a good match also. They will be running at about 45 watts per tube plate dissapation.

Pat
N4LTA

[KM1H]

1700-2000V is perfect for 572B's at zero bias. Find a real cheap pair that were pulled for low output on 10-15M but fine at 40-80M in a SB-200 or similar and they will be fine at audio.

Carl
KM1H

[N4LTA]

I read in some Eimac literature that the 4-65 is best at lower voltage - 1500 volts. It is very efficient there.

I also found another data sheet limiting plate dissapation to 45 watts in AM service -

http://www.rell.com/resources/RellDocuments/SYS_1/QB3200.pdf



Don't know if this is the same mistake copied from the RCA data.


edit NOPE - Here is The Eimac data sheet - 45 watts max plate dissapation in anode modulated service.


http://www.shinjo.info/frank/sheets/088/4/4-65A.pdf

Eimac gives a plate efficiency of 80.5 % at 1500 volts on the anode. That is pretty good at that voltage.

Pat
N4LTA

[N3DRB The Derb]

yeah 572b's would fit real well. thats the hot lick. 4-65 was designed for low band VHF use including using with mobile HV supplies, thats why it's better going at lower plate joltages. I forget who used em, Mot or GE but quite a few mobile comm units used he 4-65.

[N4LTA]

I'll bet they are more apt to take off on their own also since they are a VHF tube.

Pat
N4LTA

[KC4VWU]

I'm planning on just making this one for 80 and 40. Being a tetrode, it shouldn't be a problem at the lower freq's. I think at this point, I'm going to just start on the supply and RF deck and get it right, then build the modulator. Looks like whatever output ckt scheme is used, dead on calculations and extremely good layout is going to be imperitive to meet that efficiency factor. Maybe push-pull link coupled output with plug-in jackbar coils. I'm probably not going to be band hopping much. 

At any rate, I believe I may lay it out so that two more toobs could be added in the future. I've got a lot of reading ahead of me!

572B's sound good too, Carl. I don't have any on hand at the moment, but things could change before I get to that point.

How are you planning to set yours up Pat?


Phil

[N2DTS]

You can do push pull, or pie net using a plug in coil.
I did pie net plug in coil on the 3x4D32 rig, and did add the 3rd tube after the fact...

Push pull stuff is easy to neutralize.

Pie net, you can do one, two, three, or four tubes, not just two or four, but things can get odd having 3 or 4 tubes in pie net....

Plug in coils for high power are hard to get, old ones have crumbled, so you might have to make them.


Sounds like a very fun project.
Let me know if you need any parts, I might be able to help...

Brett

[N4LTA]

I'll run mine  two tubes in parallel with a Pi Net.

Probably 160 -80 -40 .

I have a control unit built with a variac in front of a very stiff screen supply with a optically isolated screen current trip circuit and a optically isolated plate voltage detect circuit. It also has a solid state timer to delay the plate voltage until the filaments are up. It has a regulated grid bias supply. It was designed for a single 4-400A .

My RF plate supply is variable from 1000 to 2250 volts.

My modulator power supply is build on the control chassis - 1250-1300 volts with a large 500 mA  Hammond choke  at the input of the supply. I have some Svetlana 572s and some 811As - I may go with the Svetlanas for the modulator. The iron is a big 40 pound vintage Hammond  with ceramic standoffs. It look very sturdy.

I have been playing with a homemade dual 1625 transmitter modulated with a Hammond 125GSE and three Hammond 193J choke in a modified Heising arrangement. I have a Crown  solid state amp to supply the modulation power for the 125GSE. I should be finished in time for the AM contest in early February.

I finished the driver/doubler circuit this afternoon. Should be able to generate some RF tomarrow. I have used toroid inductors throughout, including the output Pi Net.

I used a  T400 -2 a few years ago on a 4-1000 amp for 160 - 80 -40  meters with great results.

Also need to throw together a VFO for the above. Not much time.

Then I'll go back to work on the larger transmitter.

Pat
N4LTA

[KC4VWU]

Thanks for the offer Brett, because I'm probably gonna need some stuff. I know I'll need some help in the design and layout, and will appreciate any that is given. I've built some PW stuff before, but this is different in that it's not a construction article out of a handbook or mag; kinda like starting from scratch.
 
I've got so many ideas and things to do that I could get easily sidetracked on this so it's back to the books to get a schematic ready. Then I'll take a parts inventory to see what I'll need.

That layout sounds cool, Pat. You need to post some pics when your finished with it. Mine isn't going to be ready for the AM contest unless a miracle happens. I'll probably have to run tha V II; hope propagation is good.

73, Phil

[W2PFY]

Here's an Eimac PDF that has not only the spec's for the tube but practical circuits with all values of caps etc.

[N2DTS]

I never built anything by a book design, just get some ideas and design my own stuff.

I think the big thing building push pull rigs is a symetrical layout.
That can be hard to do with what parts are on hand, when I built the 812a rig I had to come up with some sprockets and chains to drive the swinging link, since I wanted a symetrical panel layout AND a symetrical rf layout.

For screen grid tubes, you need to figure out what you are going to do for screen voltage, protection, and self modulation.
I like a variac supply and a big wirewound pot as a screen dropping resistor, and an overload relay.
The dropping resistor allows the screen to self modulate and self limit the screen current.
I drop a few hundred volts in the dropping resistor, if the screen draws to much current, more voltage is dropped in the resistor, which allows for tune up without danger.
I meter both the voltage and current.

Fun

Brett

[KM1H]

I think the big thing building push pull rigs is a symetrical layout.
That can be hard to do with what parts are on hand, when I built the 812a rig I had to come up with some sprockets and chains to drive the swinging link, since I wanted a symetrical panel layout AND a symetrical rf layout.

I'll think you'll find that a symetrical layout for PP was determined by several authors to be the reason for neutralization and parasitic problems. Short leads but a slight offset difference was claimed to be better.  I havent used PP for 4+ decades so cant offer any first hand experience but I suspect it is correct when using the world above 20M. Anything works below that.

Carl
KM1H

[N4LTA]

Don't know - I have never built a PP RF amp - but I learned a long time ago - that pretty can be a bad thing when it comes to RF and high speed digital layout.

If you have every done much PC Board layout - you know what I mean.


I have had many instances of where an old jumbled up ugly construction layout worked great  - or a rats nest wire wrap design  - when it got layed out real pretty on a PC board had terrible problems with instability.

Pat
N4LTA

[W2PFY]

Short leads but a slight offset difference was claimed to be better. 

Maybe I missed something here? Do you mean that if you had two tubes in PP that one should be perhaps 1 inch or so to the front or back of a chassis?

I could be wrong again?

[KM1H]

Put the tubes wherever you want. Make the grid and plate leads slightly different lengths for one of them.

Carl
KM1H

[N2DTS]

Not being very smart at designing RF stuff (or anything else), I read the Bill Orr handbooks where it said a symetrical layout was important, and built my rig that way.
I got lucky somehow and it worked fantastic, the first and only RF deck I ever had perfectly neutralized.

I never should have sold that rig, as the motorcycle I got did not suite my tastes at all.
The most uncomfortable bike I ever owned (2005 Triumph Bonneville).
 
I traded the bike in on a rav 4, I have another bike..