Which is better design?

[W4TWA]

Hey Guys,
I wanna cut my teeth on a low power transmitter/modulator, due to the fact I would like to learn before I go qro. Which do you think is better? This 75 watt plan using a 6146 screen/plate modulated by a pair of 807's found here:  1954 handbook (arrl) pg. 162. Modulator found pg.252 of same book.

Or....this page 8 of this PDF

http://www.ami-west.com/Thordarson.TX.Guide.344E.pdf

The arrl job uses lighter power x- formers, 120 ma for x-mitter, and 90 ma for the mod.
A 6146 would like 140 ma for class c rf amp at 600 volts, according to the specs. So it seems the design would be power starved.

The thordarson seems more durabley designed, yet requires beefier power transformers rated at 225 ma. And more of them.

What do you guys think? Would th arrl design be puny and a waste of time? It sure would be easier to get the power formers for this than the thordarson design

Your thoughts?
Paul, W4TWA

[N2DTS]

Every design falls apart when it meets the junk box.
Its not like you can go out and buy most of the parts needed to build a rig.

On every transmitter, there is a weak point or limiting factor.
It can be the tubes, or the power supply, or the tank circuit components, but something is going to limit the power output.
The best designs run everything at a maximum ccs rating, no space or money is wasted in components that are oversized, nothing is running beyond what it is designed for.

I have rigs that are limited by the power supplies, voltage or current ratings are too low for what the tubes can do, and I have some that are close to all CCS maximum rated.

If you can get 700 volts and 120 ma out of a transformer choke combo, that is what you have to work with.
There is no reason you could not use a 700 volt 200 ma trans/choke setup.

Most designs you will find in handbooks work quite poorly, they were doing the same cost cutting and crappy audio designs as everyone else was using.

[KA2DZT]

First thing you have to do is see what xfmrs you have.  A 6146 at 650-700 volts on the plate will put out about 50 watts.  Run the 807s off the same supply.

So, tell us what xfmrs you have, and I can tell you just what load you can put on the xfmr.

Fred

[W4TWA]

Fred,
I have the arrl power transformers, 360 each side of ct at 120 ma, and the one for the modulator 350 each side at 120 ma. For the mod trannys I have a UTC-CVM-1, per the handbook rated for 30 watts. I also have the CVM-2 on hand rated at 60 watts. Both are vari match.

[KA2DZT]

Fred,
I have the arrl power transformers, 360 each side of ct at 120 ma, and the one for the modulator 350 each side at 120 ma. For the mod trannys I have a UTC-CVM-1, per the handbook rated for 30 watts. I also have the CVM-2 on hand rated at 60 watts. Both are vari match.

I'll assume for now that they are Stancor xfmrs.  If you have the numbers from the xfmrs I can tell better.

The two you have can work for what you're trying to build.  Both xfmrs will have to run with FW bridge rectifiers to get the 600-650 volts you need.  Problem is with FW bridge service it puts twice the VA load on the xfmrs.  To offset this heavier load, the first thing to use are solid state rectifiers.  This removes the load of the tube rectifier filaments.  A separate 6.3 volt filament xfmr for the other tubes would also lighten the load on the two xfmrs.  The next thing is you will have to use is choke input filters.  Choke input filters greatly reduce the VA load on the xfmrs.  Some xfmr current ratings (Stancor) are listed for capacitor input filters.  Using the same xfmr with a choke input filter the current load rating can be increased 30%.

Use the 60 watt mod xfmr for this set.  The smaller one can be used on another project, even can be used for a driver xfmr for a big rig.

So, what did you say the part numbers on those xfmrs were.

Fred

[N2DTS]

I looked up the design for the ARRL handbook one, and it says the power supply is a full wave bridge setup using a cheap TV transformer.
Whatever gets you 600 to 700 volts out will work.

The modulator needs work, regulate the 807 screen voltage to lower distortion, change the cap and resistor values to get some low frequency response, just a bit, and unless you want to be true to design, solid state the rectifiers and save all that filament power and voltage drop in the tubes. Saves a lot of space as well.

[KA2DZT]

A word about cheap TV xfmrs.  In most cases xfmrs out of old TVs were better than the regular catalog models from Stancor or Thordarson.  I use three RCA TV xfmrs in my HB xmtr.  They date back to the early 1950s late 1940s.  Many years ago I bought about 20 of them from one of my dealers for 2 bucks each NOS in boxes.  They even packed them with newspaper which was fun to read.  Old TV xfmrs can handle a lot of overload without any problems.

Fred 

[N2DTS]

I think they meant inexpensive.
And in 1954, they all would have been IN TV sets, not scrap like later...
My first home brew TX used one out of a dumpster.

[W4TWA]

I have a stancor 92964, and the other is a Freed R17t7218-152. As to how they are rated, I.e.condenser or choke input, I don't know.

[W4TWA]

I'm not opposed to getting other iron if need be, it's just they are not that simple to find always. With the numbers quoted on the 6146 being 140 ma, I just figured a 120 ma transformer called for in the design sounded kind of light weight. I would have expected them to call for at least a 140 ma transformer.
Is the 140 ma stated in the tube data reflect the average current drawn, or just the max seen on voice peaks?

[Steve - K4HX]

If the 140 mA is from the Class C section of the data sheet, it's average current.

[KA2DZT]

I have a stancor 92964, and the other is a Freed R17t7218-152. As to how they are rated, I.e.condenser or choke input, I don't know.

Five digit part numbers were special xfmrs made for some manufacturer of some piece of equipment, not a regular catalog model.  So, you can only work from whatever rating is on the xfmr.  The Freed xfmr may also be a special, most were.  I'll check my Freed data to see If can find that part number.  The xfmrs will work with either choke or cap input filter.  The choke input filter is much easier on the xfmr.  I would not worry too much about the current rating being a little light.  Remember, Ham use is intermittent were as the xfmrs were probably meant for continuous operation.  Bottom line to the whole issue, if the xfmrs don't get too hot, you're not overloading them.  OTOH if you're able to fry eggs on top of the xfmr, you're going to need a bigger one.

[w1vtp]

Large file size warning!!  The original link points to an 8 MB file.  Just so you know.  I applied my "reduce file size" option in my Adobe Acrobat and got it down to 5 MB

Wonderful piece of history.  I fail to see any significant advantage of the older design.  A 6146 (or better yet 4D32) based transmitter would work fine with a good antenna.

Al

[N2DTS]

From the back of the old ARRL handbook, the 6146 has 25 watts of plate dissipation, and if run at 600 volts should be around 112 ma. That is 600 volts times .112 amps for 67.2 watts input, about 52 watts output.
67.2 -52 equals 15.2 watts dissipation in the tube out of 25.

For CW, they run it up to 750 volts and 120 ma!

In real life, you can run the 6146 at 800 volts or more, I run some 4D32's at twice the voltage (1200) then they are supposed to have on them.

[w1vtp]

From the back of the old ARRL handbook, the 6146 has 25 watts of plate dissipation, and if run at 600 volts should be around 112 ma. That is 600 volts times .112 amps for 67.2 watts input, about 52 watts output.
67.2 -52 equals 15.2 watts dissipation in the tube out of 25.

For CW, they run it up to 750 volts and 120 ma!

In real life, you can run the 6146 at 800 volts or more, I run some 4D32's at twice the voltage (1200) then they are supposed to have on them.

Yup, Brett.  My EICO 720 will put that out on AM, especially the one with the solid state rectifiers - easy.  Go with modern design using a 6146 using a PI tank

Al

[N2DTS]

And Hammond still makes transformers, so full wave bridge or center tapped you should be able to get something that will do what you want for a smaller rig, they have chokes also, and a chassis to build it on.

Its nice to have a separate supply for the modulator, but not many rigs had that, only home brew that I know of, or the Eico 720/730 combo.

Its very easy to build a low power AM transmitter using tubes, only the mod transformer is a bit harder to find.

[W4TWA]

Thanks everyone for your response. I'm gonna go with the arrl job.

[W4RFM]

Go for it, if you need help finding parts, ask me.  I love this stuff.  I would much rather see you attempt to build something of your own, than be put off by people who cant walk in the store and come out with everything same brand same spec, etc. Crap, getting the parts is half of the fun. Hammond still makes new transformers for guitar amps and what ever, you would be surprised where you can find good iron.  Everybody slams e bay, but I have built a pretty good stash using that service, and I am tight with the dollar.

[WU2D]

6146 modulated by a pair of 807's or 1625's is a cinch and even if you mess up the circuit, that combo will give you something that can be a starting point to a great signal. Heck you could even mismatch the transformer and crazy connect the 807's into some kind of triode mode and have audio to spare.

[N2DTS]

Building is a lot of fun, and half the fun is making do with what you can get or what you have.
I have got plenty off ebay, you just have to decide what you want to pay.

6L6's, 807's, 1625's, and even 6146's will make good modulators.

[Opcom]

The special circuit needs a lot of grid to grid drive voltage since it is the screens that make the plate current go. I believe it is something like 800V peak screen to screen. Was stated in one of the Coastal handbooks.