40M TX - new project

[VK7ZL]

I have started work on my new 40M AM transmitter.
The power supply and modulator were salvaged from a PYE base station and I stripped the TX section off the chassis. There was no circuit information available so I drew up the one below.
There are a couple of odd things with the mod tranny - the primary windings are not balanced (see information on circuit drawing) and there is an winding that goes nowhere so any input on that would be welcome. The rest I think is fairly straight forward.

The RF deck (yet to be built) will comprise of a 6BH6 or 6AM5 oscillator, crystal locked but with provision to use a sig gen for freq agility, followed by a 7360 driving a pair of 1625's. I haven't yet decided whether to make it dual band, 160M & 40M.

Any comments on the project would be welcomed.

Bob

[Opcom]

Interesting bias and feedback circuit!

[G3UUR]

Bob,

Did you get the modulation transformer from an HF base station or a VHF one?

I wonder what size it is physically. The HF base stations were 60W, but I'm not sure whether that was input or output.

The extra windings were usually used for negative feedback to an earlier stage.

Dave.

[k4kyv]

It looks  like that unused extra bit of winding off the plate of one of the modulators, that is not connected to anything, could have been to allow the transformer to be used as a modulation autotransformer. I have a couple of autotransformers exactly like that, except they don't also have the independent secondary winding as shown in the schematic and used in the circuit.

You could put some a.c. voltage across all or part of any one of the windings, and then measure the ac voltage that appears from the midtap of the primary to each of the plate taps, then measure the ac voltage across those unused extra turns.  Now measure the DC resistance from midtap to the plate tap, and then the DC resistance across the extra turns.  If the proportion of DC resistances to AC voltages is anywhere close to the same, that tells you that the continuation of the winding beyond the plate tap would take the final amplifier plate current and the transformer would work as an autotransformer.  If the resistance of the extra turns is considerably higher in proportion to the extra voltage, compared to the voltage from midtap to plate tap, then it is there for some other purpose, such as a feedback circuit, and wouldn't take the full plate current to the final. In other words, the ohms per a.c. volt as measured should be about the same for the main primary winding as for the extra turns, if it is designed to carry the final amplifier plate current and function as an autotransformer.

What I see missing from the circuit is negative feedback round the pentode section of the 6GW8. It looks like the modulators are wired as zero-bias class-B triodes, with control and screen grids tied together with 22K resistors, similar to the familiar handbook circuit using 807s. The pentode driver stage needs to have plenty of negative feedback to satisfactorily function as a driver stage for that class-B modulator, unless the negative feedback is achieved externally with an R-C network between pins 6 and 14 of JP-1 .

[G3UUR]

Normally. when the CT of the primary is connected to another winding, that is the secondary winding for modulating the PA. However, if the other winding that Bob's used as the secondary is really intended for negative feedback its output is a bit high with a 2:1 step up. This also seems a bit too much for the PA, which would usually require about a 1.0:1.2 step up if running off the same HT voltage.

The primary inductance also seems a bit low for the application, even for Pye.

Bob's probably asleep at the moment because it's the early hours of the morning for him. I suspect we won't get any answers for a while.

Dave.
 

[k4kyv]

I agree that those inductances are way lower than what one would expect with a modulation transformer. Normally, the inductances are 10H or more on both primary and secondary.  Makes me wonder if it really is a modulation transformer.  If so, it certainly isn't designed for much of a low end, and would probably pass little audio below about 1000~. In particular, the primary inductance (@ 487 and 485  millihenries) almost suggests an rf transformer instead of an audio transformer.

[VK7ZL]

Dave,
It was a VHF unit, model TVA 528, on 100+? megs running a 6/40 in the final.

Don,
Thanks for your input. I understand what you are saying regarding measurements on the mod tranny.
The circuit was drawn from the existing layout and it looks like one end of the extra winding was never connected to anything so its purpose in this application is probably only know to the manufacturer.
I don't intend to change or modify the setup unless I have to. My main concern is what effect the imbalance of the primary winding will have on the 5/40's and does it have to be corrected in any way and how?

I have attached some pics of the original layout and the stripped down version. As you can see it has had some surgery done in the field with the tacked on filter caps. I have removed the old cans and waiting for the new caps to arrive. I will then be able to measure and adjust the voltages as required.

[G3UUR]

Bob,

I wouldn't worry too much about the imbalance in the primary resistance because the inductance is pretty close. When you've got everything up and working you might want to experiment to see which way round gives you the best positive modulation peaks with your voice. If you need to swap the modulator connections around to get a 180 degree phase change to make best use of the asymmetry in your voice, it would probably be easier to swap the primary of the driver around rather than swap the plate connections to the modulation transformer if those leads are a bit short.

I don't know the Pye TVA 528. The F27AM is the common one over here, and I'm quite familiar with the modulation transformer from that. Your 528 seems like a high power version of the F27.

Dave.

[WB6NVH]

The way transformers are constructed will nearly always prevent the winding resistance from being equal when comparing sides from a center tap, so I wouldn't worry about it.