807 Modulator-65W output

[Xnke]

Working on the audio portion of my new amateur radio transmitter, it's basically a 65W class AB1 audio amplifier.

That said-this is not a hi-fi design! Once the basic concept is established as not immediately going to melt into slag (the circuit presented here does not and actually works OK at the 45 watt power levels I've tested at) I need to limit the frequency response. I don't have a big enough 10K ohm resistor to properly load the secondary winding beyond 45W for more than a few seconds at a time.



Now, even though the desired frequency response is 100hz-10Khz, I've made little attempt at limiting the bandwidth so far. I am very open to suggestions on:

How to best limit the bandwith, restricting it to 100hz-10Khz range
How to best apply some negative feedback around the whole amplifier. I've never been particularly good at making this decision.

Now, for the details:

The 807's are running 600V anode and 300V on the screens, and -32v fixed bias on the grids. They can provide a maximum of 72W of power to the modulation transformer primary, and operate in class AB1...lots of class B but still AB1. Plate dissipation is not being violated at the idle conditions, and short of a key-down single tone test for tens of minutes (not realistic) they'll be fine.

The modulation transformer will provide between 8.4K to 9.1K plate-to-plate load for the 807's. It was actually designed for a military transmitter that used 807's as the modulator tubes. The operating point was taken from the original schematic and backed off a bit.

The 6BL8 input circuit is a very nice low-distortion design lifted from a 50W RCA audio amplifier, but it was designed for a 25v output swing. I have not gone through and checked the distortion at the 32 volt output level nor made allowances for the increased drive needed.

The modulation transformer is as follows:

807's drive a 8.4K centertapped primary winding

The secondary winding is a 10K winding designed to carry 130mA of standing RF current. I don't have a 10K resistor to load this winding with beyond 45W output right now.

The tertiary winding is a 1000 ohm centertapped winding that provides modulated screen supply voltage from the centertap to one leg, for the RF final. The other half of this winding could be used as a negative feedback source, but it will be biased at +250 to 300v, with an audio voltage swing of 206 volts peak-to-peak.

Thanks any suggestions!

[Circlotron]

Maybe also post the question in https://www.diyaudio.com/forums/tubes-valves/

Edit  -> I see you already did! LOL.

[Xnke]

Yep, it's also crossposted to an AM Radio forum as well. My old teacher is still on this forum...but he's not old, and is an internet friend from back in high school. If there's a boneheaded mistake in there, I'm sure I'll hear about it. The old forum I would normally have posted this on is down due to software problems, and has been for a while.

So far I think it'll work alright, there's just a few loose ends that I haven't tied up/can't remember because it's been a LONG time.

First suggestion I've had has been to put some kind of current limiting in place in case one of the bias potentiometers goes open circuit-I'm thinking a 1 ohm resistor in the 600V line would allow me to trip a plate current relay if the current goes over, say, 350mA? That's more current than they should draw at full power, but not a lot more than they should draw at full power. It would take a definite fault to get to 350mA.

[Theboel]

just curious why not using solidstate amp ?

[Circlotron]

just curious why not using solidstate amp ?

Similar reason to using amateur radio instead of a telephone.

[Theboel]

Ah Ok,

the glow of tubes never can be replace 

[TimFox]

Where did you obtain the modulation transformer?

[Xnke]

This transformer was on Ebay-I have another just like it I got at a hamfest but it was so scratched up it was hard to tell what it was rated at.

Modulation transformers are kinda like gemstones. You usually find them laying on the ground, but you don't just walk past one without picking it up!



YES, that does say 100hz-15Khz +/-0.5dB