807 cathode follower driver for 100TH 810 or 833A

[W2PFY]

Has anyone built the tube version of the GATES cathode follower or perhaps some other brand? Question (2) Do you know of a PC board source for same. If not, no big problem because I know how to build simple tube stuff.

I think the original version used 6BG6 tubes but I have a crap load of 807's so I would rather go that direction. I looked on line and didn't come up with anything.

I want to build two such circuit drivers over the winter. One for the desk KW and the other for the BC-610D. I never heard of anyone running such a driver for either rig?

If you happen to have a Gates driver excess to your needs, contact me on that. Then I would only need to build one.

Thanks Terry

[W7TFO]

That circuit wasn't the best performer, audio-wise.  

It needed a boatload of global feedback to make it sound good, and ate a lot of 6BG6 or 807 tubes as well.

I wouldn't waste my time on it.

Sorry.

73DG

[WBear2GCR]

Terry,

Shoot me the sketch-a-matic for the Gates design...

I've got a bunch of methods for cathode drive that work well...

Dennis, not sure why the Gates thing would have eaten up tubes?? Curious.

                           _-_-bear

[W7TFO]

Driving a pair of 833A modulators in class B as found in the BC-1G/H is a tall order for a pair of 807's, even as cathode followers.

In practice, the PC boards were crap and the tubes ran at max dissipation mounted horizontally so failures were common.

Better done via a good piece of iron and a pair of 845's or 828's a la the BC-1E/F or BTA-1K/L.

I found this to be true from many years of babysitting both the Gates (cost cut build) and RCA versions of these 1kW transmitters.  I'd get frequent calls from stations 'cause the 'sound' just wasn't there again.

The RCA was built better, but the proof of both CF designs was such that they sounded OK, they just didn't stay that way for long.  But then again, 807's were plentiful & cheap...and in a ham rig, most seem not to give a shit how it sounds.

73DG

[WBear2GCR]

Ok, without seeing the schematic, it's hard to know why there were any problems.

In general, and from strictly a technical viewpoint a CF drive, DC coupled will have a far wider
bandwidth than a transformer coupled arrangement. So, it's unclear to me why it did not sound
as good as it ought to have.

As far as their choice of tube, the mounting and the like, that's not a disqualification for using a CF drive
in an amateur situation, one does not need to follow their (Gates/RCA) practices, one ought to be able to
get excellent results with the same basic circuit.

Using a 6550 or some other tubes (depends a bit on the voltage swing required) maybe triode strapped sweep tubes that will have
a higher power rating might obviate the drive limitation of the 807. But thinking out loud, and without
looking at the drive requirements, I'm guessing that the limitation wasn't the power rating of the 807, but
the drive voltage swing requirement, which (guessing) put the plate voltage up at or over 700vdc on the 807?

Anyhow, there are a range of very good CF drive circuits used for hifi amplifiers, and they do sound very good.
So it ought to be no problem to do the same thing for a ham application.

                          _-_-

[W2PFY]

Hello Mr WA1GFZ, I tried to answer your personal message on this topic, but it says you have blocked me

[W2PFY]

Thanks to all that have replied. I know of two home built drivers that were built in this area. One was with 807's and the was the 6BG6 style. I don't know that in either case if there were any improvements to the design? I actually used that 807 version in the rig that had two 833's in it and it sounded great. The same can be said about the 6BG6 version. I however did not have any lab equipment hooked up to my receiver nor did I get any pro audio reports on these two systems. I have found that from my own experience that voice may sound great on a given system but music could sound very nasty on the very same system. I forgot what type of waveform is a real test of an amplifiers limits? Maybe it was a two tone?

Bear, perhaps we can talk you into designing a good tube type cathode follower? I have a couple designs I can send to you. One would be the original type most of are used to and the other still uses the same concept using 807 tubes but is using 100 Henry chokes in the plates of the voltage amplifier and even larger 100 Henry chokes in the cathode portion on the circuit. That is what is used in my MW-2 and is the same one that Dennis will remember that he had send me in a swap deal. I could use that unit but it is the size of an old table top TV set.

TY-PFY    

[Steve - K4HX]

833s need something like 500 volts g/g audio voltage.

[W2PFY]

833s need something like 500 volts g/g audio voltage.

Yes at 3000 volts for 1650 watts output the specs call for 400 volts G to G. Now if you need 2700 watts the G to G is 510 volts at 4 KV on the plates.

If we only needed 500 watts audio, the G to G voltage would be much lower. I run the
 Westinghouse 833's at 3200 volts and I have no problem pulling one amp plate current. Those 807's in that rig run at 400 volts on the plates.   

[WBear2GCR]

I've seen 807s run at 700vdc on the plates, so I'd expect they could make that swing in class B2... but as a follower, you'd need
to swing that voltage out of a gain stage! Therein lies the rub, I expect... still would like to see that Gates schematic.

I seem to think I downloaded it once, but did not find it thus far on this computer.

                                   _-_-

[W3GMS]

I know your taking about a tube driver, but here is a scheme Steve-QIX came up with years ago.   It will drive 833's without any issues.

http://www.classeradio.com/tube_driver.htm

Joe-GMS

[DMOD]

Here is one such circuit using 807's from a BC1:

Phil - AC0OB

[W7TFO]

Yeah, but as my late friend Charlie Kittleson always said, "Use a transistor, go to jail"

73DG

[Opcom]

one of these may help.

[WBear2GCR]

Here is one such circuit using 807's from a BC1:

Phil - AC0OB

Says "cathode drive..." on the print, but I don't see any cathode driving anything...

oops, there are two stages of 807 there, just to trick the unwary... heh.

[W9BHI]

My Gates BC-1G runs that tube driver and it has superb audio, easily makes 130% positive
modulation.
All of the resistors and caps have been replaced with film types.
The 807's are a quad matched set.
In voice service, it should last forever.
Don W9BHI

[Opcom]

The nicest thing about the Gates is using the 807 as a high voltage amplifier. This is a wonderful application example not often seen with beam tubes.

If a Gates board is found that is not too burned out, truthfully it should be cool-able to avoid (more) burning, if blown on from the underside by a small fan.

It seems like the only electrical question about using the original circuit may be current capacity. The peak current of 807 tubes is a lot more than what is given as plate current in the tube manual. Somewhere there is a formula but looking at the curves compare the plate current when 807 bias hits zero, to the grid current peak in the same part of the operating curve for the 833.

So, look at the 807, in audio service, can use 200mA per pair 'DC', and the 833 looks like it needs 250mA on peaks. On peak the 200mA average (Ec=+15; 350mA peak plate current?) of audio service is supported by one 807.

[KL7OF]

The Gates TX also has an extra winding on the mod transformer to modulate those  audio drivers .....That has to help a circuit like this.....

[W9BHI]

Here is a schematic for a solid state audio driver that Gates used in their BC-1H.
I have one of these connected (cobbled) to my BC-1G right now and it works great.
I need to get it physically mounted in place where the old 807 driver board was mounted.

Don W9BHI

[DMOD]

...It seems like the only electrical question about using the original circuit may be current capacity. The peak current of 807 tubes is a lot more than what is given as plate current in the tube manual. Somewhere there is a formula but looking at the curves compare the plate current when 807 bias hits zero, to the grid current peak in the same part of the operating curve for the 833.

So, look at the 807, in audio service, can use 200mA per pair 'DC', and the 833 looks like it needs 250mA on peaks. On peak the 200mA average (Ec=+15; 350mA peak plate current?) of audio service is supported by one 807.

I take a slightly different view of the specs and the constant current curves of the last 807 cathode follower stage.

First, the cathode follower stage has a high input impedance and a low output impedance. The input impedance is primarily limited by the grid resistance value. Here, in terms of audio response and  the grid resistance value, the - 3dB down point is ~ 20Hz.

Secondly, the cathode current capability is the sum of control grid, screen grid, and plate current.

My interpretation of the 833 grid drive requirement for ClassB audio is that it needs ~ 200 mA max. Adding up the 807 circuit's current values, the cathode current capability for this stage is > = 250mA.

Overall, this is a good drive circuit. The overall frequency limits are primarily determined by the 600 input transformer.

Phil - AC0oB

[DMOD]

Here is one such circuit using 807's from a BC1:

Phil - AC0OB

Has anyone measured the grid bias on pin 3 of XV3 or XV4 to get the voltages measured?


Phil - AC0OB