For two bucks you could do this and reduce the key-up transformer drain, the heat on the zeners and still have regulated low voltage if you want. It's just a poor man's series regulator. The NTE157 would handle a voltage difference from input to output up to 300 Volts, 20 Watt dissipation and more current than your transformer can deliver. Of course the NTE157 has to be bolted down with its insulator for heat sinking.Of course it could be just a needless complication.

Curtis Eickerman

Hi Dave and Norm, First of all the schematic for the power supply is on page
two of this thread not page one. So don't pay any attention to the voltages.
And there are two 1N4007 in series for each leg of the bridge rectifier.
The power supply is not connected to the transmitter at all yet, the only
thing connected are the two 8h 125ma chokes. In other words there is no load at all.
I have three leads feeding the Power supply board, the two high voltage leads and
the center tap. Coming out are three leads, B+ 1 and B+ 2 and the common which
connects all the ground connections. As for the zenier diodes, they are from the
old power supply. They did check out ok. For B+ 1 could I put in series two 100uf
450V caps..That would increase the voltage rateing, right?
I don't understand why the zeniers would start to smoke when there is no load
on them at all. The set up worked great before I changed the transformer.
Guess I'll have to start over...
Thanks guys for your patience,
Gary

Thanks Curtis, I like that idea.
Gary

'No load' is exactly when they would have a problem because there's nothing else pulling current across the dropping resistor so all of it, to bring it down to 300V, is going through the zeners, plus the new transformer providing more voltage to drop.

For a series pass regulator I'd suggest something like a STP4NK60Z MOSFET since, with a 600V rating, gate protection zeners, and reverse diode built in, it as close to bullet proof as solid state can get at these voltages for $1.27 (mouser). I use them for both regulators and cap multipliers.

I vote to forget any regulation on the 6AG7 osc. Period!! What problem with that osc does it solve? What's their mission?? With no load/key-up, all of the B+ 2 current goes through the zener string only limited by the 5K series R. As Flipperhome noted, the zeners are not specified in free air. When you really look into their specs, they are pretty lame in free air having a relatively small die. The last piece de resistance is that they have a positive temperature coefficient (tempco) explaining why the three measure 310 Vdc at key-up. Hams have been building unregulated power supplies for their 6AG7 osc for decades. In the one 6AG7/6L6 xmtr that I built in '94 that used two VR tubes, the only thing I got out of it was a lot less drive out of the 6AG7 osc to the 6L6. I haven't repeated that "mistake" since!!!
Dave - WA6VVL

I would guess he's convened about excessive plate voltage on the 6AG7 since I imagine the CT is putting out around 450V.

The RCA HB-3 6AG7 datasheet shows plate voltages to 475-500 Vdc. The Heath AT-1 6AG7 osc was at 400+ Vdc.. Not unusual at all. The key-up voltage in my last 6AG7 xmtr, circa '08, was 484 Vdc, with nary a problem among the dozen or so 6AG7s tested in the xmtr (reference Electric Radio, Jul.'08). Key-down voltage was 436 Vdc @ 120 Vac line. About par for the course...
Dave - WA6VVL

They may very well work at 450V but the design center max is 300V.

The plate curves 'going to' 500V is not saying that's allowable idle plate voltage and choke loaded it's going to swing both ways, you know.

One of the interesting things about tubes is they don't go 'pop' like a transistor does and what the effect is of 'too much' often remains a mystery.

Well, there's not a ham xmtr that I can think of over the years that has run their 6AG7 at 300 Vdc. The 6AG7 is my go-to xtal osc. In all the years I've used the 6AG7 as the primary osc or driver via a VFO, I have NEVER lost a 6AG7. That's 50+ yrs of experience dating back to my high school days. The HB-3 average plate characteristics for the 6AG7 go out to 475-500 Vdc. It makes no sense if the tube had an absolute max of 300 Vdc to provide design data out to 500 Vdc. Hard to believe that RCA would provide the design data if engineers weren't allowed to go there!!!! My experience shows that the 300 Vdc max is way too low. So, I'll stick with my experience as far as the 6AG7 goes...
Dave - WA6VVL

Just some follow-up thoughts... For years, design engineers have taken advantage of undocumented device parameters. Tubes are no exception, and amateurs, by their very nature, have always "pushed the envelope". Someone learned very early that you could RELIABLY operate the 6AG7 at higher plate voltages than the supposed 300 Vdc max. The rest is history. There have been many very successful amateur xmtrs that ran well over that supposed 300 Vdc max: The EF Johnson Adventurer (650 Vdc key-up), the wannabe Knight T-50 clone of the Adventurer (720 Vdc key-up), and the ever popular WRL Chief90 (520 Vdc key-up). So, after decades, at least six of them, it has been pretty much demonstrated time-after-time that the max 6AG7 plate voltage easily exceeds their stated 300 Vdc max. So I have zero heartburn recommending that Gary abandon a regulation scheme for his 6AG7 osc and connect it directly to B+ 2 sans zeners (or VR tubes).
Dave - WA6VVL

I didn't say anything about "absolute max" nor did I say it was a 'problem'. All I said is the datasheet explicitly says 300V design center max and that's probably what he's worried about.

That and looking at the plate curves to see what the 'highest voltage' on the left is isn't a plate volts rating and virtually all tubes show plate curves past the design rating, The 6L6GC, rated 500V design maximum (not design center max) goes out to 700V and the 12AX7 rated 330 V design center max goes to 500V too.

Tube ratings are also based on 'what they're for', the 'intended application', and are not necessarily the same when used for another purpose, not to mention HAMs are notorious for running things over the ratings. Intermittent operation is also less stressful than continuous.

Ok guys, I am a little bit confused, The B+ 2 is coming off the center tap of the
power transformer, which is 450vac. If I do away with the zeniers I won't have
DC going to the 6AG7, right? Don't I need some some kind of rectification?
And Curtis, your diagram using the NTE157 shows the input of up to 500VDC.
Where do I get the DC from from the CT leg of the power transformer?
Sorry I seem to be missing something here..I see what you guys mean about
no load on the PS board. I just didn't want to mount the Xfmr, hook up everything
and maybe do some real damage. What about them 100uf 450V caps? in the
B+ 1 circuit? will that voltage drop off that much under load?

Gary

Guys, has anyone suggested going to a bridge, and using the high voltage center for
1/2 B+? I haven't been following this thread, but dropping the High Level B+ is the
last thing I'd do. I'd also prefer using resistors, since they will compensate for excess
screen or plate current. IMHO. I'd also stick with choke input, to limit the voltage rise
under lighter loads.

Pete

Hi Gary,
The zeners don't rectify, they just serve to regulate the B+ 2 better when the value of that 5K series R is correctly chosen. I'm not sure that 5K is the optimum value. Concerning removing the zeners, remove them and the 5K. Connect a resistor, say 25K 10W, across the 10ufd 450V cap as a load for the 8H choke and connect the 6AG7 circuit directly to the 10 ufd 450V cap. The rectification is taken care of by your 1N4007 bridge.
Dave - WA6VVL

on the voltages of the 6ag7 in xtal oscillator service, look here: http://postimage.org/image/jty6yczr5/

Also, whatever voltage one uses for the 6ag7 it should be, imo, regulated. You can parallel tube regulators for larger current handling, why can't you do the same for zeners ?

Peter

You can't parallel zeners as their knees are different...

That is roughly 450 VDC right after the 8 H choke at the top of the cap just before the series resistor going to your present Zeners.

The regulator just replaces the present series resistor and the zeners. The other way of looking at it is the NTE157 collector connects to the choke output, the base connects to the junction of the series resistor and zeners, and the 270 VDC output now comes off of the emitter.

The center tap ends up being DC because of the action of the bridge rectifier across the whole secondary.

Again, use it or don't, whichever you feel like.

Curtis Eickerman

You shouldn't be paralleling VR tubes either and for the same reason as zeners.

OK Curtis, I'm with you now...I think I have that Power transistor.
Now, What about the 900V for B+ 1? When the Xmtr is in the spot position
SW2 Them two 100uf 450v caps start to snap..After the SW2 is closed the
voltage drops down to about 825VDC. Any ideas?
Gary

With the transmitter and power supply completely unpowered and the power supply caps discharged, what is the measured resistance to ground at the B+ line?

Unfortunately I can't see the schematic until tonight, but I suspect you can insert a little resistance between the positive output of the rectifier and the first filter capacitor and get that voltage under the limits with just a little extra loss in the key down state but at the same time protecting the capacitors from breakdown.

Curtis Eickerman