Tube Amps / Music Electronics
For current discussions, please visit Music Electronics Forum.

ampage archive

Vintage threads from the first ten years

Search for:  Mode:  

Tube voltage limits and tube regulators

1/26/2006 9:15 AM
Hiatt Collins
Tube voltage limits and tube regulators
I've been looking at tube regulators, specifically a series-pass design.  
Say I wanted to use a tube regulator to drop a B+ of 475V down to 250V for a screen supply, and wanted to use a tube like a 6AQ5 or EL84 to do it. I've got the plate and screen tied together and sitting at 475V, and the cathode and heater both sitting at around 250V.  
I've only got 225V across the tube, which would be well within either tube's rating if I had the cathode and filament at 0V and the plate/screen at 225V.  
I'm guessing that it's the voltage across the tube that matters (like in a cathode-bias arrangement), not the absolute voltage. Is this the case? Does the tube 'know' that it's sitting at a voltage higher than its ratings, even though the voltage across it is well within them?  
And now, a word from our sponsors:

1/26/2006 9:44 AM
Ray Ivers
Since you're floating the heaters too, the tube is oblivious to ground references, and can float at much higher voltages than you're using. I've got schematics of floating HV regulators using SS components rated at < 50 volts, that surely work just fine.  
FWIW, I've read that series-pass tube regulators tend to drift somewhat (probably a heat thing) as compared to SS stuff that stays rock-steady, but that may not be an issue with your circuit. Also, the 6336, 6080, and 6BX7 dual-power-triode tubes make excellent pass tubes, but they're octal-socket units.  
1/26/2006 10:04 AM
Hiatt Collins

Thanks, Ray. That's what I was hoping for.  
I've been looking at the regulators for use with a 6AV5/6AU5 output section. These tubes, as you probably know, have a listed 550V plate voltage limit, but a listed 175V screen voltage limit. I'm looking for a steady-ish voltage, but I'm just as interested in the voltage-dropping aspect of the regulator. The gibson GA-85 has something akin to what I'm considering (with appropriate differences due to filament requirements and greater voltage drop in my application).  
There are other ways that I could drop the voltage, but I've never used a tube regulator in an amp. I'm interested to see how it'll turn out.  
1/26/2006 11:01 AM

Check this out, it is an adjustable supply:  
notice how the pass tubes (parallel triode mode 6Y6's) only see the difference between the output voltage versus B+. You have gotten the similarity to a cathode bias output stage, the plate voltage rating is generally the maximum difference allowed between plate & cathode, but there are exceptions like when both plate & cathode get to really high voltages, you need to be careful about the other nearby pins and socket terminals as well as other conductors or chassis components within sparking range. The higher the voltage, the greater distance it will jump to initiate an arc.  
Also, there are two separate isolated working 6.3V heater windings, *X*, & *Y*. So the maximum heater-cathode voltage rating of the 6Y6's and 6SJ7 are not exceeded.  
When the cathode of a tube is so many volts away from B- (there is no ground, the output is completely floating except for the one banana jack marked *Ground* which is connected only to the chassis and the 3rd-prong of the line cord, so no earth ground unless you connect one intentionally), you do not want to use a heater winding which is referenced to B- or the heater-cathode voltage rating of the tube could be exceeded. When more than one tube are like this, and their cathodes are not always very near the same potential, then beyond a certain point you need a separate heater winding for each tube.  
The center-tapped 6.3V winding *Z* is not used in this circuit, it is only available externally as a bench heater supply, this winding goes straight to the 3 banana jack terminals on the right of the drawing.  
The pots & resistors in the 0A3 string and the load string are fairly high-current, IIRC 10 or 20W Ohmite wirewound, the *DC Output Voltage* has a big user knob having a fairly precise pointer, this linear pot tracks beautifully with the silk-screened scale on the control panel which goes from 200 to 325VDC for adjustment range over the full rotation of the pot.  
The *Screen Voltage Adjust* for the 6SJ7 and the *Range Adjust* are screwdriver pots, only need to adjust them when new tubes have very different bias needs compared to the one(s) being replaced. I'm using the most microphonic 6SJ7 in the world and its just fine for this application ;)  
The regulated output has a 100milliamp rating. When you are not drawing from this output, the excess current passes through the 0A3 tube instead, so it glows brightest at idle.  
As a regulated screen supply for a guitar amp circuit, when your amp starts to draw screen current, the 0A3 then dims in time with the music.  
The caps are all instrumentation-quality paper-in-oil, the 4uF are each the size of a cigarette pack, the choke is bigger than a Marshall choke.  
The whole unit is almost the size of a 50W Marshall. It rocks in its own way :)  
But it will fry screens if you have the screen voltage set too high while the B+ is coming from a conventional guitar amp supply which sags below the screen voltage for too long at a time, mainly during overdrive conditions. In an ordinary guitar amp the screen voltage is derived from the B+ so it sags more or less in sync with the plate. Not so with such a high-capacity independent regulated screen supply, screens stay steady while everything else fluctuates, if the screens glow bright enough for a long enough time, they will snap.  
Probably a good idea to get started using 1.5x to 2x the normal ohms recommended for screen stoppers for the type and brand of power amp tubes you are using.  
1/30/2006 1:35 PM
Hiatt Collins

That regulated PS would make a *serious* screen supply! That schem has some good info related to my questions, so thanks for posting it.  
I'm aware of the heater-cathode voltage limitations, so I'd be using a small (6.3V@ .5 to 1A, for example) separate filament transformer for the regulator tube, properly referenced to a voltage that would keep the tube comfortable.  
As I said, I'm not that concerned about having a rock-solid regulated voltage for the screen. I'm mostly interested in the voltage drop available, with the regulation as a bonus. I'd be running the screens at about 1/2 B+, so there's probably a very slim chance that the plates would ever get close to that low.  
1/30/2006 2:28 PM
Rob Mercure
Why not a simple gas discharge regulator or two such as the VR-150/OD3 or OA2 types - you get that nice purple glow, you can parallel several of them if you need more current, and the wiring is pretty simple. You can also do the same thing with stacked large neon bulbs and get a wild panel light out of it.  
1/31/2006 8:25 AM
Hiatt Collins

Hi Rob,  
I could, and it probably would be simpler. I've never built an amp with a tube pass regulator, so I thought I'd give it a whirl with this one.  
I know that you can put the gas regulators in series to get a higher voltage drop, but I didn't know that you could parallel them for more current handling. Do you have to use resistors to isolate them? It's my understanding that you can't parallel zeners without some sort of resistors, and I think of these gas discharge tubes as being the tube equivalent of a zener.  
Back to the regulator - If I used the 0A2 or similar, I'd probably put them in series with the screen supply to drop, say, 250-300V, rather than build a shunt regulator. I'm more concerned about keeping the screen voltage for the 6AU5/6AV5 around its limits rather than really regulating the voltage as such.  

  Page 1 of 2 Next> Last Page>>