Champ screen grid voltage (rel to plate)

[Groover]

I reduced the resistor between the 1st and 2nd B+ nodes from 10k to 3k and added a 470 ohm screen resistor. I don't hear all that much difference, (although it is hard to tell when one has to disassemble, solder, then reassemble - time to buy some "decade" boxes I guess)

[mooreamps]

I would say no. If the screen grid burns up and the screen is left floating, the tube usually redplates itself to death. The screen resistor limits the screen current when the valve is being overdriven, and prevents the screen from melting. Whatever happens you don't want that resistor to fail. (I suppose you could parallel it with a much larger resistor, like 100k, as a fail safe)

1. Then why don't you try it ? Why don't you just disconnect the screen voltage from a power tube, and then just see if your tube will "usually" redplate itself ?


2. Screen current is not a function of the tube going into saturation. Screen current flows when the wires on the screen grid become in the perfect shadow of the control grid. This can certainly happen when the tube is still hot, and it gets a couple good bumps, like when the amp is being packed away after a long gig.... This is why I teach others to look at a vacuum tube as a mechanical device, and not an electronic device. You get yourself some really well made power tubes, and they will barely conduct any screen current ; at any drive level.


3. Bruce is right on when he writes ; "if" the screen draws current. I would be listening to him when he writes these comments.


-g

[Wilder Amplification]

1. Then why don't you try it ? Why don't you just disconnect the screen voltage from a power tube, and then just see if your tube will "usually" redplate itself ?


2. Screen current is not a function of the tube going into saturation. Screen current flows when the wires on the screen grid become in the perfect shadow of the control grid. This can certainly happen when the tube is still hot, and it gets a couple good bumps, like when the amp is being packed away after a long gig.... This is why I teach others to look at a vacuum tube as a mechanical device, and not an electronic device. You get yourself some really well made power tubes, and they will barely conduct any screen current ; at any drive level.


3. Bruce is right on when he writes ; "if" the screen draws current. I would be listening to him when he writes these comments.


-g

I also disagree with "the valve will redplate itself to death if the screen resistor blows". I see the way of thinking that could support it, but due to what I've seen in my experience of having a blown screen resistor the tube will stop passing current altogether when this happens. I too have also seen amps with a screen kill standby and in those designs I don't ever recall seeing the screen being grounded when switched into standby mode.

Also, I disagree with the screen grid/control grid geometry having anything to do with causing a screen current increase. It's been a known fact that guitar amps are hard on screen grids to begin with. It's the nature of having a single rail supply in which the screens are only a few volts behind the plate. Ever play a cranked amp and as you play you see a red glow inside of the plate structure that gets brighter when you play? That's screen glow...a nice indicator of screen current increasing.

On a single rail supply, you have a plate and a screen that both have a positive voltage on them with little separation between the two positive voltages. When you hit the tube with signal, plate voltage drops as plate current rises. For the duration of the signal swing in which the plate voltage is below screen voltage (screen voltage only drops SOME due to the screen resistor voltage drop but not nearly as much as plate voltage is dropping) your screen current will increase because the positive voltage on the screen is higher than the plate voltage for most of the swing duration.

Now...take an amplifier with a dual rail supply where the screens are biased at 1/2 the plate voltage. Much less screen current because of the fact that the screen voltage is much lower than the plate voltage. As such, more current flows to the element with the higher positive voltage, and less of it gets taken by the screen grid. Due to the fact that the plate voltage is so much higher than the screen voltage in this design, the screen voltage is only above plate voltage for a much shorter duration of the swing.

Screen resistors are there to limit screen current. As a byproduct, they also cause the screen voltage to drop as screen current increases because as Bruce stated, resistance to current flow creates a voltage drop across said resistance, just as resistance to water flow creates pressure. If you think about it, it's almost as if we have an Ultra Linear design but with DC negative feedback in that as screen current rises, the screen voltage drops due to the screen resistor, which helps to reduce screen current.

[Steve Conner]

I also disagree with "the valve will redplate itself to death if the screen resistor blows". I see the way of thinking that could support it, but due to what I've seen in my experience of having a blown screen resistor the tube will stop passing current altogether when this happens.

I +1 on this.

I guess technically if the screen grid floats, it assumes whatever potential exists at that point in space inside the valve, and the valve doesn't stop working completely, rather it acts as a crappy triode.

But I expect this potential would be low, and the valve would hardly conduct at all, since in normal operation it needs screen current to hold the screen voltage up and persuade the electrons to come out and strike the plate.

Some valves have negative screen current under certain conditions, and then all bets are off. These ones could indeed run away, with the screen voltage floating above its normal value. But the only one I can think of is the 4CX250B, hardly popular in guitar amps.

[Wilder Amplification]

I +1 on this.

I guess technically if the screen grid floats, it assumes whatever potential exists at that point in space inside the valve, and the valve doesn't stop working completely, rather it acts as a crappy triode.

Exactly...it doesn't not conduct anything, but with no voltage on the screen it would definitely exhibit a super high plate resistance, in which all bets are off for plate current INCREASING to the point of redplating.

[Groover]

Would some of you knowledgeable folks please comment on my experience with this 5F2a build?

I am wondering about the effect of changing the 1st B+ string resistor, thereby raising the screen voltage, and especially wondering why I don't hear much difference when going from 10 k to 3 k in that spot.

[MWJB]

Because there probably isn't "that much" difference, neither screens nor preamps draw large currents, even though the numbers in ohms look big, the actual voltage drop may not reflect this...preamp voltage will change slightly but in either case will most likely be within "normal" perameters, you might have 30 odd more volts on the screen now (what difference do you measure?)...enough to make "a difference" but not enough to change the character of the amp significantly.

Also as you noted earlier, downtime during the part change meant time that you weren't listening to the amp, thus the "old" tone was not fresh in your mind when it was "overwritten" with the new tone...it takes time to train your ears to evaluate subtle changes in tone. If you don't really know what to expect, or what you are losing/gaining, then if the amp sounded OK before, it'll still sound OK now.

If you really want to AB test the change, mount the 10K & the 3K (or other test value)on a SPDT switch so that you can flick between them instantly. Be sure not to screw up the layout if trying this.

To significantly revoice the amp you will need to do more than change one power supply resistor to make a 20-30volt difference. Is revoicing the amp what you are after? If so, what are you looking for?

[Groover]

I am looking for more clean headroom in the power amp section... at least as an experiment. I am intending this amp to be a low volume crunch--grind machine, so if I don't like the cleaner PA I might go back, but I would like to try it.

I may be trying to get an apple to sound like an orange a little bit, because my basis of comparison is the SF Vibro-Champ, which sounds much cleaner due to TBM tone stack loss for one BIG reason, I am guessing.

[MWJB]

What's your B+? SF champs can be well over 400vdc.

If your B+ is <370vdc let's say, a 5V4 might help?

[Groover]

With 10kR in filter string B+ was 395V at 1st node. Didn't measure it at 2nd node (doh!), 346 at 3rd node (to preamp plate resistors).

I neglected to take measurements after making the change. I'll get back to you.

[ThermionicScott]

It takes a decent power increase for an amp to be audibly louder -- it's the way our ears work. I'm guessing you gained a watt or two by raising the screen voltage.

My 5F2A puts out about 3 clean watts with a 5Y3, and it doesn't seem much cleaner or louder when putting out 4.5 watts with a 5AR4 rectifier.

But the 'scope tells me it's making more power.

- Scott

[Groover]

I wasn't expecting to hear a loudness difference as much as I was expecting to hear less distortion.

[ThermionicScott]

I wasn't expecting to hear a loudness difference as much as I was expecting to hear less distortion.

They're facets of the same thing -- with more unclipped power, it should sound cleaner at the same volume.

[Wilder Amplification]

They're facets of the same thing -- with more unclipped power, it should sound cleaner at the same volume.

Care to elaborate?

[ThermionicScott]

Care to elaborate?

No problem. What isn't making sense?