strange bias problem

[Firestorm]

Using 1k screen resistors won't affect sound much, will it?

Higher screen resistor values tend to make the amp a little more touch sensitive. If you make them very large, they can reduce high frequency response a little, but I don't think you'd be able to hear the difference between 470R and 1K.

[skyboltone]

Ok, thanks!

Using 1k screen resistors won't affect sound much, will it?

No

[Tdale]

What about using 1.5k resistors? That's the only ones I got that are 5W.
Will that work, or is that too far off the ideal value?

Tommy

[skyboltone]

What about using 1.5k resistors? That's the only ones I got that are 5W.
Will that work, or is that too far off the ideal value?

Tommy

Put them in there and see. You will not hurt anything and I doubt you'll tell any difference tonally. The five watt thing is up for discussion too. We haven't argued about that around here for some time. I'm all for starting a fresh fight.

The RCA tube manual tells us that a pair of 6L6GT tubes in AB1 at idle, 450 volts on the plates, will draw 116ma of plate current and 5.6ma of screen current. But you aren't operating in AB1. You are much closer to AB2 or even straight class B with about 78ma combined plate and screen current. So, just to set parameters at something reasonable for this fight lets say you are biased at 450 volts to 37.5ma plate and 1.5ma screen current. So using Mr ohms law and figuring that our choke has 50ohms of dc resistance we'll combine the choke and your original 470 screen resistors to make 520 ohms difference between plate and screen supplies. So let's solve for voltage drop across the resistor first. E=IXR. Or .0015 X 520 = .78 volts. Just to keep the argument clean and correct we need to subtract out the choke voltage drop. Using ratio/proportion we'll multiply .78 volts by .0961538 to arrive at .075volts dropped in the choke and .705 volts in the screen resistor. This is not perfect because we can't use R/P for a precise figure. We need to figure the choke drop then the screen resistor drop but it's close enough. Ok so we have .705 volts dropped at the screen resistor.

This agrees with measurement doesn't it. Notice how at idle the screen and plate voltages appear identical? P=IxE or .0015(current through the resistor) x .705 (volts dropped by the resistor) = .001 watts. So far we're ok with an ordinary 1/4 watt resistor right? Now let's put some signal to the amp and work everything again.

We don't have a set of values for class B so we'll use the AB2 numbers to see what's happening. We could use the worst case AB1 numbers but let's be consistent. The AB2 chart says we draw 205ma on the plate and 16ma on the screens at 47 watts out. I'm gonna round up that screen current just for grins and call it 20ma. First resistor E. E=IxR or .02x470=9.4 volts. Quite a bit more eh? P=IxE or .020x9.4= .188 watts. So this calculation shows we can get by with a 1/2 watt screen resistor in this circuit. But wait! If I leave it there, I will be termed scurrilous at best and asinine or worse at worst. Let's remember that it is really hot under there. Resistors don't like it hot. Especially the old carbon comp that Fender and Ampeg used back in the day. They used two watt resistors which look for all the world like today's five watters. They're big. Those carbon comps drift high as they age and are exposed to heat, so the voltage drop and resistor I will rise with time. And they will fail. You should see that the implication of these calcs is that the higher the screen resistor vlue the higher its power rating should be, but it never gets THAT high. Work it out.

Just as sure as tomorrows another day some folks will chime in here that they've seen screen resistors go up in smoke. Ok. What else was going on I wonder? Probably oscillation Anyway. I use 2 or 3 watt metal oxide power resistors and I think BobI uses 2 watters. No smoke yet.

So Tommy. Play with a variety of values and wattages and then when you've made your decision on what you like the best, solder in some 3 watt metal oxides for piece of mind and go about your business. I maintain that screen resistors should be of sufficient value to drop some voltage at idle and at full roar.

Dan

[Tdale]

Thanks alot!

I got some metal oxide 1k, but I'm not sure if they're 2 or 3 watt. The 5W I got are the white ceramic ones, so they should really hold up well!

I'm selling it tomorrow, and I want to make sure that the screen grids won't go bad, so I wanted to take out the 470 and replace them with a higher value.

I'f you're saying that 1.5k won't be dangerous or hurt the amp technically in any way, I might go for that, just to be sure..

Tommy

[skyboltone]

Thanks alot!

I got some metal oxide 1k, but I'm not sure if they're 2 or 3 watt. The 5W I got are the white ceramic ones, so they should really hold up well!

I'm selling it tomorrow, and I want to make sure that the screen grids won't go bad, so I wanted to take out the 470 and replace them with a higher value.

I'f you're saying that 1.5k won't be dangerous or hurt the amp technically in any way, I might go for that, just to be sure..

Tommy

Should work fine.