Proper 6V6 Bias for 5E3

[scstill]

About a year ago I built a scratch 5E3 from the original Fender schematic.
I used the 250ohm cathode resistor with a pair of “matched tested” RCA 6V6.
It sounds really good at low volume (don't have much opportunity to crank it).

I measured the bias to cool and match it with different tubes lying around.
Ibias = (B1 Voltage at OTcenter tap - Voltage plate)/Resistance center tap to plate
The table below are the values I got. All tubes except the JJs are 50’s vintage bought over time as tested.
Seems the best is case 6 but the tubes are different brand although they are nice looking vintage coin base types
Case 4 runs cool with original design tubes but is somewhat mismatched.
Case 3 is a little warmer with design tubes but still mismatched
Case 5 uses tubes committed to car radio conversion project, first time they were played in guitar amp
Case 7 uses modern JJs (so big) matched within 2ma but these tubes don’t match the vintage look this amp is seeking

Before I commit...
I wanted to see if there is any advice on these measurements and design approach.
What do you think is best?

[sluckey]

Take the two highs and two lows and set them aside. Put the remaining ten tubes in a hat, mix them up, and blindly pull two tubes out of the hat. Put them in the amp, button it up, and build another amp.

[scstill]

Take the two highs and two lows and set them aside. Put the remaining ten tubes in a hat, mix them up, and blindly pull two tubes out of the hat. Put them in the amp, button it up, and build another amp.

There are only 8 total tubes, so if I set aside 4 the blind draw is on on 4 remaining
Do you like the 470 Cathode Resistor with the blind tube draw?

[sluckey]

Oops! I overlooked the fact that you were also changing cathode resistors. And, in fairness, you didn't say you only had 8 tubes. That changes everything.   

So, put all eight tubes in the hat. Blindly pull out two tubes. Plug them in the amp. Use the 250Ω resistor. Button up. Play it like Keith. Honky Tonk Woman want hot bias and don't care about matching.   

[HotBluePlates]

... I built a scratch 5E3 ...

... I wanted to see if there is any advice on these measurements and design approach. ...

You forgot to subtract the voltage from cathode-to-ground, as the tube only "feels" voltage from cathode-to-plate when considering dissipation.

     -  Case 1:  Vplate1 = 384.2, Iplate1 = 0.044A.  384.2v x 0.044A = 16.9w (rather than 17w)
     -  Volts across the 250Ω near 21v?  (384.2v - 21v) x 0.044A = 16w
     -  Dissipations might be less-bad than you think, across the board.


After measuring voltages in vintage Fender amps I've come to find you can't trust the schematic voltages are correct.  But let's look at the 5C3 Deluxe, since there are voltages on the schematic.

     -  18v shown across the 250Ω cathode resistor: 18v / 250Ω = 72mA.

     -  Preamp Current Draw:  (308v screen supply - 280v preamp supply) / (10kΩ dropping resistor) = 2.8mA preamp current

     -  Screen Current Draw:
        (360v plate supply - 308v screen supply) / (10kΩ dropping resistor) = 5.2mA screen + preamp current
        (5.2mA screen + preamp current) - (2.8mA preamp current) = 2.4mA screen current
            -  This screen current (which is for both tubes, or 1.2mA per tube) sounds a little low to me, but let's roll with it.

     -  Total Plate Current = (72mA Cathode Current) - (2.4mA Total Screen Current) = 69.6mA
     -  Plate Current per tube = 69.6mA / 2 = 34.8mA

     -  Plate Dissipation = [(360v Plate) - (18v Cathode)] x (34.8mA Plate Current) = 11.9 watts


What's the big difference between the 5E3 Deluxe and the 5C3 Deluxe?  Well, there's actually a lot of differences, but in the power supply the dropping resistors were changed from 10kΩ/10kΩ to 5kΩ/22kΩ.

Lowering the resistor feeding the screen supply node raises the screen voltage, from 308vdc on the 5C3 schematic to your measured 347vdc.  Higher screen voltage causes higher plate current.  If you simply revery to 10kΩ/10kΩ as in the 5C3, your plate current will come down while still using the 250Ω cathode resistor.

There might be a slight reduction of power output, but chances are you won't notice as the output tubes also become easier to drive.  Translation: the amp sounds as-loud at lower settings of the Volume control.  Most folks these days are buying/building amps then complaining how loud they are & trying to attenuate them.  Chances are you'll still want an attenuator after switching to the 5C3's power supply arrangement.

I think your experiments were good, trying larger cathode resistors to tame plate current.  But if the 5E3 is over-hot it's because Fender jacked up the screen voltage without also raising the cathode resistor value.  Either way, cathode-biasing far into Class AB just doesn't work well, so it's probably reasonable to lower the screen voltage and run the amp Class A.

I've found other places where Fender's "last version" of an amp circuit actually created problems rather than solving them, so the 5E3 isn't necessarily a "better Deluxe tone" than one of the earlier circuits.  I sometimes wonder if Fender monkeyed the 5E3 Deluxe to try to be a little bit louder than the 2x6V6 Harvard, and then dropped the issue with the switch to fixed-bias in the brown Deluxe (while at the same time changing the Harvard circuit to use up tweed Princeton chassis) and having the 6G3 Deluxe take over the role the tweed Vibrolux covered.

[Bieworm]

I once played in a band where even my vibro king wasn't loud enough so I bought an 80W bogner shiva. But my opinion is that when you play in a band where an 18W or 5E3 isn't loud enough..run away!
If you need over 30W in a live situation the crowd don't enjoy the gig like you do. So..a lowish volume like a 5E3 is plenty loud and you've got a bigger chance to keep your eardrums :)

[scstill]

I once played in a band where even my vibro king wasn't loud enough so I bought an 80W bogner shiva. But my opinion is that when you play in a band where an 18W or 5E3 isn't loud enough..run away!
If you need over 30W in a live situation the crowd don't enjoy the gig like you do. So..a lowish volume like a 5E3 is plenty loud and you've got a bigger chance to keep your eardrums :)

you have inspired me....
This 59 Tweed Hi Power Twin will be a future build (watts baby).
Too bad it will never help me sound like JoeB (couple of these used for his stage amp)
https://truevintageguitar.com/blogs/previous-inventory/1959-fender-twin-amp-5f8-a

[scstill]

What's the big difference between the 5E3 Deluxe and the 5C3 Deluxe?  Well, there's actually a lot of differences, but in the power supply the dropping resistors were changed from 10kΩ/10kΩ to 5kΩ/22kΩ.

Lowering the resistor feeding the screen supply node raises the screen voltage, from 308vdc on the 5C3 schematic to your measured 347vdc.  Higher screen voltage causes higher plate current.  If you simply revery to 10kΩ/10kΩ as in the 5C3, your plate current will come down while still using the 250Ω cathode resistor.

Thanks, I was hoping for the great info in your post when I submitted this post. Very helpful and appreciated.

But I cannot find the 5C3 schematic with voltages. Could you provide link?

My next build (for which this experiment was done) will be esentially a 5C3 with a 2nd channel 6sj7 Pentode preamp (like Gibson BR-6F).
Exception is there will be a 6x5 as rectifier.

[HotBluePlates]

But I cannot find the 5C3 schematic with voltages. Could you provide link?

Same link.  The voltages are shown only on the layout portion.

[scstill]

But I cannot find the 5C3 schematic with voltages. Could you provide link?

Same link.  The voltages are shown only on the layout portion.

Thank you, did not see that.
So on the 5C3, how do they get to 85v at the preamp 6SC7 plate yet it is 157v at the PI 6SC7 plate? Both of these come from the same 280v source with the same 250k to drop. Is the plate current that different for the two 6SC7's? although its really small current .49ma vs .78ma

[sluckey]

V1 is grid leak bias with .8V for bias (grid to cathode). V2 is cathode bias with 1.9V for bias (grid to cathode). V1 is conducting much harder, current is higher, therefore plate voltage is lower.