Adding 5879 to SE - help with low voltages

[Jonas]

I'm attempting to add 5879 to a Single Ended amp using similar values to the Gibson GA-40 (CH 2) set up. I have 328V at the node but only 32V at the plate and 41V at the screen?

Do I need to adjust the power supply or am I mixing the single back in incorrectly?

I've tried a couple different 5879's with same voltages.

[shooter]

but only 32V at the plate

this is the "V1" position?  (I'm not a pentode knower so that might be low)??


If so
the "goal" of V1 is to bring the input milli-volt signal to a useable Volt or 2, keeping the "noise floor" as low as possible.  as long as it's "center biased" enough to swing clean ~5VAC I never worry much about vdc at plate.
I use AU7's typically, they hover around 80vdc. 

[PRR]

> only 32V at the plate and 41V at the screen?

Is that actually bad?

Only 3 of your 8 examples have voltage. The ratios of resistors and the tube's ratio of Ia to Is set the plate voltage. For this tube, ratio Rs/Ra of 6 or 8 make roughly center bias. Ratio like 3 makes low plate voltage. You are trying ratio near 2. No wonder your plate is laying low.

[Jonas]

Thanks PRR - I changed the values to #6; much improvement. Can you explain center bias?

[shooter]

since sleep evades me here's my take;


you're trying to bias the tube so that you can swing the input AC signal, clean in both directions, amplifying the AC without distorting.
the Cathode VDC, "controls" the negative 1/2, while the plate, the positive half.
get the cathode to close to ground, it starts clipping
get the plate to close to B+ it starts compressing

[tubeswell]

Centre-bias is the middle of the load line with respect to where the (signal) grid voltage curves cross the load line. To provide a hypothetical example for a typical preamp tube, take curve Vg0 as the left hand end of where these curves cross the loadline and (say) curve Vg-4 as the right hand end of where the curves cross the loadline. Center-bias would be Vg-2. I.e., the signal grid voltage can swing 2V either side of the idle point (Vg-2) without being clipped. (And, if your aim is a clean signal, the tube should be set up so that) Centre-bias also (simultaneously) happens to provide the point at which the plate voltage can swing either side of the tube’s plate voltage idle point (I.e., 141V in example #6) without being clipped.

[Jonas]

thanks tubeswell, explanation makes sense I'm looking at a few different TDS for the 5879 and trying to follow along and identity the load lines.

I see the RCA TDS recommends grounding pins 2 and 6 in all applications to shield grid 1 and plate from heater when using AC heaters. Then RCA TDS calls out N/C on these pins on the pinout diagram. My 5879 is not RCA brand so I'm leaving these pins ungrounded

[tubeswell]

thanks tubeswell, explanation makes sense I'm looking at a few different TDS for the 5879 and trying to follow along and identity the load lines.

I see the RCA TDS recommends grounding pins 2 and 6 in all applications to shield grid 1 and plate from heater when using AC heaters. Then RCA TDS calls out N/C on these pins on the pinout diagram. My 5879 is not RCA brand so I'm leaving these pins ungrounded

Even if 2 and 6 are NC inside the tube, grounding the socket pins can still help reduce AC induction between pins. So feel free to experiment with it

[PRR]

> Then RCA TDS calls out N/C on these pins

A shield is normally connected only one end, thus IS "NC" by pin-out logic.

I'd ground them.