Is the voltage supposed to drop across the bias splitting resistors?

[Iplayloud]

I have never paid much attention before to actual bias voltage because I use the 1ohm resistor method. I just decided to probe around and realized I got -48v before the split and -38v after.I thought bias voltage was static and not being spent. Why the drop then?

[Leo_Gnardo]

I have never paid much attention before to actual bias voltage because I use the 1ohm resistor method. I just decided to probe around and realized I got -48v before the split and -38v after.I thought bias voltage was static and not being spent. Why the drop then?

Most of the current is being 'wasted' in the voltage divider that allows you to select/adjust the bias voltage. Not to worry about this waste, it's just a milliamp or two or three depending on the resistor values in the divider. Very little current goes to grids as long as your output tube(s) are working properly, mere microamps.

[g1]

You mean the split for the 2 halves (push and pull)? 10V does seem significant, which circuit are you looking at?

[J M Fahey]

No and no.

What he means is that he measures -48V before the "bias splitting resistors" which are the typically 220K ones feeding the power tube grids, and he measures -38V "after" , meaning closer to the grids.

Which is normal, he's creating a new voltage divider with those 220K and the 1M multimeter internal resistance to ground.

What I asked sibo to do in his amp for some 100 posts.
I was interested in knowing whether there actually was bias voltage to begin with, and its value; Stan wanted to check whether it actually reached the tube grid pins, not caring that much about the voltage measurement precision but checking path integrity.

Oh well.
2 sides of the same coin.

[Iplayloud]

What he means is that he measures -48V before the "bias splitting resistors" which are the typically 220K ones feeding the power tube grids, and he measures -38V "after" , meaning closer to the grids.

Which is normal, he's creating a new voltage divider with those 220K and the 1M multimeter internal resistance to ground.

Exactly, the amp is a Fender style clone built in '71. Like a princeton with the half-tail PI but with a pair of 6L6. I don't see the interest of doing that instead of using a full blown PI and 6v6 as both cases yield 20 something watts.. except the half-tail stays clean all the way up? Anyhow, it sounds pretty good. So indeed I measured directly across the 220k splitting resistors and there is 0.0v
Case closed

[g1]

Which is normal, he's creating a new voltage divider with those 220K and the 1M multimeter internal resistance to ground.

Thanks JM, that makes sense. If memory serves, I've never measured such a big difference across those resistors, but I've never used a meter with a low input resistance like that either (my meter is 10meg input). I will double check next time I'm in some similar amp.
I have run across the issue measuring the gate voltage of mute FET's, not sure what kind of meter I was using then.

[Enzo]

Never really noticed it on 6L6 grids, but the LTP grids are about impossible to measure directly because of meter impedance.