Tube Amps / Music Electronics
|For current discussions, please visit Music Electronics Forum.||The sunn still shines online!|
|2/13/1999 7:25 PM|
||conversion to BF Twin bias questions|
I'm working on blackfacing an SF Twin here (1975, 100W, master volume and boost removed) that is doing some weird stuff in the PI/bias supply. The amp, prior to modification, was totally stock. I replaced all the electrolytics, all the screen/grid resistors (1% carbon films), and have changed things around in the phase inverter section. The standard stuff has been done (plates changed from 2 x 47k to 82k/100k, 330k to 1meg, 68k to 220k, etc.) The resistors used here are 1% metal films.
Here's where things are a little odd: in changing the bias supply to BF specs (level instead of balance), going with the BF specs (27k resistor on pot, 470 ohm from bias supply, changed diode to 1n5048 however) the current through 2 tubes, at maximum bias, is only 13ma. No problem, decreased bias resistor to 8.2k, which got things into the 30ma range, but now I'm getting two different bias readings; using the output transformer shunt method, the right side reads 48ma (24ma/tube) when the left reads 61ma (30.5ma/tube). Suspecting it might be the tubes, I swapped in a quad that I knew for sure were matched within 1ma; same thing.
Question is, I figure this is a sympton of why Fender went with the bias balance in the first place, but where is the imbalance coming from? Is it a quirk in the output transformer, or an imbalance in the PI section (which would necessitate a change in the 220k bias resistors?)
Any insight appreciated.
|2/13/1999 8:50 PM|
You don't have a PI balance problem, unless the coupling caps are leaking badly.
You DID rewire the balance to level adj on the bias pot, right? (never assume anything...)
|2/14/1999 6:45 AM|
Yup, the bias pot has been rewire to a level adjustment, using the three lugs on one side.
I'm starting to think it's a problem with the transformer... any ideas?
|2/14/1999 10:59 PM|
Measure the bias voltage on each grid.
Make sure they're the same.
|Page 1 of 1|