bias set for aleph

Helo everybody!
Help me please

this are the readings of voltage acros source resistors in my aleph 2 shown for one chanal. I`d like to ask some one who build this amp (GRolins perhaps or anybody else) for fast check of values. I`m geting 2,6A on positiv chan. and 2,7A on output devices in total. I`m litle low but what is concerning me is not equal reading on resistors. I`d like to know if I`m in the resonable limits of recomanded values. If I`m not or even if I`m what is the way to corect it (R19 ?). Voltages acros R14(392ohm) and acros colect. and emitor of MPSA18 are normal. Voltage acros R11 (source resis. on Q3-221ohm) is a litle to high 6V (4-5 recom.). DC is 43,8 + 43,8 -. Oh, one more thing regarding source resistors. On the pos. chanal side the resistor conecting main board and fet board (source to R20) have the largest value of six fets, on the negativ chanal side there is resistor in the same position (conecting to the main bord) but with the lowest value of six uotput fets (normal?). Othervise amp is working only few ours but without any problems clean and stady. Heat sinks (2 piec. 200x300x40mm 10mm base plate, per chanal) alowing me to hold my hand on them for 5-6 seconds. Are the aleph`s phase depandent?

Values of source resist. voltages
+chanal 0,43 0,41 0,46 0,43 0,43 0,40

-chanal 0,48 0,42 0,48 0,39 0,48 0,45

R11 (221ohm by Q3) 6V
R14 (392ohm) 4,2V
colect - emit MPSA18 4,12V

I`ll aprishiate any ansver - thanks!
rgds Marjan
 
Turns out that the bias is sensitive to the temperature of the outputs. When I first cranked up a single channel of the Aleph 2 on the water-cooled system, the outputs were stone cold...and the bias was quite low. About 1.8A, as I recall. I drove myself nuts trying to figure out what I'd done wrong. Couldn't find anything. Ran the bias up using R19, and all was well. Then I brought the second channel online and the water temperature increased a bit. The bias for the first channel popped up to about 10% over the nominal value. I sat back and thought it over for a bit, then ran the second channel's bias up to match the first, and have been running them overbiased ever since.
Nelson's production versions ran quite hot, and the resistor values (R19, etc.) balanced the circuit quite nicely, I'm sure. However, I wanted to run at a lower temperature both for esthetic reasons and because I'm using TO-220 devices (less wattage capability than TO-3s or TO-247s), so I went with the water-cooled thing. Everyone will end up with slightly different bias values, depending on how they go about their heatsinking.
Yes, the sound does change as the bias goes up--more is better--but watch your temperature.
Some bias variation from device to device is normal. The tighter you match the devices, the less the bias will vary.

Grey
 
Hi!
N.P. & G. thanks!
Yesterdaj I made the mesurment of the heatsinks and I`m runing at 61°C. I supouse thet could be good enough. Mesurment is made without top cover plate (300x200mm 10mm thick aluminium) which should decrise the temp. for degre or two. After 10 ours of inrun and conecting my Electrocompaniet EC4.6 preamp to an aleph with balanced mode instade of SE RCA, the sound become what I expected from such of design. I think it shoud be beter by the time go by. I`m using Magnetostatic`s SMG-c (4 ohm)so my only consern and question is corelation betwen bias seting and impendance stability in aleph design. Magnepans are not the easeest burdon to care.

"Suma sumarum" I`ll leve it the chanel one on inruning an finish the second block. After that I`ll see if any actions needed.

Thanks again!
rgds Marijan

P.S
I`m using IRFP240 ST (Moroco - same lot) in TO-93 package - similar to to-247 but with the 45°degre cut of on the top side of FET, not insulated on the mounting side with plastic like 247 - I used grease+mica+grease