Here are photographs of a 45 tube amplifier built 2 years ago with a GSG board.
http://www.diytube.com/phpBB2/album_pic.php?pic_id=171
http://www.diytube.com/phpBB2/album_pic.php?pic_id=172
http://www.diytube.com/phpBB2/album_pic.php?pic_id=173
The Amplifier uses 2 Hammond 166M2 filament transformers to supply 2.5 VAC to the 45 tubes. The center tap on the secondary winding for each filament transformer connects to pin 5 on J3 or J4 on the GSG board. D1 through D8 are not populated. Likewise, C11 through C14, R19 and R20 are not populated since the Hammond 166M2 transformers supply filament power.
The Amplifier uses 4.3k resistors for R15a, R15b, R15c, R16a, R6b and R16c. The parallel combination of 3 resistors per channel makes a bias resistance of 1.43 K.
The amplifier uses an EDCOR XPWR106 (275-0-275V @ 175ma & 6.3V @ 4A) power transformer. Since there is only one 6.3 V winding it is necessary to put a jumper wire from J3-1 to J4-1 and J3-2 to J4-2 to supply 6.3 volts to the 6SL7 and 6AX5-GT.
A second Hammond 156R choke replaces R17a and R17b. With the EDCOR XPWR106 power transformer, the power supply section produces 300 volts at B+
Plates from the type 45 tubes connect to the 5 K-ohm taps on the James transformers.
The moderate B+ value was selected with hope of extending the life of globe tubes that are over 70 years old. A standby switch was added to energize the filament transformers prior to energizing the EDCOR power transformer. Power supplied to the primary windings of the filament transformers and the EDCOR also passes through MF72-120D15 inrush current limiters.
The 45 data sheet indicates that a cathode to plate potential of 250 volts delivers a plate current of 34 mA with a grid bias of -50 Volts. 34mA * 1.433 k-ohms (cathode resistance) = 48.7 volts. Subtract the 48.7 volts from 300 volts at B+ to leave 251.3 volts of cathode to plate potential. The math works out close enough and the amplifier sounds great.
Keith