Mainline voltage check [resolved]

[Fredo]

Hi, folks:

I just completed a mainline build. The resistance check on manual page 72 completes ok. The voltage check on page 73 completes with slight variances. I would like to make sure that these variances are acceptable.

Measurements with tubes inserted:

Power transformer output measured at Power Supply board T6 and T7: spec = 135v  my reading = 127v

+275vDC on Power Supply board: spec = 275v  my reading = 280v
+6.3vDC on Power Supply board: spec = 6.3v  my reading = 5v
IA on A side of C4S board: spec = 275v  my reading = 280v
IA on B side of C4S board: spec = 275v  my reading = 280v
Breg on Regulator board (both sides): spec = 220v  my reading = 220v
-reg on Regulator board (both sides): spec = 0v  my reading = 0v
Kreg on Regulator board (both sides): spec = 8-12v  my reading = 9v

Measurements with tubes removed:

+275vDC on Power Supply board: spec = 275v  my reading = 349v
+6.3vDC on Power Supply board: spec = 6.3v  my reading = 6.3v
IA on A side of C4S board: spec = 275v  my reading = 349v
IA on B side of C4S board: spec = 275v  my reading = 349v
Breg on Regulator board (both sides): spec = 220v  my reading = 348v
-reg on Regulator board (both sides): spec = 0v  my reading = 0v
Kreg on Regulator board (both sides): spec = 8-12v  my reading = 1.4v


Notes:

1. Mains reads 125v, which is 5v higher (4.166%) than spec. This might explain the 280v Power Supply reading.

2. Tube filaments cause a heater voltage drop from 6.3v to 5v.

Are these readings within acceptable variances?

Thanks,

Fred

 

[Paul Birkeland]

When you are measuring 5V coming out of the regulator, to what range do you have your meter set?

That voltage should literally be exactly 6.3V, but could show up as something else if you're on the 1000V scale for instance. 

Your measurements with the tubes removed showing 6.3V on the output makes me nearly certain that you need to resolder every solder joint on that big power supply PC board. In particular, those 1N5820 diodes take a lot of heat to solder properly.  Put your soldering iron tip on a joint, add a little bit of solder, then count to ten.  What you're looking to see is the solder suck down into the joint.  It's very easy to flow some solder into the joint, but without enough heat, the flux from the solder will actually form a barrier against the diode lead (the last Mainline I repaired had this problem).

When you have no load, there is enough capacity to provide proper output from the regulator, but with a load, there is too much resistance to provide adequate voltage.

-PB

 

[Fredo]

Thanks, Paul for your excellent advice. I resoldered everything on the power supply board. I had originally soldered for 6 seconds at 650 degrees F. I resoldered for a full 10 seconds at 700 degrees. I watched the solder flow and the flux boil up and out of the joints.

I'm using Cardas Quad Eutectic Silver Solder (proprietary blend of silver, copper, tin & lead), melting point 364 degrees F. It seems to flow as well as standard solder, but take longer to get the flux out of the joint.

New voltage measurements are as follows:

With tubes removed, the +6.3v on the power supply board measures 6.27v. With tubes inserted, it measures 6.24v. This is a .03v voltage drop under load; meter scale is 20v.

With tubes inserted, the +275v on the power supply board measures 287v, as do IA and IB on the C4S boards.

Bregs still measure 220v and Kregs still measure 9v. Biasing is set at 145v.

So the question is: are these readings adequate or do I need to attain precisely +6.3v and +275v?

I could resolder again, but I am concerned about damaging parts and the board.

Thanks very much for your help,

Fred

 

[Paul Birkeland]

You're on the money now.  6.24V is more than close enough to 6.3V, and is within the tolerance of the resistors used to set the voltage, but certainly a huge improvement over 5V.  The 287V of raw B+ will vary with the incoming power line voltage, so there's not much to worry about there.

-PB

 

[Fredo]

Great!

What fun!

Thanks for your help,

Fred