Power supply question

[bluefoxicy]

Hi guys, I'm looking at the power supply stage in the Valve Junior:

http://www.sewatt.com/files/sewatt/StockVJ_1-3.pdf

I have two questions.

1) It looks like R10 and R11 act as a voltage divider. Why not use a bigger R10 and no R11? This appears wasteful (dump some input voltage to ground). Would it put too much stress on the PT?

2) HTF does adding a 220Ohm resistor at R10 produce a 14V drop at B1, 32V at B2, and 15V at B3? I can appreciate the voltages but something is wrong with this circuit. Only the power supply changed between V1 and V2.

[stokes]

R 11 is a "bleeder" resistor,it bleeds the B+ when you power off the amp.The drop at B2 is across the 4.7k not the 220 ohm.B3's drop is across the 10k.The voltage drops at these points is also affected by the current draw that each cap is feeding.The screen on the power tube has more current than the current of the preamp tube,that is why the 4.7k seems to drop more volts than the 10k.The power supply is not "isolated" so you have to figure everything in the circuit into the draws.

[Enzo]

Right, R11 is a bleeder resistor - it discharges the filter when you turn the amp off. It draws 1ma from the power transformer - that won't stress anything. Ohm's Law will tell you that.

R11 and R10 do not make a voltage divider. A voltage divider starts with a voltage cource and two resistors in series across it. Then the voltage at their junction is a portion of the total voltage by their ratio. But here, there is only on resistor across the supply, R11. R10 is just in series with the load.

Stokes, scroll down to version 2 of his schematic. The 14v drop he sees is from the added 220 ohms in the input pi filter stage. SO the plate supply is 354v in version1 and 340v in version 2.

Again, this is simple Ohm's Law. You added 220 ohms in series with the whole circuit. 14 volts across 220 ohms means 64ma of current is flowing. By adding R10 and C9, you have a second stage of filtration, making the input a pi filter instead of a simple filter. This does reduce the voltage a little, but more important it will reduce the ripple a lot, and that is very important on a single ended amp.

[bluefoxicy]

Ah, I thought a voltage divider was two parallel circuits with different resistors to direct some voltage in one direction and some in another. The stuff at R6 and R7 is often referenced as a voltage divider by modders (to cut the signal in half).

Normally I have 100uF for C6 there (I don't use this amp off the shelf). Sometimes 200uF. Trust me I know about ripple, it sucks stock here; 22uF is not enough.

[Don Moose]

The stuff at R6 and R7 is often referenced as a voltage divider by modders (to cut the signal in half).

R6/R7 is a voltage divider - the voltage goes in at the C1/R6 node and comes out at the R6/R7/VR1 node. So effectively, the two resistors in the divider are R6 and the parallel combination of R7 and VR1.

Vout = Vin*(R7||VR1)/((R7||VR1) + R6) = Vin*(1M||1M)/((1M||1M)+1M)
Vout = Vin*(500k/1.5M) = Vin/3

... not worrying about the pot.

Back to the voltage divider question - in the other schematic, the voltage goes in at the Bridge/R11/C6/R12 node ... and the bottom resistor would be effective DC resistance of the rest of the preamp.

Hope this helps!

[Enzo]

And in fact VR1 is all by itself a voltage divider - a variable voltage divider. Nothing says one voltage divider circuit can't have another within itself.

Your idea of parallel resistors branching would not be a voltage divider, but it could be thought of as a current divider. I've never heard the term used, but why not.

You don't really send voltage down a wire. You apply voltage to one end of it, and current flows down the wire. You can't separate voltage and current, but they are not the same thing.

And to further nit pick, I would reserve the term parallel for where two things are connected together at both ends, not merely one end. They may be drawn facing parallel on the paper, but parallel has a specific meaning electrically. So branching circuits are just that, branching, but though the resistors be side by side and have a common end connection, they would not be parallel.

[bluefoxicy]

And to further nit pick, I would reserve the term parallel for where two things are connected together at both ends, not merely one end. They may be drawn facing parallel on the paper, but parallel has a specific meaning electrically.

right, they branch but they're not parallel. They both reach ground eventually but

[Don Moose]

Your idea of parallel resistors branching would not be a voltage divider, but it could be thought of as a current divider. I've never heard the term used, but why not.

I last heard the term in school twenty-some years ago, and suddenly remembered it late last night after my last post in this thread.

You don't hear it much because it doesn't really have much practical application as a design or debug concept.

[Enzo]

Cool.

I learned electronics in the 1950s, it was all tubes, transistors were just a novelty, and everything was voltage. We didn't deal in currents much then, other than power supply.