problem with dc filament supply

[spy]

I'm repairing a Matamp King Street Major and it has a problem with the dc filament power supply. The first two preamp tubes are feed with dc. Classic psu with bridge, el. caps and a 7805 regulator. I don't know what happened in the past but the trimmer that sets the exact 6.3V is missing! And I suppose a resistor, too but I'm not sure. When I got the amp, there were 5V on the filaments!
Does anyone have the schematic? I don't know the trimmer's value or if something else is missing. The circuit is a little bit strange and I can't understand what's going on. It's not the classic circuit with a voltage divider and one resistor at the ground pin of 7805. I'm not good with drawings but if anyone is interested I can try to describe it.

[nickb]

I'm not familiar with this amp but it could be that the regulator is wired as a constant current source. See page 23 of the 7805 datasheet http://www.fairchildsemi.com/ds/LM/LM7805.pdf.

[spy]

I'll try to describe it!

At the input of 7805 there is an el.cap after a bridge as a rectifier, at the output of 7805 another el. cap, negative side of both caps are connected with the ground pin of 7805. All these negative (-) sides, bridge, el.caps & regulator are not at the ground of the rest amplifier. One side of trimmer is at the output pin of 7805, the other side at the ground pin of the regulator and the centre tap is at the main ground of the amplifier!

[nickb]

It allows you to adjust the DC offset between the ground and the heaters.

[spy]

Have you seen anything similar? Can you adjust the output voltage to be 6.3V?

[Amp Kat]

I'll try to describe it!

At the input of 7805 there is an el.cap after a bridge as a rectifier, at the output of 7805 another el. cap, negative side of both caps are connected with the ground pin of 7805. All these negative (-) sides, bridge, el.caps & regulator are not at the ground of the rest amplifier. One side of trimmer is at the output pin of 7805, the other side at the ground pin of the regulator and the centre tap is at the main ground of the amplifier!

Most regulator circuits have some additional filtering circuitry and sometimes a resistor or pot to vary the voltage but in the case of the 7805 the most you can get out of that regulator is 5 volts so whether that is what the heaters are using is not all bad as it's a safe range for a preamp tube but the concern is current draw because the 7805 is only capable of max 1 amp and should be heat sinked so it depends on how many tubes. However you will never get 6.3 volts DC out of a 7805 without a voltage doubler circuit so obviously they must be running at 5 VDC.

[J M Fahey]

Connect the LM7805 ground leg to ground through 2 1N4002 or 1N4148 diodes in series (forward biased).
Instant LM7806.3 !!!

[spy]

That's how I ended! With 3x 1N4007 and I have 5.8Vdc with tubes, without tubes in place I have 6.8Vdc. But I'd like to know what Matamp had in mind!?

[nickb]

Sure - it's in that data sheet see fig 12. R1=100 1 watt ohm R2= 25 ohm 1/3 watt, (better test first!)
JM Fahey's way is better, I think.

[spy]

Matamp's is not like figure 12! R2 is missing and the trimmer is at the place of R1, trimmer's centre tap is at the ground.

[Steve Conner]

I guess this contraption is running off the existing 6.3V heater winding. There's not enough headroom to get 6.3V DC out of the regulator, so they settled for 5V. The trimmer on the output is probably some sort of horrible "hum balance" for the remaining AC heater tubes.

What is the problem with the amp exactly? I think the DC heater part of it was working as designed, until you broke it with the diodes.

[spy]

The AC heater tubes are totally separate from the dc heater supply! The amp did have some other problems but I saw this one, too. As it seems, someone took off the trimmer over there and I'm trying to find out why. Also, I believe 5V are too low for the heaters.

[Steve Conner]

If you want 6.3V, you probably need to re-engineer the circuit with Schottky diodes and a low-dropout regulator. Or get rid of the regulator altogether.

The fact that the voltage drops to 5.8V under load means that the regulator is dropping out, so it's no longer effective at getting rid of ripple.

[spy]

You're probably right as at the input of the regulator I have around 7,5V. What if I'll leave it as it is? The amp is quiet right now.

It's not difficult to get rid of it but will it be better?

[Amp Kat]

You're probably right as at the input of the regulator I have around 7,5V. What if I'll leave it as it is? The amp is quiet right now.

It's not difficult to get rid of it but will it be better?

It will probably run quieter at 5 volts than at 6.3 because there isn't as much thermionic emissions but the tube will run fine and actually help it's longevity in the process.