Sound City Concord - Bias Too High - Insufficient Negative Voltage

[Grantorino]

Hello EL34 World.

I am restoring a Sound City Concord amplifier from the 1970's. I have replaced many components, including the big filter cans, electrolytic capacitors, diodes, etc. I replaced the 100uf 64v bias cap with 100uf/100v.

I am unable to achieve a cool enough bias on the new matched pair of EL34 power tubes. The bias trim pots are at the most counter-clockwise setting (most negative). Through the limiter, the 1ohm resistors I installed on pin 1 measure the bias at around 34mA - this jumps to 44mA without the limiter, which is too hot. The raw bias node was around 41.4v before replacing the bias cap - it's now around 39v.

I have heard mention of a voltage doubler circuit - I have spare 100uf/100v capacitors and uf5408 diodes - but am unsure whether this mod is compatible with the PT. There is no resistor to ground on the bias rail, as I have seen in certain other Sound City amplifier threads on this site.

I have attached pictures of the power section and bias rail.

Any advice and instruction to lower the bias/negative voltage of this amplifier would be much appreciated. I'm happy to take any measurements that are needed.

Thank you kindly.

[Grantorino]

Here is the schematic for this amplifier - The Sound City Concord.

[AlNewman]

I believe this is the resistor to ground for the bias pots.
Increasing this should increase negative voltage.

[pdf64]

I believe this is the resistor to ground for the bias pots.
Increasing this should increase negative voltage.

I don't see how that could happen, what's the mechanism?

I don't think the schematic can be correct, there's no ground reference for the heaters, no other end for the bias winding.

[AlNewman]

The schematic is hard to read and doesn't have many component values, but I can see by the pictures that there are 2 brown wires from the transformer, one goes to ground, and the other goes to the bias circuit.  I don't see any extra parts in the bias circuit that would point that the schematic from that point is incorrect.

The grid voltage is drawn from the wipers of the pots.

If it's a winding on a transformer, wouldn't the point to ground create a voltage divider and elevate voltage, much like a tail resistor on a PI?

[Grantorino]

I appreciate the input from both of you, AINewman and pdf64.

Regarding the resistor AINewman identified - this is downstream of both the trim pots. I am not sure it acts as the raw bias ground reference.

The bias cap itself is connected to ground.

It seems to be a simple bias design of Rectifier -> Cap -> Pot -> Grids.

Installing a dropping resistor in series between the raw bias and the trim pot has the opposite of the desired effect, making the voltage less negative under load.

I previously attempted a voltage doubler circuit from the brown wire to the bias circuit. It was shorting to ground and again, made the voltage less negative. (I think the AC winding needs to be floating)

Browsing other threads, I found this suggestion by sluckey (different amp model, still Sound City):

I would add a 12.6v transformer. Easiest, no brainer way to do it.

https://el34world.com/Forum/index.php?topic=22852.0

For wiring, he states to:

Mount the transformer near the bias cap. Look at the bias diode. Cut that dark brown wire connected to the banded end of the diode and tape the cut end of the wire. Connect a transformer black lead to the banded end of the diode. Connect the other black lead to chassis ground at the positive end of the bias cap. Tape off the green yellow wire. Connect the two green wires to the existing filament string (pin 2 and 7) of the nearest power tube.

From what I understand, this creates a half-wave bias supply to provide some more negative voltage.

Does anyone have experience with such a mod? Or perhaps some insight as to the suitability for this amplifier?

Thanks again.

[pdf64]

To me, it seems nuts to add a transformer rather than implement a voltage doubler.
Leave that end of winding connected to the chassis common for half wavedoubler, or lift it for full wave doubler. Whichever, it's just 2 diodes and 2 caps.
The trick is to implement it correctly this time https://www.electrical4u.com/voltage-doubler/

Just to state basic good practice, remove valves and use a light bulb limiter until the circuit is proved good.

[Grantorino]

Thank you pdf.

I’m not ashamed to say i’m learning everyday.

I will lift the grounded bias wire and include it in a voltage doubler circuit.

I will report back once installed.

All the best.

[Merlin]

Alternatively you can use a half-wave doubler, which might be easier to implmenet in this scanario, since you only need to lift one end of your bias rectifier diode and insert two extra components before it, leaving the ground intact.

[Merlin]

I found this easier-to-read snippet of the Concord PSU on google images, I've drawn on the exra two components (ignore the big red circle, I didn't draw that)

[pdf64]

... I will lift the grounded bias wire and include it in a voltage doubler circuit. ...

As Merlin shows, it's not essential to lift that wire, you can leave it connected to the chassis if implementing a half wave voltage doubler.

[Grantorino]

I installed a full-wave voltage doubler in the following configuration:


• Brown 2 was lifted from the chassis ground, tied to a new lug on the rail.
• Both capacitors negative to chassis ground, positive of cap 1 to brown 1. Positive of cap 2 to brown 2
• Anode of Diode 1 to brown 1, cathode to raw bias node. Anode of diode to brown 2, cathode to raw bias node.


This resulted in a less negative voltage - from -40 to -8.

I believe a full-wave voltage doubler is incompatible because Brown 1 and Brown 2 are part of one winding, not out-of-phase taps. Or perhaps I wired it incorrectly.

I am now considering a half-wave doubler, as suggested by Merlin. Perhaps you could offer some guidance as to the wiring. I picture it as such:


• I lift the cathode of Diode 1 (originally to Brown 1) as Merlin suggests.
• Connect the cathode of Diode 2 to the cathode of D1
• Connect the negative of Cap 2 to the cathode junction of the diodes, cap 2 positive to the raw bias node
• Cap 1 negative to raw bias node, positive to ground. Brown 2 remains connected to chassis ground.

Thanks again everyone.

[Merlin]

Looks like you wired your full-wave voltage doubler wrong, the two caps should not both be going to ground. This is partly why I recommend a half-wave doubler; beginners always have trouble following the full-wave diagram.

[pdf64]

At least there's been no exploding cap - yet

Seriously, as you're getting things mixed up, that is a risk.

[Grantorino]

I appreciate the insight!

As a beginner, perhaps you could nudge me in the right direction and suggest whether my above description of the half-wave doubler is correct and save me blowing a cap :)

Otherwise, I may attempt the full wave doubler again with the correct cap orientation. Though I do suspect the bias wires are not out-of-phase, but from the same winding. There was continuity between the two browns.

[Merlin]

• I lift the cathode of Diode 1 (originally to Brown 1) as Merlin suggests.
• Connect the ANODE of Diode 2 to the cathode of D1
• Connect the CATHODE of Diode 2 to ground.
• Connect the negative of Cap 2 to the junction of the diodes, cap 2 positive to the transformer winding.
• Cap 1 negative to raw bias node, positive to ground. Brown 2 remains connected to chassis ground.

See correction.

[Merlin]

Though I do suspect the bias wires are not out-of-phase, but from the same winding. There was continuity between the two browns.

That's normal, the transformer winding might have such low resistance that it will beep on continuity test.

[Grantorino]

Praise Merlin! I sit at your feet at learn.

With the half wave voltage doubler, i’m seeing -80v which has allowed me to bias the power tubes at 33.0mA and 33.2mA respectively.

Next i’m going to tackle the low volume, hum and distorted guitar signal that was present before tackling the bias issue.

I appreciate the guidance very much. This is a great forum to learn from!

All the best.

[AlNewman]

Regarding the resistor AINewman identified - this is downstream of both the trim pots. I am not sure it acts as the raw bias ground reference.

The bias cap itself is connected to ground.

Actually, I made a mistake.  Since there is no dropping resistor upstream of the bias pots, there's no way to increase maximum negative voltage.  Meaning all the pots are in series with the resistors, and increasing the tail resistor would only increase the minimum negative voltage you could see adjusting the pots.

That being said, -80V seems high for EL34 grids.  What are you getting for plate voltage?

[Grantorino]

Identifying pin 3 as the plates, i measured each socket at 468V.

Identifying the screens as pin 4, they measured 465V.

Pin 5, identified as the control grid, measured -42 and -41v.