Runaway Bias Voltage Kit MK III

G'day!

So, I've bought a set of MK III kits, minus chassis, built my own spacious chassis, had them powdercoated, and I have just finished building the first amp.

So, it comes to Biasing time.

When I've first plugged it in and switched on, there's no bias voltage, whatsoever...
I soon realised this was simply because the 3-point post was glued to the chassis, which was powdercoated.
I Soldiered in a separate ground lead from the ground post to the closest ground point.

Now I have the complete opposite problem. The amp is switched on, and bias voltage slowly builds up, until 5 seconds in, it's to 2 volts.
I immediatley Googled the issue, finally finding a thread on this forum from a few years back.
One of the first posts said to pull all the tubes out and check that there's -40 to -65 volts on pins 5 & 6 on the power sockets.
As the thread goes on, the guy having the issue says he has that voltage at those pins.
Long story short, I have no current whatsoever at pins 5 & 6.
I have re-set every joint in and around the bias circuit.

Could one or more of the Bias supply components been damaged / blown when I had no ground?

Cheers,
Nick

When transferring a particular design to a new chassis or layout it can be real easy to forget something. What I did to avoid as many problems as possible is to draw it out on paper first. Think of it as a puzzle and make sure all of the pieces that are supposed to be connected are connected. The one area where you have a little room for improvement and changes is all of your ground connections. Go over your schematic and make sure all of your ground connections are secure. If possible, create a star ground where most if not all of your ground connections come together. Ideally this should be close to the power supply caps.

Also, take a break. Get away from it for a couple of days to refresh your mind. A little time away will make many mistakes pop out and obvious. Trust me when I say I've had plenty of Duh! moments. And of course be careful!

I'm sure it could be a lot of things, but I just went through the same problem with one of the MK IIIs I recently rebuilt. After a lot of tube switching and voltage testing I found a bad solder joint on the small ground wire between my new SDS power supply board and the common speaker post where it was grounded. This was a very tight hard to get to area and I didn't make a good enough connection to the board. As soon as I re-soldered it, the bias held steady.

I would go through every solder connection with a fine tooth comb and test each connection by pulling on the wire as mine looked good but wasn't.

I had a problem similar to that when I rebuilt my ST-35. I had intentionally isolated the input ground from chassis, which I shouldn't have. This caused a 'float' as it was explained to me and allowed the bias voltage to rise. Grounded the input ground to chassis and problem resolved. Sounds like it could be the same issue with your powder coated chassis.

Bob Latino wrote:You must have a diode as part of the Mark III's bias system replacing the selenium rectifier.

1. Check the AC voltage at the point where the RED/BLACK wire connects to the diode > You should have about 55 volts AC to chassis ground

2. Check the NEGATIVE DC voltage on the other side of the diode to chassis ground. You should have about -70 to -75 volts DC

3. Check the NEGATIVE DC voltage on the OTHER side of the 1000 ohm main bias resistor to chassis ground. You should have maybe -70 to -75 volts DC

4. Check the NEGATIVE DC voltage from eyelet 2 on the driver board to chassis ground and see if you get any DC voltage at all.

Somewhere along the line in this chain from 1 to 4 - you have lost your bias voltage ..

Bob

Bob - Either side of the Diode were measured as OK.

It's only until I got to the negative DC voltage on the other side of 1K resistor that I found I was in trouble...
Every manual and pictorial on the net states that it's a 1K resistor.
However, my manual clearly states that it's a 4700 ohm resistor, and there is no voltage whatsoever on the other side of it. (Lug 3 of 3 lug terminal)

Could this possibly be causing a runaway bias?

There is also a measly voltage of -0.01 V DC on eyelet 2 of the driver board.

Cheers,
Nick

Hi Nick,
I would suggest checking the grounding of the (3) lug terminal strip to your chassis. The center lug
# (2) must be grounded to the chassis. Since your chassis is powder coated...did you remove the coating where this terminal strip is mounted to the chassis insure a good electrical ground?

In the absence of this ground....you will not be able to set the bias.

I'm assuming your having the same problem with both amplifiers?

Kevin @ Dynakit

Bob - Will do that first thing tomorrow, I'll update the thread when I do. Thanks!!!

Kevin - I have hardwired a ground lead from lug 2 to the nearest ground point, and I haven't started on the second amp yet.

Thanks very much to everyone, I'm getting closer!

-Nick

Did you install the diode for the bias the correct way to get a negative voltage. Did you use the red/black wire going to the bias and the red/yellow wire going to ground? Did you wire the bias electrolytic capacitors with the positive lead going to ground? Can you post a clear picture of the bottom of the chassis.

Sal

Update; First amp is working fine!

Finished the second amp just now, got to Biasing, only to find that the pin voltage when the pot was turned to it's lowest extremity was 2.9V.
I've worked out that it's the 4.7K 1/2 watt resistor which I had to replace, because the one in the first amp was blown. (In your case 1K 1/2 watt.)
The negative voltage drops across it - from -77V to -66V, or thereabouts...

I've tried 3 of them so far, all with the same result. I've got another 5, They're metal film, and about half the size of the ones that came with the kits.

Is this just another case of Cheap Chinese Syndrome???

Cheers
-Nick

Metal film resistors act like small fuses when too much current is applied. Carbon composite tolerate overages much better.

pmarcin wrote:Metal film resistors act like small fuses when too much current is applied.  Carbon composite tolerate overages much better.

So why is it reducing the voltage, and not dead-heading it?

Update; I have sourced, and installed a 4K7 1/2 watt carbon composite resistor, and it makes no difference...

I've re-checked everything in the Bias supply circuit... Could it possibly be a bad pot?

(Another) Update; Had a play around with all connections, and swapped out JJ GZ34 for a Genalex I literally got in the mail about 5 minutes ago...

Also had a poke around again with the Soldering Iron... Re-set a few joints... She works!!!

Once again, thanks to all in this thread, and all on this forum...

Will definatley be poking around here a bit more often!!!

-Nick

So, looks like I've spoken too soon...

The second amp is about half the volume of the first...
I've tried swapping all tubes over between the two, I've tried swapping sources, and speakers...

It takes twice as long to start producing sound as the first when I initially switch it on.

Any ideas?
I'm thinking caps...

Also, would this be caused by the voltage drop across the 4K7 resistor I've swapped out a million times???

Cheers,

-Nick

Yeah, I had to splice extra lengths on most of the cables, and I've been meaning to have a tidy up under both of them...

What can I say??? I'm a Machinist, not an electrician...

I don't think the forum is letting me post photo's as my account is less than a week old...

Going to strip the second amp and rebuild it tonight...

Wish me luck!!!

-Nick

Nick,
Send me the photos & I will take a look at them and also try to post these here on the forum. Send to: dynakitparts@aol.com

Kevin @ Dynakit

Just finished rebuilding the 2nd amp with all new resistors.

Still nothing... I had this thing apart down as far as the 240V mains circuit, and it's still nowhere near the volume of the 1st amp...

I'm thinking Output transformer???

_Nick

Nick,
Since you have one MK III working correctly....I would use this as your reference to compare voltage readings to the suspect amp. Please be aware that you are working with high
(lethal) voltages. You can also measure & compare resistances at various points in the circuitry. This should be done with the power "unplugged" and the can capacitor fully discharged. If your have a capacitance meter, I would check all the on board capacitors (C1) through (C6).

Refer to Page (9) of your MK III assembly manual for the Voltage Check Points.

I still await photos of your wiring of this amplifier. Top & bottom views.....Send these to: dynakitparts@aol.com

Regards,

Kevin @ Dynakit

So, due to a few family issues, I have only just got back into these amps...

Swapped the output transformers, which made no difference at all.
Today, I have swapped out the driver boards.

The quiet driver board in the loud amp resulted in yet another runaway bias, untill the tubes started to buzz quite loudly at about 1.8 volts.

The loud driver board in the quiet amp resulted in no tubes lighting up at all.

I have triple checked the wiring to the driver boards with no fault.

I'm not going to bother putting a multi-meter across anything, because I'm going to strip it down, yet again, and rebuild both from scratch, side by side, simultaneously.

3rd times the charm.

I'd recommend rebuilding the problem one to match the one that works rather than tearing them both apart.