2A3 Monoblocks

[Cressy Snr]

OH BLOODY HELL

Just had a horrible thought.

Normally with fixed bias we have unregulated HT so the bias usually tracks the changing anode voltage with the music signal, so that the actual difference remains constant, meaning the operating point does not move around.

If the HT is clamped with regulator, then the DC op point is going to be all over the place unless the bias supply is regulated too.


It ain't gonna work in its present form is it, not without a negative reg on the bias supply

Ah well back to autobias it is.

FOOL!

[Nick]

Not sure I follow the last post Steve. I agree, that if you have something like a 6c33c then regulated bias and regulated B+ are vital, but with a low gm valve like the 2a3 it should be fine. I have not paid attention to this thread as mush as I should (no time), but if you have the 2a3 grid on the other side of the coupling cap from the driver, and a 1M grid leak like the examples show, then you will just charge the cap up. Is there a circuit of what you have done somewhere?

[Cressy Snr]

OK here is what I had

It is back as cathode bias until I get my head around what is causing these probs.
It is a shame as fixed bias seems cleaner in the high treble.



Steve

[ed]

maybe a cap where the 10k is in the bias supply?

[Nick]

I can see why it blocks, but not why it takes 15 minutes to return. The grid has something like 50k to ground. Once the grid current starts it will run away and lock up, and the fixed bias doesn't help as there is nothing to try and limit cathode current.

[Cressy Snr]

I can see why it blocks, but not why it takes 15 minutes to return. The grid has something like 50k to ground. Once the grid current starts it will run away and lock up, and the fixed bias doesn't help as there is nothing to try and limit cathode current.

Sorry about the 15 min myth. It was just the time I left it to cool before starting it up again. It comes back much faster than that.

The data sheet gives a max grid cct resistance of 50K I thought that would be allright.

Obviously not.

I'll have to replace the trim pot.
I have some 10K ones, perhaps they'd be a better bet than the existing PD.

Either that or just forget it.

[Paul Barker]

Obviously you have to change your coupling cap. Now your -3db is 30hz.

[Cressy Snr]

Obviously you have to change your coupling cap. Now your -3db is 30hz.

Yeah I realise that Paul.

Makes it seem hardly worth the bother.
I'm going rapidly off the whole idea.

Sitting listening to it now back in cathode bias, as I type.
I keep asking myself, whether I really need to do anything to it anymore.

[pre65]

I keep asking myself, whether I really need to do anything to it anymore.

And how many time have you said that ?

[Cressy Snr]

I keep asking myself, whether I really need to do anything to it anymore.

And how many time have you said that ?

Exactly so why not admit I can't leave the bloody thing alone

What I need is another serious project that'll make me keep my hands off the monoblocks.

Those 813s are still sitting in their dark little boxes. Then there's the OT100s.
I've still got my 801As too, plus a single STC 4304CB that could do with a mate.

I need to save for some iron. A decent 10K pair of output transformers would be a good place to start.

[ed]

Steve

I'm sorry to belabour your thread but being a bit naive I havn't seen that exact bias layout before.....The reason for my earlier post was that I've seen the bias hooked into the driver load/grid leak line with a cap beneath it.....

in my ignorance I thought this decoupled the driver load from the bias source...In your scheme the driver load is through the bias divider circuit....does this really not matter??? maybe I'm just making life more complicated than it really needs to be

[Cressy Snr]

Steve

I'm sorry to belabour your thread but being a bit naive I havn't seen that exact bias layout before.....The reason for my earlier post was that I've seen the bias hooked into the driver load/grid leak line with a cap beneath it.....

in my ignorance I thought this decoupled the driver load from the bias source...In your scheme the driver load is through the bias divider circuit....does this really not matter??? maybe I'm just making life more complicated than it really needs to be

Hi Ed all I did was make a mini full-wave rectified power supply for the bias.
Maybe it is doing it like that, that is the root cause of my problem as all the resistance I use in the adjustment potential divider apart from that above the wiper is in series with the grid.. I can't see how I'm going to get the grid resistance low enough to get good adjustment but avoid runaway.

Even if I get it low enough, it will only make the output valve harder to drive. Can't win really. You hardly see these types of DHT with fixed bias
wonder why
Are you using fixed bias on your 2A3 amp?
If so maybe you could let me see the schematic?

Half wave rectification is one way of getting the volts down without the need for a big value PD I suppose.

[ed]

Hi Steve

Like I said I'm a bit naive with this stuff but fwiw this is where I started with fixed bias...the 2 basic types(1 for the squeemish, in fear of pot failure, and 1 for the brave) :

[ed]

I don't use it on my 2a3 but this is what I do with my sv572, which is what caused my question in the first place:

[ed]

and here is the actual circuit..I'm currently messing with the difference between 1 cap power supply and clc, the thinking being that the ripple mst do something.....I don't have any figures yet.