[pvico]

There is not much space to put bigger capacitors there. Moreover, my schematic is incomplete on the power supply side. There is another 4N60 MOSFET between those 2 caps adding either some kind of regulation or just an enable/disable function. I don't know what drives the gate of that MOSFET

And there is a signal from the 12.6V (11.7V measured) through an optocoupler (IC2) back to the switching controller IC.

EDIT: on power up, HT rises to 250V and within seconds drops to 179V, probably after the tubes have heated and conduction takes place.

 

[CZ4A]

The increase in Cathode cap is to stabilize that voltage drop for bass sustain and bass punch,
Which is optimized with larger Cathode Electrolytic value of 1000uf.

This is due to the "larger" cathode cap value causing the effect of holding the cathode voltage constant, in relation to current demands,
So the tube is kept at optimal gain.

Also, at such large capacitance of 1000uf, it should compensate for any performance or noise issues of the combo of Q2&IC3 wich, as "pvico" noted, are china parts which set the tube bias at 25ma.

Interesting, that probably explains why I didn't hear any difference when I replaced the stock Nichicon cathode caps with Elna Silmics of the same capacitance (after I replaced the coupling caps w/ WIMA 0.47uf caps).

@pvico Awesome to see all of the investigation and schematics you have made of this amp! Quick question, what does "HT" stand for?

 

[pvico]

Quick question, what does "HT" stand for?

HT is commonly used to designate a DC high voltage supply. You sometimes see B+ as well

 

[CZ4A]

HT is commonly used to designate a DC high voltage supply. You sometimes see B+ as well

Ah, thanks. I'm familiar with B+ but haven't seen HT used to mean the same thing before.

 

[jekjek]

I have one thing to add
The amp sounds finer with 12ax7 tubes
When i use 12sn7, i lost some small background details

 

[Ctritical Bill]

Thanks pvico for the schematic.
I measured the B+ and it was 185v not 210v. Also, the heater voltage was 11.9v so I don't know if that affects your calculation of the bias current.

 

[pvico]

I measured the B+ and it was 185v not 210v. I don't know if that affects your calculation of the bias current.

Yes, I just measured HT (B+) at 179V and the heater voltage at 11.7V (with the original tubes). I have updated the schematic accordingly. Here (Belgium), the mains voltage is 230V. If you're in the UK, it's more likely 240V which can explain the difference.

I don't know if that affects your calculation of the bias current.

No, the bias current is set by R29, R30 and IC3 which has a fixed 2.5V reference. So, the bias current is independent of HT.
Look at it this way: IC3 will do whatever it can on the gate of Q2 so that its ref input is at 2.5V. R29 and R30 in parallel have a value of 100 ohm. 2.5V / 100 ohm = 25mA

 

[Maxx134]

This is all excellent info, thanks pvico.
As for the bias, this is important to know so we can pick optimal tube for that bias point..

Anyways, from my testing, the 6V6 were the most optimal,
Until I found out about the 6F6...

Its output is roughly similar to that of the more well-known type 6V6,
but it is considerably more sensitive(!),
So it is even better at micro detailing from my own observations.

Also, the "42" tube with adapters was also a close choice for best tube,
But for now, the best tube choice I like are the 6F6, specifically the Russian 6Ф6С / 6F6C Cobra...
Thanks to member "Kryl007" for that recommendation.

Before the Russian tubes came in,
My best choice I was using was the Tung-Sol 6F6.
Tung-sol had great juicy overall sound,
But the treble detailing of the Russian tubes pushed it into the lead and made my amp more resolving so it overcome any losses anywhere else,
Like the output transformers, as we know all transformers have some loss and that's why the good ones are expensive.

So far I am really almost dumbfounded how good this little amp is sounding ..

 

[pvico]

Look at it this way: IC3 will do whatever it can on the gate of Q2 so that its ref input is at 2.5V. R29 and R30 in parallel have a value of 100 ohm. 2.5V / 100 ohm = 25mA

Q2 with the help of IC3 and R29/R30 is a constant current sink.
The bias current will be 25mA independently of HT or of the tube used.
It will also actively counteract the current variations, mostly in the bass frequencies, this probably explains the original poor bass.
Bass dramatically improved (almost too much...) since I replaced the cathode caps by Nichicon Muse 1000uF and the coupling caps by Mundorf silver gold oil .47uF.

My 6F6's are on their way from Ukraine

 

[pvico]

When I used the Tung Sol KT66 Reissue's, the right tube glowed noticeably brighter than the left one. Intrigued, I measured the voltages at the heater pins of both tubes: the left one had 5.50V and the right one 6.09V.
While 6.09V is within specs (6.3V +- 10%), 5.50V is too low and this can lead to cathode poisoning, degrading the tube over time. However, I did not notice a significant difference in audio output between the left and right sides (but I'm not an expert audiophile).

I removed the tubes and powered their heaters from my bench power supply set on a stable 6.3V. One tube had .890A, the other .918A, so there is a slight imbalance between the 2 tube heaters.

Because the heaters are connected in series, that imbalance is exacerbated: the more they heat up, the more the heater resistance increases which further increases the imbalance (the TS KT66 heater has a resistance of 2 ohm when cold increasing to 7 ohm when at temperature, all tubes react in a similar way).

If you have the possibility to do so (if you have more than 2 tubes of the same model), try different combinations of them so that they glow in a similar way.

WARNING​

If you want to measure the heater voltages while the amp is powered, do that only if you know what you are doing. The HT voltage can kill you!

 

[Maxx134]

Q2 with the help of IC3 and R29/R30 is a constant current sink.

Excellent..
I always Iike to share info so that others may learn whats going on in a circuit so to try understand the choices these designers make in thier circuits..

This was done in the Little Dot Supermod thread,
so others can learn instead of being in the dark as to whats going on.

If you have the possibility to do so (if you have more than 2 tubes of the same model), try different combinations of them so that they glow in a similar way

Yes this seems common to tubes.
The heater coil will light up different pace on every tube.

Plus, if the heaters of the tubes are in series like you say on this amp,

Because the heaters are connected in series, that imbalance is exacerbated

then for sure they will never light up same time, hehe.

The heaters are mini coils by nature not even.
Sometimes the cold resistance of the heater will be the issue, as it will be lower than the hot resistance,
Which is the case if the ends of the heater is sticking out of the cathode sleeve and or visible.

Cold heater resistance causes short term inrush current and localised heating creating quick flash or bright glow.
You sometimes see the difference lighting up faster, as there is a short current 'burst' creating a "heater flash".
I have one set of Mullard tubes that scared crap out of me when they did this on every startup lol.

So, Once the coil heats up the resistance increases and the current falls and heating becomes uniform over the whole length of the heater wire.
Anyways, "heater flash" is not an issue.
Instead, it is an indication of NOS and old, very early production years,
which are usually considered as tubes of a higher quality than New stock current production types .

 

[Maxx134]

When I used the Tung Sol KT66 Reissue's

I also still have the same reissue tubes, they look dam pretty.
I also have the big 6L6 and others I rolled into this amp...
I need to make a list..
The 6F6 sounds better still.

 

[baronbeehive]

Excellent..
.............

Yes this seems common to tubes.
The heater coil will light up different pace on every tube.

Plus, if the heaters of the tubes are in series like you say on this amp,

then for sure they will never light up same time, hehe.

The heaters are mini coils by nature not even.
Sometimes the cold resistance of the heater will be the issue, as it will be lower than the hot resistance,
Which is the case if the ends of the heater is sticking out of the cathode sleeve and or visible.

Cold heater resistance causes short term inrush current and localised heating creating quick flash or bright glow.
You sometimes see the difference lighting up faster, as there is a short current 'burst' creating a "heater flash".
I have one set of Mullard tubes that scared crap out of me when they did this on every startup lol.

So, Once the coil heats up the resistance increases and the current falls and heating becomes uniform over the whole length of the heater wire.
Anyways, "heater flash" is not an issue.
Instead, it is an indication of NOS and old, very early production years,
which are usually considered as tubes of a higher quality than New stock current production types .

I always wondered about this since seeing a friends tubes flash on startup, then my Amperex 12AX7's did the same thing but not the Amperex EL84's. It's fun when you get used to it but the first time it happens you freak out.

That's also useful to know about the heater glow. I've always swopped around the tubes to find the right combination but something puzzled me lately. I was trying to equalize the sound output from each power tube by ear and by the glow and I didn't realize that although I had a stronger glowing tube on one channel it sounded weaker. So that explains that, now I understand I was going by the heater glow not tube trandsconductance.

Edit:
When I said in a previous post that the transients were sharper I really meant faster, clearer and more detailed, because they do not sound sharp.
And in that respect it is better than my headphone amp - the modded Little Dot MKVI+.

Damn I really want one of these babies now to go with my PA0901A APPJ!

 

[jekjek]

Thank you Maxx134 and pvico for your contribution to this thread
I learn a lot more about this amp and cancel my plan to measure the heater voltage

 

[Maxx134]

Edit:
When I said in a previous post that the transients were sharper I really meant faster, clearer and more detailed, because they do not sound sharp.
And in that respect it is better than my headphone amp - the modded Little Dot MKVI+.

From what I have learned about those amps, is their clarity and detailing is mostly attributing to the tubes and bias points, once the amp is sufficiently moded, as the circuit design is very good.

Especially the issue of soundstage and detailing is dependent on bias.
In Your specific case, the driver tube bias need to be boosted and you will be shocked how much more it can go....