I don't understand the phobia so many people have towards negative grid bias, it eliminates all of these issues and if you use a pot you can easily optimize the operating point on a per tube basis.
There is nothing more transparent than removing a large, not so great capacitor in the signal path and replacing it will a very small value resistor or a piece of wire..
http://www.chimeralabs.com/images/axiom.sch.jpg
It uses damper diodes.
Large/old style of 'Red light zone' ?
Interesting circuit, though.
This amplifier has low measured distortion and a very lean, tight sound which a lot of audiophiles like. I think vacuum diodes have their place in tube audio design. I'd give it a shot if I were you. You might like it.
John
I don't understand the phobia so many people have towards negative grid bias, it eliminates all of these issues and if you use a pot you can easily optimize the operating point on a per tube basis.
There is nothing more transparent than removing a large, not so great capacitor in the signal path and replacing it will a very small value resistor or a piece of wire..
Kevin,
Have you ever used an adjustable regulated negative supply for grid bias?
John
There is nothing more transparent than removing a large, not so great capacitor in the signal path and replacing it will a very small value resistor or a piece of wire..
Kevin, true that you are removing capacitor from one place, but you are introducing it to another (mainly negative supply filtering).
The one on bias supply wouldn't (shouldn't) be in the main signal current loop(s).
So it's still positive in this regard, I think.Is that for me?
Why? ( I like tight and fast, but I don't like lean.
)It's the circuit and concept attract me, I haven't had any chance to give it a listen. And I myself don't like to build such an complex amp...
CCS bias controls the wrong thing: average current rather than quiescent current. In this respect it is worse than a resistor! It also changes the LF phase response as a bypassed CCS is a high-pass filter, while a bypassed resistor is a lead-lag. However it might help prevent mains voltage variations from causing woofer wobble.
I don't understand the phobia so many people have towards negative grid bias, it eliminates all of these issues and if you use a pot you can easily optimize the operating point on a per tube basis.
There is nothing more transparent than removing a large, not so great capacitor in the signal path and replacing it will a very small value resistor or a piece of wire..
Well, it's rather difficult to make a direct coupled grid bias circuit. Whether or not DC is a smart thing to do, is a whole other discussion.
I agree, removing components off the signal path is a good thing - that's why I'm doing DC.
Those VR tubes got me thinking again.
So I came up with this idea, from ground up to 45 cathode:
ground -> two rechargeable batteries in parallel -> two 0C3's in series -> 45 cathode
The batteries would stop all feedback - two in parallel would ensure that the current (about 37mA) wouldn't strain them too much as it'll be hot enough inside the chassis. The 0C3 regulators would make the cathode about 210 volts positive, giving good bias if the 45 grid is at about 150 - 160 volts.
So this would give essentially fixed bias, no caps, no resistors in the cathode circuit. Also (I'm destroying my hifi reputation here) I think the 0C3's look really really nice glowing.
This would work, right? Any negative sonic effects to take into account?
So I came up with this idea, from ground up to 45 cathode:
ground -> two rechargeable batteries in parallel -> two 0C3's in series -> 45 cathode
The batteries would stop all feedback - two in parallel would ensure that the current (about 37mA) wouldn't strain them too much as it'll be hot enough inside the chassis. The 0C3 regulators would make the cathode about 210 volts positive, giving good bias if the 45 grid is at about 150 - 160 volts.
So this would give essentially fixed bias, no caps, no resistors in the cathode circuit. Also (I'm destroying my hifi reputation here) I think the 0C3's look really really nice glowing.
This would work, right? Any negative sonic effects to take into account?
0C3 specs are:
Voltage Regulator
Maximum Supply Voltage ........................ 133 V
Regulated Voltage ............................. 105 V
Maximum Current ............................... 40 mA
Minimum Current ............................... 5 mA
Regulation (Min to Max Current) ............... 2 V
Voltage Regulator
Maximum Supply Voltage ........................ 133 V
Regulated Voltage ............................. 105 V
Maximum Current ............................... 40 mA
Minimum Current ............................... 5 mA
Regulation (Min to Max Current) ............... 2 V
OK, so you'll be running it right at the ragged edge. I'd be cautious.
If you want to keep the glow (and the noise and the voltage instability) of the gas tubes, you could use them as a reference at the base of a bipolar (or the gate of a FET), run 10mA through them, and let the solid state device handle the power tube's current.
If you want to keep the glow (and the noise and the voltage instability) of the gas tubes, you could use them as a reference at the base of a bipolar (or the gate of a FET), run 10mA through them, and let the solid state device handle the power tube's current.
A couple of thoughts here:
In your zeal to acquiesce to my and other people's suggestions for fixed bias you might be heading down a blind alley. I have done this and it's a good learning experience, however your initial instincts might provide a better result.
I was an early advocate of cathode based battery bias with Nicads, (10+ yrs ago) but have since changed my mind. I think this in many cases compromises the sound more than simple RC cathode bias does, and is grossly inferior IMHO to fixed bias applied in the grid circuit.
Capacitor quality and performance tends to decrease as value increases, restoring some semblance of equality seems to require exponentially larger expenditure on cathode bypass caps which is why I am such a proponent of fixed bias, notwithstanding the often significant gain in power utilization efficiency, heat generation, and tweakability. Small value high quality capacitors are easily achieved for use in ac coupling whereas larger values are not. (That said I have designed several good DC coupled two stage amps with large film caps and power resistors in the output tube cathode circuit with hum canceling - nothing over 3Wrms output however.. And I am working on another such design.. All this stuff goes to Asia.)
Sorry to belabor the above points, I guess this is one of the dangers of "design by forum.." Ultimately it is your choice to decide what works for you, take all obviously subjective design stuff (not safety related though!) with a grain of salt, and do what you want. Let us know how it works out.
- Rechargeable batteries in parallel aren't guaranteed to share the current equally. Some (many) types also have rising impedance with frequency. A single set of cells rated at 20X the continuous current flow in your circuit will essentially trickle charge and last a few years. (Anything above 800mAH should be OK.) Sanyo Cadnica are the best sounding of the bunch I have tried and have the lowest internal impedances I've seen.
- The dynamic (AC) impedance of those gas tubes in series is not as low as you would think, and they generate significant noise. You can't capacitively bypass them in a meaningful way for audio frequencies either or they will become relaxation oscillators. Dynamic impedance in these tubes might be as much as a few hundred ohms each.
- Gas tubes have a significant tolerance on the initial voltage value they regulate to. You will not know your exact operating point and it cannot be accurately predicted in advance.
In your zeal to acquiesce to my and other people's suggestions for fixed bias you might be heading down a blind alley. I have done this and it's a good learning experience, however your initial instincts might provide a better result.
I was an early advocate of cathode based battery bias with Nicads, (10+ yrs ago) but have since changed my mind. I think this in many cases compromises the sound more than simple RC cathode bias does, and is grossly inferior IMHO to fixed bias applied in the grid circuit.
Capacitor quality and performance tends to decrease as value increases, restoring some semblance of equality seems to require exponentially larger expenditure on cathode bypass caps which is why I am such a proponent of fixed bias, notwithstanding the often significant gain in power utilization efficiency, heat generation, and tweakability. Small value high quality capacitors are easily achieved for use in ac coupling whereas larger values are not. (That said I have designed several good DC coupled two stage amps with large film caps and power resistors in the output tube cathode circuit with hum canceling - nothing over 3Wrms output however.. And I am working on another such design.. All this stuff goes to Asia.)
Sorry to belabor the above points, I guess this is one of the dangers of "design by forum.." Ultimately it is your choice to decide what works for you, take all obviously subjective design stuff (not safety related though!) with a grain of salt, and do what you want. Let us know how it works out.
Last edited:
Or, use them as a reference voltage for a pentode that can take the current at the required voltage.
Honestly though, I'm pretty much set on using arrays of LED's for Cathode bias. Still cheaper than a power resistor and AudioPhool capacitor, Free power on indicator, Great AC characteristics, and damn near constant voltage drop.
A brilliant idea! Anyone who thinks of that must be a genius!!
The LEDs also act as an overload indicator- the flickering in brightness when I drive my amp into AB is very noticeable, and even more so when it clips.
When used in cathode there is no load for it. And it is rated for twice higher current than usual LED. However, some minimal current too keep it glowing have to be supplied by an additional resistor, to prevent shut down when an output tube cuts off, otherwise nasty clipping guaranteed.
Why it oscillates? Because higher voltage is needed to ignite it than to sustain glowing. If current is always higher than needed to keep it glowing it won't oscillate.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.