cathode bias circuit question

[gary h]

Hi all,

I'm having trouble understanding how the cathode bias circuit recommended for the Aikido line stage is implemented. I thought that cathode bias was to raise AC voltage up so the negative swing was above zero. The scheme I'm working with references it to 1/4 B+ voltage. Here's the excerpt that is confounding me: "...since one triode stands atop another (in the same envelope), both triodes can never share the same heater-to-cathode relationships."

I have 2 dc regulator boards feeding tube heaters, I had originally planned on each board serving the tubes for each channel. I'm wondering now if I can merge both dc boards as one supply, insert the cathode bias circuit, and then on to my tubes. Here's a diagram, what are your thoughts?

gary

[mashaffer]

Actually the point of bias (whether cathode or not) is to make the DC potential of the cathode higher than that of the grid which sets the DC current through the tube. It will also determine how high and low the AC signal can go before serious distortion of course as too large a signal would drive the tube into grid current (grid higher than cathode) or cutoff.

I am not familiar with the context of your quote but I interpret what they are saying to be that since each tube has only one heater then if the same bottle is used for the top and bottom tube the potential between the heater and cathode number one can not be the same as that between that same heater and cathode number two. The only way to do that is to use one bottle for the top and another for the bottom and use two separate heater supplies.

You could use one bottle for the lower tubes in both channels and another for the upper tubes. Then use one DCPS for the top tube and the other for the bottom tube. The reference each one slightly higher than their respective tubes cathode so that rectifier and regulator hash is less likely to get into the audio.

I am not sure I understand how any of this would effect a cathode bias situation other than needing to know what you cathode voltage will be.

Your diagram makes me think you are looking at fixed bias however as you seem to be doing a voltage divider on the B+ which would not be needed for normal cathode bias. Maybe if you showed the entire circuit it might be more clear to me what you are trying to do.

I am also wondering why we are going to the trouble of DC heaters on a line stage.

Oh, wait, I think I see. What you have labeled as a cathode bias circuit is just the VD to reference the heater supplies. But you have both heater supplies tied to the same reference voltage... so are you using two just to get enough current capacity?

[ray_moth]

What you are calling a 'cathode bias' circuit in your diagram is actually a heater bias circuit, which will result in the heaters being elevated about 75v DC above ground. For a single 6SN7 doing duty as both the top triode and the bottom triode, this should be OK. The 6SN7 has a heater-cathode voltage rating of -200v to +100v, so 75v is a happy medium. You should increase the bypass cap in the heater bias circuit from 0.1uF to 10uF.

[pchw]

Gary,

What JB said about biasing the filament to 1/4 of the B+ is very much for those using his PCB. In his PCB, each tube is "wired" as the top and the bottom. If you are doing wire-to-wire, you can follow what Mike said - use one tube for the bottom triode of both channels, and another tube for the top triodes. This way, you only need to bias the top triode to whatever optimal B+ that the 6SN7's are happy; however, this also mean that you will need 2 filaments supplies which don't seem to be a problem to you :-)

Ray, JB uses a 0.1uF cap to bypass the heater bias to ground. I have seen people recommending up to 100uF. What is the reason that you use 10uF? I kind of don't quite understand the rationale of different capacitance and m curious the impact of different value to the circuit.

[athos56]

On a related note, would this heater bias have any effect on a 6D22S rectifier using the same 6.3v heater string?

[pchw]

According to Svetlana's data sheet, 6D22S's heater to cathode range is +200 to -600. As long as the bias point is within this range, it will be OK.

[athos56]

I thought thats what that ment! Ahh slowly but sure I'm picking this up. Thanks!

[gary h]

For me it's very slowly and not very surely. Thanks for the help guys. Ray, could you talk a little about the bypass cap, why 10uF instead of .1uF?

Heater bias is correct. My mistake, but won't that then elevate the heater to cathode voltage relationship equally for both triodes within the tube? How does this discourage electrons from moving from the cathode of one triode to the plate of the other? I assume that this is the point here. Is this what is referred to as hash?

gary

[koolatron]

The current draw from that heater bias supply should be exceedingly low (close to zero in all but catastrophic cases). There is no good reason for excessive capacitance on that node, unless you have extra ginormous capacitors that you are looking to waste.

By the time you are dividing B+ for heater biasing, you are ideally already be starting with a very clean source voltage. The only extra noise you'll be seeing is directly from the resistors used in the divider chain, and even then all you're doing is using that voltage to reference an AC heater tap. If you notice it, I'll be surprised.

If I understand correctly, the real reason for that 0.1uF capacitor is to provide an AC ground for the heater reference. If there's a small amount of AC on the center tap (due to its not being precisely in the 'center'), it'll find an easy path to ground through that cap instead of perturbing your bias voltage or getting into B+.

-k

[Sheldon]

Originally posted by gary h why 10uF instead of .1uF?

The corner for with that resistor string and a 0.1uF cap will be about 20Hz, which will filter most line hum. But no reason not to knock it down more. 10uF should be plenty, but this is DIY, and it's fun to experiment. Try the 0.1 and try it with 10uF clipped in parallel. If you can't hear a difference leave in the extra or not.

Originally posted by gary h
won't that then elevate the heater to cathode voltage relationship equally for both triodes within the tube? How does this discourage electrons from moving from the cathode of one triode to the plate of the other?

No. You have one heater element, but two independent cathodes per dual triode. As mentioned earlier, if the triodes from one tube are stacked in the circuit (one on top the other, which is usually done with the Aikido so that you can independently change input and output tube types), then one cathode will be above the filament potential and one below. With a single tube, one usually raises the heater above the cathode, because that's slightly better, and easier, than having it lower. But having the heater at lower potential than the cathode is almost as good for noise cancellation. Example curves are floating around here somewhere.

This has nothing to do with electrons moving from one cathode to the opposite plate (which is very low, otherwise putting two triodes in a single envelope wouldn't work). This only concerns fluctuations in the heater potential from influencing the cathode.

Sheldon