Although I understand the drawback of displacing distortion from the output to the input stage of an amp, I have this urge to try using a cathode follower in a low power amplifier output. The two stages of 6SN7 also make me feel all warm and fuzzy because of how often it appears in classic DHT designs.
There's a little serendipity with the above schematic as it looks to me like I can swing 400V ptp and the Vhk of the 6L6GC is +/-200V. Probably good for a bit under 4W into an 8 ohm speaker (also want to try headphones with and without a resistor to load the primary).
I've got all the parts and tubes aside from the Hammond chokes (outputs are Edcor XSE15). Anyone care to talk me out of this or suggest tweaks?
Yesssss‼ A good decoupling cap in parallel to the 400ΩR, otherway you will lost almost all signal across it.
Hi Osvaldo. Thank you! I didn't consider that. The transformer will reflect a 5k impedance in series with the cathode resistor, so I don't think I'd lose that much AC signal (or am I missing something?). But I can't think of a reason not to bypass it.
400Ω / (5000Ω + 400Ω) = 400/5400 = .074, then about 7.4% of the signal will be lost in it. But still worse is that it increases the output impedance looked by the reflected load through the transformer, loosing the main advantage of the cathode follower.
OK, I learned this thanks for your "ancestors" and teachers; Seely, Wallman, Valley, Terman, and many other electronic pioneers.
Perhaps if you like falsely increased bass response, leaving the resistor unbypassed, will reinforce speaker's resonance in the bass range. Undesirable for me, but I see that someone like this behavior.
PD: Why not to try a pentode cathode follower, with higher gain and lower output impedance?
Perhaps if you like falsely increased bass response, leaving the resistor unbypassed, will reinforce speaker's resonance in the bass range. Undesirable for me, but I see that someone like this behavior.
PD: Why not to try a pentode cathode follower, with higher gain and lower output impedance?
Last edited:
A circuit like this, but with the output transformer in place of the "resolver stator" (G. Valley, H. Wallman - Vacuum tube amplifiers - 1948, page 350).
I would use the choke for the cathode, and from it, bootstrap the load anode resistor in the 6SN7 stage to improve linearity and gain.
My only idea.
I would use the choke for the cathode, and from it, bootstrap the load anode resistor in the 6SN7 stage to improve linearity and gain.
My only idea.
Attachments
Why not ground the transformer and use fixed bias; it shouldn't be much of a risk with this low gm valve.
Actually, that idea of Osvaldo looks very interesting.
Actually, that idea of Osvaldo looks very interesting.
Last edited:
Depending on the trafo's DCR, perhaps no external bias nor resistor needed, but the guy didn't say anything about it.
I'm not sure. The cathode follower gives almost the same voltage output as in the input, so no voltage gain. And current gain in a triode and in a pentode of the type we are speaking about, I don't believe that they be too different.
I was curious about the practical limits for performance in an SE amp with a transformer so I built a follower amp with a mosfet as a power device and a low-distortion driver. See here for details.
I think the dominant source of distortion in the amp is output transformer nonlinearity, but I have yet to definitively show that. (I blew the output device before I could take comparative distortion measurements on the primary and secondary of the output transformer and haven't repaired it yet, but I am fairly certain there is much less distortion on the primary than what I am getting out.)
Anyway, you should be able to make a very nice SE follower amp if you do it right.
I think the dominant source of distortion in the amp is output transformer nonlinearity, but I have yet to definitively show that. (I blew the output device before I could take comparative distortion measurements on the primary and secondary of the output transformer and haven't repaired it yet, but I am fairly certain there is much less distortion on the primary than what I am getting out.)
Anyway, you should be able to make a very nice SE follower amp if you do it right.
Of course they will be different, and drastically so. A pentode will be able to swing to a much lower Va-k without positive grid drive. In the circuit above, there will be much more swing capability in a pentode to drive the primary and thus more voltage swing on the secondary and more power out.
Edit: It doesn't matter what the voltage gain is if swing is limited by the Vg-k=0 line in one and not the other. Output swing will hit a wall at Vg-k=0. Your comments about current gain don't make sense to me. These aren't BJTs.
Last edited:
I have to look back at my plate curves (on another computer), but I believe the 6L6 was hitting Vg=0 before the 6SN7 driver. Though it wasn't by a large margin with the operating points I chose.
The pentode cathode follower is a cool idea. I am assuming the Vhk shouldn't be violated for the output cathode follower and this limits the maximum swing if it is at a fixed bias though. Does that Vhk apply to AC signal swing or just DC?
Thanks for chiming in SpreadSpectrum. I was hoping you'd see the smoke and come running.
The pentode cathode follower is a cool idea. I am assuming the Vhk shouldn't be violated for the output cathode follower and this limits the maximum swing if it is at a fixed bias though. Does that Vhk apply to AC signal swing or just DC?
Thanks for chiming in SpreadSpectrum. I was hoping you'd see the smoke and come running.
Vhk applies to both AC swing and DC. Just use a separate filament winding to heat the filament of the output tube and reference it to the cathode through a modest series resistance. (Say 4.7K)
I would try a sweep tube as a pentode connected output follower - should work better than a 6L6 due to its lower internal losses, higher transconductance, etc..
On my Unity-Coupled amp I used a separate small transformer for the heater of each KT88 because I was concerned about heather-cathode insulation stress. I just connected the transformer center tap directly to the cathode.
If you go the route of pentode cathode follower, you have to figure out the screen supply. It could probably be done with an appropriately-sized cap that holds a charge throughout the AC swing. Otherwise, you would have to make a floating screen supply which would be considerably more complicated. The voltage between screen and cathode needs to remain constant, which means that the screen voltage needs to be able to go above B+ for part of the swing.
I sidestepped the whole issue by using a mosfet, which doesn't require a screen supply.
If you go the route of pentode cathode follower, you have to figure out the screen supply. It could probably be done with an appropriately-sized cap that holds a charge throughout the AC swing. Otherwise, you would have to make a floating screen supply which would be considerably more complicated. The voltage between screen and cathode needs to remain constant, which means that the screen voltage needs to be able to go above B+ for part of the swing.
I sidestepped the whole issue by using a mosfet, which doesn't require a screen supply.
For this amp I won't be able to include more heater supplies. The power supply is a standalone regulated PSU with 6A of 6.3V heater capacity. I have it referenced to 1/6 B+ at this point, but that can be changed easily. Thanks to kevinkr for his help with that on another thread.
I'm kind of building this in reverse🙂 The power supply was a separate project so that I could learn about series-pass supplies. With the fixed heater bias taken into consideration, I think my limiting factor would be the Vhk of the 6L6GC.
Or maybe I scrap the cathode follower for now and just build a regular single-ended amp. Seems so simple by comparison.
I'm kind of building this in reverse🙂 The power supply was a separate project so that I could learn about series-pass supplies. With the fixed heater bias taken into consideration, I think my limiting factor would be the Vhk of the 6L6GC.
Or maybe I scrap the cathode follower for now and just build a regular single-ended amp. Seems so simple by comparison.
- Status
- Not open for further replies.