• WARNING: Tube/Valve amplifiers use potentially LETHAL HIGH VOLTAGES.
    Building, troubleshooting and testing of these amplifiers should only be
    performed by someone who is thoroughly familiar with
    the safety precautions around high voltages.

'nipple' in waveform - What is it?

Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.
Interesting stuff.

kenpeter, I like that circuit, completely new to me. What job does R1 do (18R)?

(R1+R2) / R1 = Mu
(18+1800) / 18 = 101 (ratio as setup for AX7 use)
Attenuate cathode feedback influence to match plate Mu influence.

The reason ordinary cathodyne presents unity gains and unequal Z's is
overly strong cathode feedback. Cathode Mu'=1 is unequal to plate Mu.

One way to fix: Special triode of plate Mu=1, to match cathode Mu'=1.
Was actually done at least once, though forgot this example's part#.
With effective feedbacks from both ends, gains were -0.5 ~ +0.5 .
Not unities, as experience with cathode dominated feedback might
lead one to assume.

Alternate way to fix: Attenuate cathode Mu' to match the triode's plate.
In my circuit, this alternate method was attempted. Gains -50 ~ +50 .
 
Last edited:
Interesting stuff.

kenpeter, I like that circuit, completely new to me. What job does R1 do (18R)?

MelB, $100 a pair for grid chokes is more than I want to spend on this particular amp. I would like to try those chokes, though.

I noticed that you used the 6CG7 with only 3mA per triode. The rp will be well over 10k used like that. Does that make a difference? I thought we want low rp in a cathodyne.

Also, what's the advantage to the paralleled 12AX7 at the input? Perhaps this was a monoblock design, so rather than leaving a triode half flapping in the wind...

DIYBras, this amp will be run without gNFB, if possible. Maybe it'll get -6dB in the end. The schematic is straight from the simulation, so has an unrealistically high value resistor (R20) in series with the feedback resistor, taking it out of circuit. In other words, the amp's running open loop in the schematic. So yes, I'm only seeing that nipple using a high mu triode for the cathodyne, and with no negative feedback around the amp.

My goal here is to get a pair of 6L6-triode output tubes to sound good and make 10W output, and to use as few parts as possible to get there. I want simple, self-adjusting and totally reliable. I found that 6L6-triodes are much more linear running at 360V Vp and 60mA Ip than at the more usual 450V Vp and 45mA Ip. With a grid bias of only -30V, they should be easy enough to drive. EL34-triodes would be better, but I don't have any and they're more expensive than the Russian 6P3S.

I've modeled the circuit with 5965, 6SN7, 5687, EL86-triode and EL84-triode as the phase splitter. According to the models, the two best performing are the 5965 (albeit with that nipple when the 6L6s overload) and the EL84-triode, which looks really, really good (no nipple). Actually, EL84-triode looks like it drives the snot out of the output stage. The problem is that the EL84-triode will pull 20mA per channel from the power supply, while the 5965 draws only 4.5mA per channel. That's a big difference. I might have to compromise and go for 12AU7 or 6SN7, with about 7 or 8mA per channel.

What is the relationship between high-ish rp in the cathodyne and stability when the output stage starts to overload? It seems the lower rp triodes like 5687 or EL84-triode yield symmetrical clipping from the 6L6s, well into overload. Switch to the 5965 or 12AT7 and the nipple appears once the 6L6s overload. I don't think it's a coincidence.

I agree that an LTP with source followers would be better. However, I don't want to mess with a negative supply in this build. Hence the cathodyne, which yields near-perfect balance with few parts and little effort. But I see that the penalty is poor performance once the output stage overloads.
--

Yes I paralleled the 12ax7 just so it was not flapping around. I have 3 sockets in my chassis so I can always add another stage now or maybe change the phase splitter if I want.
3 mA seems to be plenty to drive even a KT120...check the 10K square wave at the speaker outs. 12ax7 to cathodyne to power tube is a very good sounding circuit to me. :) (No global negative feedback)
 

Attachments

  • 10Khz.JPG
    10Khz.JPG
    670.2 KB · Views: 61
Status
This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.