I'll try to explain......
The gas tube needs a higher voltage to "strike" than "maintain". This is just like a common wire ended neon. This is negative resistance.
Have you never made a relaxation oscillator? The simplest is a neon in parallel with a capacitor.
A gas tube in series with a choke is equivalent.
The gas tube needs a higher voltage to "strike" than "maintain". This is just like a common wire ended neon. This is negative resistance.
Have you never made a relaxation oscillator? The simplest is a neon in parallel with a capacitor.
A gas tube in series with a choke is equivalent.
Re: I'll try to explain......
And why is it that you think this can be accomplished with resistance, but not with a choke of equivalent DC resistance? Why would the choke care if the DC voltage goes up, down, or stays the same? The choke only starts to act "chokelike" in the presence of frequency. Is there something I'm missing?
dhaen said:
And why is it that you think this can be accomplished with resistance, but not with a choke of equivalent DC resistance? Why would the choke care if the DC voltage goes up, down, or stays the same? The choke only starts to act "chokelike" in the presence of frequency. Is there something I'm missing?
Well, as I quoted, if "all of the signal" (AC) is dropped across the resistor in the 1st drawing, that means the 0A3 is a short for AC purposes. And you do not argue that the choke could be there all by itself - in other words as the "load" for the CF, right?
So, what is there to cause oscillation? All I see at audio freq is a choke tied to cathode of a cathode follower.
Maybe you could explain your view with some detail?
So, what is there to cause oscillation? All I see at audio freq is a choke tied to cathode of a cathode follower.
Maybe you could explain your view with some detail?
Joel, you're hard work today.
I suggest you go away, and read up on "relaxation" oscillators.
Then consider the conditions in that circuit at various points of the signal cycle.
Take into account that the cathode current can vary from almost zero to a max value.
Consider the release of energy from the inductor.
I suggest you go away, and read up on "relaxation" oscillators.
Then consider the conditions in that circuit at various points of the signal cycle.
Take into account that the cathode current can vary from almost zero to a max value.
Consider the release of energy from the inductor.
Ok... with the capacitor across the gas tube, you have a situation where, if we have enough capacitance, the anode of the gas tube can be swung on the negative cycle into cutoff. Since the tube is not performing its regulator duties anymore the cathode is charged positive. When it reaches the strike voltage, the gas tube clamps down in reverse, and then you repeat the whole process over again. Sound about right? (I skimmed it quickly).
I'm not sure I see how the choke, present only at the anode, can accomplish that.
I'm not sure I see how the choke, present only at the anode, can accomplish that.
Why wouldn't this work? Power Output?
I actually was thinking along these lines also.
What voltage / current are you going to bias those 6080's at?
Any idea on what type of coupling caps you will use?
You could probably get away with less than a 1H choke, I'd go with higher power handling and less H. Also, perhaps go with both a fixed and cathode bias scheme.
The only problem with this type of design is need a really hum free noise free power supply capable of 1A or so at 135v.
It would be a nice room warmer though!!!
Ever thought about using EL509's instead?
I actually was thinking along these lines also.
What voltage / current are you going to bias those 6080's at?
Any idea on what type of coupling caps you will use?
You could probably get away with less than a 1H choke, I'd go with higher power handling and less H. Also, perhaps go with both a fixed and cathode bias scheme.
The only problem with this type of design is need a really hum free noise free power supply capable of 1A or so at 135v.
It would be a nice room warmer though!!!
Ever thought about using EL509's instead?
Results
John,
I breadboarded the gas-tube coupled stage over the weekend. No oscillation, or wierd effects at all with the choke (150Hy) - although it was a good deal louder than using the single 2.5k resistor.
I used music from CD, and a 2V RMS 1kHz sine wave to test.
So, there must be something, er.... incomplete about your theory?
John,
I breadboarded the gas-tube coupled stage over the weekend. No oscillation, or wierd effects at all with the choke (150Hy) - although it was a good deal louder than using the single 2.5k resistor.
I used music from CD, and a 2V RMS 1kHz sine wave to test.
So, there must be something, er.... incomplete about your theory?
Re: Why wouldn't this work? Power Output?
135V on the plates, -30V on the grids.
Hopefully none - see the "gas tube coupling" discussions above.
On the output it would have to be an electrolytic of course, although I've done that before and have no objections to it.
It is a fixed + auto bias scheme. The choke does drop 5-6 volts.
No, because I just got sixteen 6080's for $2 each from a guy in Canada, off eBay.
RichardJones said:
135V on the plates, -30V on the grids.
Hopefully none - see the "gas tube coupling" discussions above.
On the output it would have to be an electrolytic of course, although I've done that before and have no objections to it.
It is a fixed + auto bias scheme. The choke does drop 5-6 volts.
No, because I just got sixteen 6080's for $2 each from a guy in Canada, off eBay.
Joel,
More seriously: You may not notice it. It may not happen on your prototype, it may not happen if you build it. But it will happen sometime.
Engineering is about designing something that's stable, and gives consistent results. For that we have rules. One design rule is to keep a good phase margin against oscillation.
I'm sure even you would not put caps aross a gas tube; would you? It might "not" oscillate.
Cheers,
BTW, What current are you running through that gas tube?:
No, it might not. But with some rate of change of voltage, I assure you it will oscillate, if only for a few cycles, until the conditions change. That'll be nice for the music
More seriously: You may not notice it. It may not happen on your prototype, it may not happen if you build it. But it will happen sometime.
Engineering is about designing something that's stable, and gives consistent results. For that we have rules. One design rule is to keep a good phase margin against oscillation.
I'm sure even you would not put caps aross a gas tube; would you? It might "not" oscillate.
It's not my theory. And, well, I never said it was 49.
Cheers,
BTW, What current are you running through that gas tube?:
...are you sure
about SE OTL ? There's a time and place for SE circuits and I don't think this is one of them , sorry mate . That big electrolytic sitting on the output should really be in the power supply , I would think the cap would fudge the sound far more than a transformer ever would . I find paralleled triodes sound a lot worse than a bigger single (especially 6080/6as7 which more than likely will have ****-poor section matching) . 6C33 are about $10 each at the moment at some vendors . How about something differential 'semi' transformer coupled , say 6c33's with cathode transformer loading (transformers wired as autoformers) ? You could lose the loop feedback as well as the large and nasty big output coupling caps . Even a mains toroid could be used as the output tranny for a prototype , they work well and I'm using one at the moment . This should achieve simplicity and low output impedence , remember loop feedback encourages high order harmonics and phase shifts etc etc etc . If you want to inject 2nd harmonic with this type of circuit , re-jig your line stage with a lower value of anode load . No flames : please note , these are serious constructive comments
316a
about SE OTL ? There's a time and place for SE circuits and I don't think this is one of them , sorry mate . That big electrolytic sitting on the output should really be in the power supply , I would think the cap would fudge the sound far more than a transformer ever would . I find paralleled triodes sound a lot worse than a bigger single (especially 6080/6as7 which more than likely will have ****-poor section matching) . 6C33 are about $10 each at the moment at some vendors . How about something differential 'semi' transformer coupled , say 6c33's with cathode transformer loading (transformers wired as autoformers) ? You could lose the loop feedback as well as the large and nasty big output coupling caps . Even a mains toroid could be used as the output tranny for a prototype , they work well and I'm using one at the moment . This should achieve simplicity and low output impedence , remember loop feedback encourages high order harmonics and phase shifts etc etc etc . If you want to inject 2nd harmonic with this type of circuit , re-jig your line stage with a lower value of anode load . No flames : please note , these are serious constructive comments
316a
Re: ...are you sure
Oh darn! I guess I have to be nice then...
Just kidding. Thanks for the comments 316a. (is that the lower filament current version of yourself? )
Why SE? Just to work on something. I'm actually inclined to agree with you that it's less than ideal for OTL. But I do have to disagree with your run-with-herd commentary on electrolytic output coupling. I just don't hear any of the nastiness others seem to.
Cheers.
Joel
316a said:
Oh darn! I guess I have to be nice then...
Just kidding. Thanks for the comments 316a. (is that the lower filament current version of yourself?
)Why SE? Just to work on something. I'm actually inclined to agree with you that it's less than ideal for OTL. But I do have to disagree with your run-with-herd commentary on electrolytic output coupling. I just don't hear any of the nastiness others seem to.
Cheers.
Joel
electrolytic caps
...have no place in the signal path and belong in the power supply . You can't tell me that a whacking great big cap like this is going to be transparaent sounding ? If so what have you just taken and where can I get some of it ? Also where can I purchase the said transparent sounding big electrolytic ? When I was developing my current SRPP amp I used a 22uF 450V to get the thing working , once this was removed and a 22uF polyprop was fitted , well , you can guess . Remember , no flames or snide comments , please .
316a
...have no place in the signal path and belong in the power supply . You can't tell me that a whacking great big cap like this is going to be transparaent sounding ? If so what have you just taken and where can I get some of it ? Also where can I purchase the said transparent sounding big electrolytic ? When I was developing my current SRPP amp I used a 22uF 450V to get the thing working , once this was removed and a 22uF polyprop was fitted , well , you can guess . Remember , no flames or snide comments , please .
316a
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.