Your choice of EF86 over a 6AU6 is a poor decision. Technically the EF86 maybe a superior audio pentode, but it is a low current one which will struggle to drive the low impedance feedback node without distorting. A 6AU6 is more than capable of handling a standing current of 10mA which will make its output signal a relatively small proportion of the standing current and ensure the least variation in the gm and hence the most linear output signal.
The more standing current the better and even the 6AU6 could be considered marginal in this respect.
Shoog
The more standing current the better and even the 6AU6 could be considered marginal in this respect.
Shoog
Shoog,
Two questions. First, SV83 is a more sensitive tube to drive than EL84. I don't know enough to know if this will make a difference in your objection to EF86? Second. If I decrease the local feedback and add a global feedback loop would that fix the problem?
And M Greg, in practice is Shoogs above objection actually a problem?
Two questions. First, SV83 is a more sensitive tube to drive than EL84. I don't know enough to know if this will make a difference in your objection to EF86? Second. If I decrease the local feedback and add a global feedback loop would that fix the problem?
And M Greg, in practice is Shoogs above objection actually a problem?
Last edited:
Hey Mach - when was the last time you actually designed a plate to grid amplifier ?Even one more reason why it shouldn't work according to the self appointed cognoscenti.
Shoog
It would do and it looks a reasonably nice set of curves. However you then have an unused triode hanging around.
I have had no issues with using a 6AU6 - but my amps have all been PP so its not totally applicable. There are very few ideal candidates out there really and that is why the 6AU6 keeps coming up.
Shoog
Last edited:
CCS?
artsalo,
I sold those output transformers to a guy I found in the swap meet forum. I've already ordered a new pair. The new ones will be 18 watt, 150ma with a 2.8k primary and a 6 ohm secondary. And my power transformer is good for 250ma.
I personally, would like to get beyond the RH without adding too much complexity. I would also like to use as little feedback as possible in future projects. That is why I'm wondering about the feedback?
Every RH looks roughly the same. A preamp tube front and center and two outputs on either side. I'd like to see more balance, visually. So to that end four tubes, two on each side. The purpose would be to use the second half of each triode as a CCS as either a supply or a sink? I know that a CCS would reduce distortion. But would it also flatten frequency response? And if so what would that do to the damping? artsalo you mentioned at the beginning of this thread that there isn't enough feedback in the circuit to properly damp the bass driver. I don't understand how feedback affects damping, yet. But would damping then become even more a problem? The whole idea is to build an amp that sounds good without giving up any more gain than absolutely necessary?
artsalo,
I sold those output transformers to a guy I found in the swap meet forum. I've already ordered a new pair. The new ones will be 18 watt, 150ma with a 2.8k primary and a 6 ohm secondary. And my power transformer is good for 250ma.
I personally, would like to get beyond the RH without adding too much complexity. I would also like to use as little feedback as possible in future projects. That is why I'm wondering about the feedback?
Every RH looks roughly the same. A preamp tube front and center and two outputs on either side. I'd like to see more balance, visually. So to that end four tubes, two on each side. The purpose would be to use the second half of each triode as a CCS as either a supply or a sink? I know that a CCS would reduce distortion. But would it also flatten frequency response? And if so what would that do to the damping? artsalo you mentioned at the beginning of this thread that there isn't enough feedback in the circuit to properly damp the bass driver. I don't understand how feedback affects damping, yet. But would damping then become even more a problem? The whole idea is to build an amp that sounds good without giving up any more gain than absolutely necessary?
A couple general questions if I haven't tried your patience too much. For Schade type feedback am I right in assuming that the coupling cap and the feedback resistor form an RC network? And is there an inherent benefit over conventional GFB?
And the voltage regulator in place of the cathode resistor? I have a very limited understanding of transistors and solid state in general. But LT1086, which was suggested I use, sets current by maintaining the voltage dropped across the device, like a cathode resistor, rather than by regulating current directly is that right? If that is the case as B+ rises and falls, (from the wall outlet) won't current also change? And if that is the case wouldn't it be more effective to regulate B+ and then just use a cathode resistor? But that wouldn't give individual automatic bias adjustment for each tube would it? So the best plan would be to regulate both?
I apologize in advance if this is too many questions.
Kevin
And the voltage regulator in place of the cathode resistor? I have a very limited understanding of transistors and solid state in general. But LT1086, which was suggested I use, sets current by maintaining the voltage dropped across the device, like a cathode resistor, rather than by regulating current directly is that right? If that is the case as B+ rises and falls, (from the wall outlet) won't current also change? And if that is the case wouldn't it be more effective to regulate B+ and then just use a cathode resistor? But that wouldn't give individual automatic bias adjustment for each tube would it? So the best plan would be to regulate both?
I apologize in advance if this is too many questions.
Kevin
I can see no benefit in using a voltage regulator over a simple resistor to set the bias. In both cases what you are listening to is the bypass cap which will dominate the sound. A voltage regulator works when you need to set a very precise current, such as in a PP output stage, but the SE amp see's no benefit at all.
Schade feedback can be done as a direct coupled amplifier - but its not straightforward. This feedback only takes in the output stage (though it places demands on the driver) so it is a true form of local feedback. My experience is that it improves the bass considerably (lower output impedance) and allows a pentode to sound like a triode whilst maintaining its power capability (if driven adequately). None of the deadening associated with over exuberant application of gNFB is experienced. I have applied nearly 100% Schade feedback in some of my designs - without experiencing the loss of detail and life you get with even moderate levels of gNFB. Ultimately though you would have to do the comparison yourself and draw your own conclusions.
Shoog
Schade feedback can be done as a direct coupled amplifier - but its not straightforward. This feedback only takes in the output stage (though it places demands on the driver) so it is a true form of local feedback. My experience is that it improves the bass considerably (lower output impedance) and allows a pentode to sound like a triode whilst maintaining its power capability (if driven adequately). None of the deadening associated with over exuberant application of gNFB is experienced. I have applied nearly 100% Schade feedback in some of my designs - without experiencing the loss of detail and life you get with even moderate levels of gNFB. Ultimately though you would have to do the comparison yourself and draw your own conclusions.
Shoog
SV83 to 6550
Thanks Guys for all the help. I really do appreciate it and I've learned quite a bit from this thread. Enough so that I've found that SV83 is not a good candidate for this design. Mostly because its peak voltage is the same as its max plate voltage 330. So I've decided to change horses in mid-stream.
The original purpose of this project was to build a powerful SE amp without having to resort to very expensive triodes like 330b. So what I have decided to do is build using a single 6550C as the output.
6550 using fixed bias so bias adjustment can be done with a single pot for each tube without having to open the amp.
One thing I learned researching the previous idea is that the only thing that distinguishes RH design is its feed back method. In fact now that I think about it every one I've seen uses the same 470k feedback resistor and the same .22uf coupling cap.
I still want it to be an all pentode amp though so I have to chose a voltage amplifier yet. I have 6AU6, EF86, 6AN8, 7199 and SV83 to chose from. I'l check out the RH universal for ideas on that.
And, as always, thanks for all the help so far!
Thanks Guys for all the help. I really do appreciate it and I've learned quite a bit from this thread. Enough so that I've found that SV83 is not a good candidate for this design. Mostly because its peak voltage is the same as its max plate voltage 330. So I've decided to change horses in mid-stream.
The original purpose of this project was to build a powerful SE amp without having to resort to very expensive triodes like 330b. So what I have decided to do is build using a single 6550C as the output.
6550 using fixed bias so bias adjustment can be done with a single pot for each tube without having to open the amp.
One thing I learned researching the previous idea is that the only thing that distinguishes RH design is its feed back method. In fact now that I think about it every one I've seen uses the same 470k feedback resistor and the same .22uf coupling cap.
I still want it to be an all pentode amp though so I have to chose a voltage amplifier yet. I have 6AU6, EF86, 6AN8, 7199 and SV83 to chose from. I'l check out the RH universal for ideas on that.
And, as always, thanks for all the help so far!
Last edited:
Consider the inexpensive and good Russian 6ж9п (a.k.a 6Z9P and 6J9P) as a driver, and consider using fixed bias in the 6550 based output stage, gets rid of an LF pole in the response and eliminates all of the effects of a less than ideal capacitor in that location. (I no longer use cathode bias in anything..)
Since this is moving away from the RH conceptually perhaps you'd like to rename the thread? If so send me a PM..
Since this is moving away from the RH conceptually perhaps you'd like to rename the thread? If so send me a PM..
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.