I guess these circuits with fixed voltage references are technically
not gyrators. Gyrators have a fixed voltage drop which must be
greater than the AC swing needed. The example gyrator in the first
post of this thread uses a voltage divider between Vin and Vout
to set the gate voltage midway, which makes the total voltage drop
2 * the Vgs threshold voltage. This sort of gyrator is used to remove
ripple from a power supply. It can also be used in the anode circuit
of an amplifier if the fixed voltage drop is made larger by increasing
the ratio of the resistors.
The fixed voltage circuits, above and in the other thread, are for
operating a tube with a fixed anode voltage but high AC impedance.
Sort of like a gyrator and regulator in one.
Calling them gyrators may be a bit misleading, as the applications
are not exactly the same.
So what shall we call these things that are sort of like gyrators but
with fixed (or adjustable...) DC voltage? What about the mu-follower
and symmetric (anti-triode) variants? Is it just a constant voltage
gyrator (mu-follower, anti-triode)?
Michael
not gyrators. Gyrators have a fixed voltage drop which must be
greater than the AC swing needed. The example gyrator in the first
post of this thread uses a voltage divider between Vin and Vout
to set the gate voltage midway, which makes the total voltage drop
2 * the Vgs threshold voltage. This sort of gyrator is used to remove
ripple from a power supply. It can also be used in the anode circuit
of an amplifier if the fixed voltage drop is made larger by increasing
the ratio of the resistors.
The fixed voltage circuits, above and in the other thread, are for
operating a tube with a fixed anode voltage but high AC impedance.
Sort of like a gyrator and regulator in one.
Calling them gyrators may be a bit misleading, as the applications
are not exactly the same.
So what shall we call these things that are sort of like gyrators but
with fixed (or adjustable...) DC voltage? What about the mu-follower
and symmetric (anti-triode) variants? Is it just a constant voltage
gyrator (mu-follower, anti-triode)?
Michael
Michael Koster said:
My understanding is that to meet the original definition the circuit has to work both ways , put a cap on the output port and the input looks like an inductor , swap in an inductor and the input looks like a cap.
Wavebourn said:
Maybe it's just because I'm a rank beginner and struggling to remember what's what, but I think it's important that the name gives a description of what distinguishes the thing being named from any other. If the important characteristics are the stable voltage and high AC impedance isn't that what you want the name to reflect? Is it really a current source ?
Michael Koster [/i][B] Here's what I would find useful as a modular bit of circuit . . . . . Cheers said:
The amp is already breadboarded in monoblocs , cathode bias on both tubes and cap coupled. I had thought it would be good to direct couple but with CCS loads the plate voltages of the 417A's I have are all over the map. I was pretty happy when I saw your circuit as I think it may be the perfect answer.
I'm happy to try pretty much anything but as you say, with direct coupling I'll need more B+ so it makes sense to start with the present cap coupled set-up first and then move up from there. I'd certainly like to try any ideas you suggest.
Thanks Guys!!
Hearinspace said:
It depends.
A guy living in Siberian forest knows well beasts, birds, trees, flowers, by name, and knows well their unique characters. When he goes from one to another point in the forest he does not need to name each and every path through the bush, between trees.
A guy living in the American town coming to Siberian forest have to know names of each path so he can come to one point to another without being afraid to be lost.
Even famous and legendary hunters from towns who are coming to the forest to get meat, skins, honor, need to know names of roads.
But native Aborigines don't need to name each and every path in the wood. So, when some hunter say, "Hey, I went that path 40 years ago! You have to learn from me!", it sounds funny for a Siberian Aborigine. He can choose another path any time he goes, depending on circumstances...
I am not famous... I am not a hunter... I just live here, breath this beautiful electronic air... And wish to everyone to enjoy it for free.
I would still call it a SVCS: it is a voltage stabilizer on DC, but a current source on AC. I prefer descriptive names that reflect functions.
Thanks Anatolly! I'm really stoked by this. Большое спасибо !!
OK, now I have time enough to pester you a bit. 
Can you tell me something about the circuit choices you made? First, I assume the 5731A is there because I mentioned I'd been thinking of using it. But the only reason I had them was that I got them a few years ago as (possibly) improved replacements for MJE350's in an older CCS circuit. They do sound better than the 350's but it wouldn't surprise me if there's a better sounding alternative by now AND as you're depth of knowledge is so much greater I can't let it go without asking, if starting from zero, what other transistor (or other device) would you have chosen?
The other question right off the bat is why enhancement mode Mosfets above and specifically why the IRF9521 ? Again, not criticism just wanting to understand the thinking behind the choice.
Thanks Wavebourn !
Can you tell me something about the circuit choices you made? First, I assume the 5731A is there because I mentioned I'd been thinking of using it. But the only reason I had them was that I got them a few years ago as (possibly) improved replacements for MJE350's in an older CCS circuit. They do sound better than the 350's but it wouldn't surprise me if there's a better sounding alternative by now AND as you're depth of knowledge is so much greater I can't let it go without asking, if starting from zero, what other transistor (or other device) would you have chosen?
The other question right off the bat is why enhancement mode Mosfets above and specifically why the IRF9521 ? Again, not criticism just wanting to understand the thinking behind the choice.
Thanks Wavebourn !
Transistors here must not sound, they must regulate. There is a broad range of available devices, and their regimes would depend on the device used.
Why your favorite transistor?
Because you have it, and it will work well in the application.
Why enhancement mode MOSFETs?
Because it was simpler to use them here. Their G-S cut-off voltages summed determine a reference that will be used to compare a voltage drop across the cascode. Since they vary from device to device, you have the pot that will be used as well to vary a voltage on your tube's anode to to make feel happy it and another one that it drives.
Why IRF9521? Because I have it right now in front of me on my laptop; it has low capacitances and more than plenty of volts and amperes specified by the manufacturer for this application.
The top one I've found searching a datasheet catalog online, it again has optimal parameters for your application, plus a good safety margin.
Zener string? Again, you may use Zeners available for you; I selected Zeners again for optimal price/performance ratio.
There are many ways to skin the same cat. Those who exploit the single topology struggling to find the best parts for it are called "Legendary Designers". I am not. For different parts different topologies would be more optimal, or different regimes in the same topology. For different regimes different parts would be optimal. And so on. I am free to vary parts, topologies, regimes, in order to get end results I need. I could use more transistors or less of them. I could arrange them differently to get similar results. But before designing something I carefully define end results, limits, conditions. No problem can be solved without the comprehensive conditions given. That's why I was asking you about voltages, transistors, currents, tubes.
Why your favorite transistor?
Because you have it, and it will work well in the application.
Why enhancement mode MOSFETs?
Because it was simpler to use them here. Their G-S cut-off voltages summed determine a reference that will be used to compare a voltage drop across the cascode. Since they vary from device to device, you have the pot that will be used as well to vary a voltage on your tube's anode to to make feel happy it and another one that it drives.
Why IRF9521? Because I have it right now in front of me on my laptop; it has low capacitances and more than plenty of volts and amperes specified by the manufacturer for this application.
The top one I've found searching a datasheet catalog online, it again has optimal parameters for your application, plus a good safety margin.
Zener string? Again, you may use Zeners available for you; I selected Zeners again for optimal price/performance ratio.
There are many ways to skin the same cat. Those who exploit the single topology struggling to find the best parts for it are called "Legendary Designers". I am not. For different parts different topologies would be more optimal, or different regimes in the same topology. For different regimes different parts would be optimal. And so on. I am free to vary parts, topologies, regimes, in order to get end results I need. I could use more transistors or less of them. I could arrange them differently to get similar results. But before designing something I carefully define end results, limits, conditions. No problem can be solved without the comprehensive conditions given. That's why I was asking you about voltages, transistors, currents, tubes.
Hearinspace said:
I don't know. I have a bag of them from Supertex. http://www.datasheetcatalog.org/datasheet/supertex/IRF9521.pdf
Search for something with less input capacitances; high breakdown voltages are not needed (actually, 60V is more than plenty since about 10V maximum will be between source and drain), maximal current up to 2A. I am sure you will find some. If it's cut-off voltage will be different we always can recalculate values of resistors, but as designed tolerances are wide like barn gates (everything adjusts one pot), so don't worry.
About other devices, and some possible improvements of this draft, let's wait for Kenpeter, Michael Koster, or other people to contribute their ideas.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.