Yes, the Aiikido output stage is a current sourced cathode follower. Which is a bit different than a White cathode follower, in that both the top and bottom tubes are biased identically, thus ensuring 50% B+ across each tube. This is a feature that permits noise cancellation in the Aikido topology.
I used a simple CLC supply on my "first" preamp attempt - 15uf/30H/30uf. Considered choke input, but worried about some stories that "swinging" ones are necessary, and that ordinary chokes will "buzz". Even if true with heavy draws, I doubt that any choke, used at rated current, would give trouble at preamp loads. Any thoughts on this? Bill.
Hi,
Guilty as charged...
First of all, this is not meant as a "universal" 300V series reg. but rather a circuit optimized for what I wanted it to do in the context of the phono preamp it was designed for.
Second, it's been a while since I worked on it so, bear with me...
Response time was better this way. At the expense of regulation, so all in all it's a bit of a trade off.
Keep in mind that the 12AX7A wasn't really designed for cascode service either...
That too had something to do with it, I just don't remember what it was exactly.
Excellent. You figured it out quite well.
The idea was to obtain penthode like gain figures from the cascode(-ish) circuit without having to source some expensive and obsolete low noise penthode.
The only twin triode I know of that betters the 12AX7A in this respect is the unobtainable (or almost anyway) Brimar ECC807.
That one has a mu of about 150 IIRC and an odd pinout as well.
It was used by Rogers in some preamps, they often had some Brimar odd-balls in their designs.
You may find some single triodes, even a nuvistor, I think with a mu above the 100 mark but I digress...
As for the exact gain....it was formidable but I can't seem to recall the exact figure.
Some of the ideas that went into the regulator were "borrowed" from some of Tim de Paravicini's work on professional studio preamps and the famous Studer tapedecks, not really his inventions but not often seen in the production of commercial gear anyway.
Other than that, you may have some luck finding some similar ideas in old (1960-1980) publications of British Electronics World magazine and similar weeklys.
Those large caps are absolute not essential for the reg to work properly.
They were the best caps we could find at the time, low ESR and all but I'm sure you could improve the preamp with better and smaller values: MKP a la Solen or Unlytic even BGs if you're so inclined.
Again, within the context of the preamp, those seemingly hugue caps are there so the circuit proper is running from a virtual battery supply.
Feel free to experiment. I recall I did and that's what I ended up with as best in my system. (Subjective, I know)
That circuit which is an integral part of the phono preamp, is a White follower with plate stopper (avant la lettre if I may say so), reminescent of a cascode follower.
The Aikido circuit which should be viewed as a gain block with optimized PSR works by combining the SRPP (note that the signal is DC coupled from the plate of the lower tube, not the cathode of the upper one, to obtain a better voltage division) as a gain block cum voltage divider followed by a White CF plus resistive voltage divider in a FB loop resulting in optimized PS noise rejection, a gain of the mu of the first triode divided by a factor two and a very low Zout after the WCF.
As a block the circuit has much improved PSRR compared to just the SRPP and WCF alone. The PSRR being the achilles heel of most totem pole circuits.
The theory behind it can be found at John Broskie's excellent site as I'm sure you know.
Read his expose on the WCF first, his optimization of and...if you don't have a splitting headache by then, read about SRPP ("decontructed" or similar title)....then about noise reduction techniques for SE circuits.....You get the picture.
Cheers,
Guilty as charged...
First of all, this is not meant as a "universal" 300V series reg. but rather a circuit optimized for what I wanted it to do in the context of the phono preamp it was designed for.
Second, it's been a while since I worked on it so, bear with me...
Response time was better this way. At the expense of regulation, so all in all it's a bit of a trade off.
Keep in mind that the 12AX7A wasn't really designed for cascode service either...
That too had something to do with it, I just don't remember what it was exactly.
Excellent. You figured it out quite well.
The idea was to obtain penthode like gain figures from the cascode(-ish) circuit without having to source some expensive and obsolete low noise penthode.
The only twin triode I know of that betters the 12AX7A in this respect is the unobtainable (or almost anyway) Brimar ECC807.
That one has a mu of about 150 IIRC and an odd pinout as well.
It was used by Rogers in some preamps, they often had some Brimar odd-balls in their designs.
You may find some single triodes, even a nuvistor, I think with a mu above the 100 mark but I digress...
As for the exact gain....it was formidable but I can't seem to recall the exact figure.
Some of the ideas that went into the regulator were "borrowed" from some of Tim de Paravicini's work on professional studio preamps and the famous Studer tapedecks, not really his inventions but not often seen in the production of commercial gear anyway.
Other than that, you may have some luck finding some similar ideas in old (1960-1980) publications of British Electronics World magazine and similar weeklys.
Those large caps are absolute not essential for the reg to work properly.
They were the best caps we could find at the time, low ESR and all but I'm sure you could improve the preamp with better and smaller values: MKP a la Solen or Unlytic even BGs if you're so inclined.
Again, within the context of the preamp, those seemingly hugue caps are there so the circuit proper is running from a virtual battery supply.
Feel free to experiment. I recall I did and that's what I ended up with as best in my system. (Subjective, I know)
That circuit which is an integral part of the phono preamp, is a White follower with plate stopper (avant la lettre if I may say so), reminescent of a cascode follower.
The Aikido circuit which should be viewed as a gain block with optimized PSR works by combining the SRPP (note that the signal is DC coupled from the plate of the lower tube, not the cathode of the upper one, to obtain a better voltage division) as a gain block cum voltage divider followed by a White CF plus resistive voltage divider in a FB loop resulting in optimized PS noise rejection, a gain of the mu of the first triode divided by a factor two and a very low Zout after the WCF.
As a block the circuit has much improved PSRR compared to just the SRPP and WCF alone. The PSRR being the achilles heel of most totem pole circuits.
The theory behind it can be found at John Broskie's excellent site as I'm sure you know.
Read his expose on the WCF first, his optimization of and...if you don't have a splitting headache by then, read about SRPP ("decontructed" or similar title)....then about noise reduction techniques for SE circuits.....You get the picture.
Cheers,
Hi,
Unless I'm mistaken most of the chokes for use in tube circuits you commonly find at most tube vendors actually are "swinging" chokes.
It's the plate chokes that need much more care in execution as these have to be, among other things, wide BW.
Something PS chokes don't have to be.
Such chokes you probably only find at specialized OPT winders, in essence it's not all that more complicated than a SE OPT really, such a Sowter in the UK, Bertolucci in Italy and similar sources.
To me at least, it would be easier if PS inductors would just be called "chokes" (which is what they do: to choke") and plate load inductors be called, well, plate load inductors for lack of a better term.
Cheers,
Unless I'm mistaken most of the chokes for use in tube circuits you commonly find at most tube vendors actually are "swinging" chokes.
It's the plate chokes that need much more care in execution as these have to be, among other things, wide BW.
Something PS chokes don't have to be.
Such chokes you probably only find at specialized OPT winders, in essence it's not all that more complicated than a SE OPT really, such a Sowter in the UK, Bertolucci in Italy and similar sources.
To me at least, it would be easier if PS inductors would just be called "chokes" (which is what they do: to choke") and plate load inductors be called, well, plate load inductors for lack of a better term.
Cheers,
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.