It's interesting to see that several builders on theis thread have been sort of working in parallel - I too had a 6C45P WOT, and also a very vanilla 12B4A as my last two line stages, and the Aikido is looking like it is gong to be the "keeper" so far.
Mine uses 4 x 6SN7 (I had a lot of them around and not too many other dual triodes to choose from in my stash, plus I usually like how they sound) and the single heater supply is raised to 125V using a 400V B+ power supply. My heater supply is 12VDC and I made two series pairs of 6SN7's and then put the two pairs in parallel. This is all so I could use the power supply I am already using for my phono stage and previous line stage, which has the 12VDC. That made the project really easy to get going.
Note that with two heater windings at different DC levels you will want to allocate the tubes "top" and "bottom" rather than "Stage 1" and "Stage 2." This seems to contradict a couple of the Aikido principles, namely using the same tube's halves in one stage to hopefully keep a match as the tube ages, and also, there is likely to be a different tube you'll want to put in the 2nd stage than the tube in the 1st, right? In other words, Aikidos like mine with 2 of the same tube are likely the exception, not the rule.
So what I'm saying is you'd be more likely to want to allocate the tubes "stage 1" and "stage 2" than "top" and "bottom."
I guess you could also allocate the tubes "Stage 1 left and right top," "Stage 1 left and right bottom" . . . . share a tube between channels so as to preserve the ability to roll different triodes in each stage . . . .
Another comment: This low parts count circuit is SO easy to build point-to-point, I'm really surprised there is this much interest in a PCB for it.
-j
Mine uses 4 x 6SN7 (I had a lot of them around and not too many other dual triodes to choose from in my stash, plus I usually like how they sound) and the single heater supply is raised to 125V using a 400V B+ power supply. My heater supply is 12VDC and I made two series pairs of 6SN7's and then put the two pairs in parallel. This is all so I could use the power supply I am already using for my phono stage and previous line stage, which has the 12VDC. That made the project really easy to get going.
Note that with two heater windings at different DC levels you will want to allocate the tubes "top" and "bottom" rather than "Stage 1" and "Stage 2." This seems to contradict a couple of the Aikido principles, namely using the same tube's halves in one stage to hopefully keep a match as the tube ages, and also, there is likely to be a different tube you'll want to put in the 2nd stage than the tube in the 1st, right? In other words, Aikidos like mine with 2 of the same tube are likely the exception, not the rule.
So what I'm saying is you'd be more likely to want to allocate the tubes "stage 1" and "stage 2" than "top" and "bottom."
I guess you could also allocate the tubes "Stage 1 left and right top," "Stage 1 left and right bottom" . . . . share a tube between channels so as to preserve the ability to roll different triodes in each stage . . . .
Another comment: This low parts count circuit is SO easy to build point-to-point, I'm really surprised there is this much interest in a PCB for it.
-j
True..but I have never built anything this complex and the resistor density in mine is ....well quite dense
An externally hosted image should be here but it was not working when we last tested it.
the output 0,47uF polyprops went out after the first day...ragged highs...so I put in some Auricaps...and the ragged highs are gone now.
Just a comment on the Aikido operation and tube half allocation.
It is important that the two halfs of each twin triode are allocated top and bottom of each stage... the Aikido effect is a balancing of the non-linearities of the triode stages against each other on the top and bottom of each stage - it is this that generates its exceptional linearity... since it is much more likely that the two halfs of a twin triode are matched than the two halfs from different tubes it is important to use them together in each stage.
Of course if you only intend to use one type of twin triode and can guarantee that they are dynamically matched then this doesn't matter...
ciao
James
It is important that the two halfs of each twin triode are allocated top and bottom of each stage... the Aikido effect is a balancing of the non-linearities of the triode stages against each other on the top and bottom of each stage - it is this that generates its exceptional linearity... since it is much more likely that the two halfs of a twin triode are matched than the two halfs from different tubes it is important to use them together in each stage.
Of course if you only intend to use one type of twin triode and can guarantee that they are dynamically matched then this doesn't matter...
ciao
James
Bas - to me, that's what a layout SHOULD look like, it looks like you've built the whole thing with what, about 4 inches of hookup wire, in total? That's one thing I like about this circuit, it's all component leads for the most part.
I just popped the top of mine to snap a picture but now I can't find the digital camera - maybe my wife knows where it is . . . anyway it looks pretty much like yours except I used cheap 1/2 watt R's from radio shack and I have a couple of terminal strips, a selector switch, and a volume control in there.
-j
I just popped the top of mine to snap a picture but now I can't find the digital camera - maybe my wife knows where it is . . . anyway it looks pretty much like yours except I used cheap 1/2 watt R's from radio shack and I have a couple of terminal strips, a selector switch, and a volume control in there.
-j
http://www.tubecad.com/2005/July/15July2005.pdf
latest tubecad blog is about aikido. At the end of the article, John talks about group buy of PCB. Maybe someone here should talk to John and coordinate the attack.
latest tubecad blog is about aikido. At the end of the article, John talks about group buy of PCB. Maybe someone here should talk to John and coordinate the attack.
I mailed John last week asking if we could make a PCB. He may or may not have received the mail (I have not received an answer). The fact is that is seems to be offering to make up a PCB. We might as well mail John individually and close this thread? That would be bummer for our diyAudio member bonny_kjs who has put in a few hours work already.
Could you do that? OTOH...the pcb COULD be a way for him to support his BLOG..the importance of such a great website should not be taken for granted. And we should do all we can to ensure John keeps feeding us with info/education. I don't think it would be good to idea to compete with his PCB. Although it won't make him rich...I would happily accept a surcharge on the PCB to support his blog! Even more interesting would be a PCB for an Aikido phono stage come to think of it! Would be interesting to enter it into the ETF 2005 phono stage shootout!
Done.
To: editor@glass-ware.com
Subject: Aikido PCB Groupbuy
From: **************
Message:
Thanks for the great Aikido schematic. I and many at diyaudio.com are very interested in a PCB groupbuy and have a thread
http://www.diyaudio.com/forums/showthread.php?s=&threadid=60585
discussing such possibility. One member even designed a PCB. Please take a look at the thread, as there are suggestions that you may be interested.
Personally, I am suggesting Loctal sockets and balanced Aikido, as teflon loctal sockets is only about $2.50 each on Ebay and NOS 7N7 and 7AF7 are cheap, in comparison to their Octal cousins.
cc: post at above thread at diyaudio.com
To: editor@glass-ware.com
Subject: Aikido PCB Groupbuy
From: **************
Message:
Thanks for the great Aikido schematic. I and many at diyaudio.com are very interested in a PCB groupbuy and have a thread
http://www.diyaudio.com/forums/showthread.php?s=&threadid=60585
discussing such possibility. One member even designed a PCB. Please take a look at the thread, as there are suggestions that you may be interested.
Personally, I am suggesting Loctal sockets and balanced Aikido, as teflon loctal sockets is only about $2.50 each on Ebay and NOS 7N7 and 7AF7 are cheap, in comparison to their Octal cousins.
cc: post at above thread at diyaudio.com
Anybody know how to calculate the output impedance of these Aikido stages? Is it just rp+Rp||rp+Rp or is that too simple?
Once you know that, designing a phono stage is not too big a deal. Here's what my build-up looks like.
The inp[uts are on the left, the selector at top left, the volume pot at top center, and the output jacks are bottom center & bottom right. The yellow rectangular cap is the power supply noise injection cap: that signal is shared between both channels.
-j
Attachments
Bas,
It is much more linear than using even Gary Pimms CCS. By between 20-60dB more linear...
The CCS represents a perfect load for the triode (at least GPs CCS get as close to a perfect load as humanly possible) so you end up with the triode producing the lowst distortion that a single triode can. However there are still distorton producing mechanisims at work in the triode that the use of a CCS doesn't negate.
JBs 'Aikido' is to balance the non-linearity of one triode against the non-linearity of the second triode in such a way as to leave a very linear net effect. For matched triodes and all oher things being equal 60dBs of additioal distortion removal can be achieved. For the average twin triode better than 20dB is acheived.
ciao
James
It is much more linear than using even Gary Pimms CCS. By between 20-60dB more linear...
The CCS represents a perfect load for the triode (at least GPs CCS get as close to a perfect load as humanly possible) so you end up with the triode producing the lowst distortion that a single triode can. However there are still distorton producing mechanisims at work in the triode that the use of a CCS doesn't negate.
JBs 'Aikido' is to balance the non-linearity of one triode against the non-linearity of the second triode in such a way as to leave a very linear net effect. For matched triodes and all oher things being equal 60dBs of additioal distortion removal can be achieved. For the average twin triode better than 20dB is acheived.
ciao
James
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.