I am no expert on the circuit.. though i have built a few versions. BUT, it depends on what you are trying to achieve: eg. how uch gain, how low an impedance, etc. The first tube pretty much dictates the gain, and dominates the tonality and response over the second section, so linearity is good there, and low internal noise and low microphonics. Ti will also dictate the gain for the most part, so for high gin, pick hi mu tubes like 12ax7, or a pentode.. it is a totem amp, so you'll get roughly 1/2 the mu of the input tube as the overall gain.
For the second stage, which is a balanced totem pole, use SRPP logic to figure that one out. the top element is the cathode follower, and the bottom is the pseudo CCS, and the PSU noise input. Low noise and microphonics are important there too, but maybe not quite as critical as the input stage. This stage is the impedance buffer, so the tube will dictate the Zo of the total circuit. Also, the bottom triode sees half the PSU noise, so the gain has to be such t hat it amplifies that properly to get the desired noise-nulling effect.
hope that helps, maybe someone else could expound a bit.
For the second stage, which is a balanced totem pole, use SRPP logic to figure that one out. the top element is the cathode follower, and the bottom is the pseudo CCS, and the PSU noise input. Low noise and microphonics are important there too, but maybe not quite as critical as the input stage. This stage is the impedance buffer, so the tube will dictate the Zo of the total circuit. Also, the bottom triode sees half the PSU noise, so the gain has to be such t hat it amplifies that properly to get the desired noise-nulling effect.
hope that helps, maybe someone else could expound a bit.
I've been watching some of the Aikido postings with interest too and was wondering what everyone's favorite tube-compliment choices were, that they felt gave them the best sound?
(I don't want to hijack anyone else's thread, but I've been wondering this similar question myself and it seems superfluous to start a speperate thread for something so close)
(I don't want to hijack anyone else's thread, but I've been wondering this similar question myself and it seems superfluous to start a speperate thread for something so close)
Hi,
dr._sleep wrote:
"I am currently running RCA clear top 6FQ7/6CG7 's as input tubes into
cryo'd Tung Sol 5687's, very nice."
I think that's a good choice (although I believe using the same type
for both positions can be quite satisfactory too). I'm running 7E6
(what?) into 6N30P (aka 6H30P), which is basically a pretty similar
lineup. The 7E6 is a close relative to 6J5 which is half a 6SN7 which
is an octal 6FQ7 (pretty much). And the 6N30P is much like a
higher-transconductance 5687.
I'd say using a valve with small paracitic capacitances for the first
spot is a good idea, as it will typically be preceded by a volume pot
or attenuator. Hence an increased source resistance facing the first
valve, so small miller effect is good (i.e. in practice medium-mu, low
current valve) for good bandwidth. And it doesn't hurt (if you want
a "neutral" tone) that this family of valves are very linear. (Another
obscure valve to try here would be the tiny 9002, which at least
looks good on paper.)
Using a more powerful valve for the CF/noise-reduction stage isn't
really necessary for normal cable runs, but if you want to plug a
high-impedance headphone directly into the output, this will work
with a 5687 or 6N30P running at 20+ milliamps - but not so well
with a "lesser" valve. I designed my Aikido with headphones in
mind, and it works a charm. Of course, if using an OPT good current
capability is somewhat less important, but I still think a chunky
output valve is a good idea for this too.
BTW, I did the opposite of cryogenics to my 6N30Ps: I put them in
the oven at 150 degrees C for 24 hours. Seems to have helped
with microphony, which can be bad with some exemplars. But
this could be coincidental. Oven baking can also reduce bad cases
of blue glow (gas) if you're lucky.
Morgan L
dr._sleep wrote:
"I am currently running RCA clear top 6FQ7/6CG7 's as input tubes into
cryo'd Tung Sol 5687's, very nice."
I think that's a good choice (although I believe using the same type
for both positions can be quite satisfactory too). I'm running 7E6
(what?) into 6N30P (aka 6H30P), which is basically a pretty similar
lineup. The 7E6 is a close relative to 6J5 which is half a 6SN7 which
is an octal 6FQ7 (pretty much). And the 6N30P is much like a
higher-transconductance 5687.
I'd say using a valve with small paracitic capacitances for the first
spot is a good idea, as it will typically be preceded by a volume pot
or attenuator. Hence an increased source resistance facing the first
valve, so small miller effect is good (i.e. in practice medium-mu, low
current valve) for good bandwidth. And it doesn't hurt (if you want
a "neutral" tone) that this family of valves are very linear. (Another
obscure valve to try here would be the tiny 9002, which at least
looks good on paper.)
Using a more powerful valve for the CF/noise-reduction stage isn't
really necessary for normal cable runs, but if you want to plug a
high-impedance headphone directly into the output, this will work
with a 5687 or 6N30P running at 20+ milliamps - but not so well
with a "lesser" valve. I designed my Aikido with headphones in
mind, and it works a charm. Of course, if using an OPT good current
capability is somewhat less important, but I still think a chunky
output valve is a good idea for this too.
BTW, I did the opposite of cryogenics to my 6N30Ps: I put them in
the oven at 150 degrees C for 24 hours. Seems to have helped
with microphony, which can be bad with some exemplars. But
this could be coincidental. Oven baking can also reduce bad cases
of blue glow (gas) if you're lucky.
Morgan L
Rather than start another thread, I am also looking at the choices for a 9 pin Aikido. This is a present for my brother, who has been into tubes for the last 30 years. Simple line amp, moderately long cables @ 15 ft / 5 meters.
I have on hand:
8 6H23pi
8 6H6pi
16 6N1p
4 6H30 (for another project)
4 5687 (for another project)
16 12AX7
4 12AU7
The 6H23pi / 6DJ8 / 6922 seems at home in position 2.
Position 1, I was thinking of the 6H6pi, because of their Mu of 22.
The gain of ~ 11 would be OK.
My other choice would be to buy a couple of 6CG7/6FQ7.
My problems more ascetic. I would use a 6FQ7, but there isn't a cheep Russian equivalent I am aware of, and it seems wrong to use a lower RP tube such as the 6H6pi in slot 1.
Any comments welcome.
Doug
I have on hand:
8 6H23pi
8 6H6pi
16 6N1p
4 6H30 (for another project)
4 5687 (for another project)
16 12AX7
4 12AU7
The 6H23pi / 6DJ8 / 6922 seems at home in position 2.
Position 1, I was thinking of the 6H6pi, because of their Mu of 22.
The gain of ~ 11 would be OK.
My other choice would be to buy a couple of 6CG7/6FQ7.
My problems more ascetic. I would use a 6FQ7, but there isn't a cheep Russian equivalent I am aware of, and it seems wrong to use a lower RP tube such as the 6H6pi in slot 1.
Any comments welcome.
Doug
Hi,
Current thoughts have been towards John Broskie's circuit - but it seems strange that tubes providing dis-similar pairs have not been considered.
If one considers this type of tube, and looks at the circuit as "upstairs" and "down-stairs", then it is very easy to acknowledge the filament-cathode limitation and use Morgan Jones' "thingy" circuit for the top-tube. It just may be that separation between channels will improve...
I have been looking at the 6CM7 - got that that 6J5/6SN7 half for the input, then a lower impedance second-half if you are pursuing drive. 6GF7 is a starter also.....
Currently playing with HT regulators/snubbers. Previously I have built a clone of the SP10, but I doubled the regulators - each tube enjoying its own......channel-wise. Seems more correct somehow......
I am truly a fan of Audio Research - and that site that provides glimpses of circuitry. Pondering over the SP10 and SP11 circuitry where the series HT regulator went from tube to mosfet. Appears that there is a school of thought that the mosfet-circuit "rings" and results in a "steely" sound. Love to share comment in this regard......
Graeme
Current thoughts have been towards John Broskie's circuit - but it seems strange that tubes providing dis-similar pairs have not been considered.
If one considers this type of tube, and looks at the circuit as "upstairs" and "down-stairs", then it is very easy to acknowledge the filament-cathode limitation and use Morgan Jones' "thingy" circuit for the top-tube. It just may be that separation between channels will improve...
I have been looking at the 6CM7 - got that that 6J5/6SN7 half for the input, then a lower impedance second-half if you are pursuing drive. 6GF7 is a starter also.....
Currently playing with HT regulators/snubbers. Previously I have built a clone of the SP10, but I doubled the regulators - each tube enjoying its own......channel-wise. Seems more correct somehow......
I am truly a fan of Audio Research - and that site that provides glimpses of circuitry. Pondering over the SP10 and SP11 circuitry where the series HT regulator went from tube to mosfet. Appears that there is a school of thought that the mosfet-circuit "rings" and results in a "steely" sound. Love to share comment in this regard......
Graeme
I just finished putting my Aikido together.
I use Brimar 12AU7 for the first stage (1K ohm cathode resistance)and Tungsol 5687 for the buffer (470 ohm cahtode resistance). The PSU is tube regulated B+ (HP400's design), and the 12Vdc is regulated with current limiting). PSU is separate from the line amp. I run this line amp at +300Vdc. A 0.22uF film capacitor is placed on each PCB between B+ and ground for better treble.
Output capacitor is 1.0uF MIT Cap.
I removed those B+ noise cancellation components because I believe my PSU is clean enough. Also, I want to keep the circuitry clean and simple.
The high frequency response is approx 250kHz at -3dB. Slightly rounded at the top of the rising edge of the 10kHz square wave waveform. The total gain is approx 8 times.
Sounds pretty good with very strong bass, warm vocals and rather clear highs. I run it with my low gain 300B monos - one stage 6J5 drives 300B.
Johnny
I use Brimar 12AU7 for the first stage (1K ohm cathode resistance)and Tungsol 5687 for the buffer (470 ohm cahtode resistance). The PSU is tube regulated B+ (HP400's design), and the 12Vdc is regulated with current limiting). PSU is separate from the line amp. I run this line amp at +300Vdc. A 0.22uF film capacitor is placed on each PCB between B+ and ground for better treble.
Output capacitor is 1.0uF MIT Cap.
I removed those B+ noise cancellation components because I believe my PSU is clean enough. Also, I want to keep the circuitry clean and simple.
The high frequency response is approx 250kHz at -3dB. Slightly rounded at the top of the rising edge of the 10kHz square wave waveform. The total gain is approx 8 times.
Sounds pretty good with very strong bass, warm vocals and rather clear highs. I run it with my low gain 300B monos - one stage 6J5 drives 300B.
Johnny
I brought my Aikido Amp to my friend's house today. It was compared with his DIY Marantz M7 line amp. It is very obvious that the sound of Aikido is too compressed and dull.
When I get back home, I added a 100uF BG capacitor to bypass the bottom cahtode resistor of 12AU7. Now it has more dynamic and bright.
I know that now I have deviated too far from the original design. However, I believe it's the fun being a DIYer to experiement different things in order to make the amp sound right to my ears.
Have fun,
Johnny
When I get back home, I added a 100uF BG capacitor to bypass the bottom cahtode resistor of 12AU7. Now it has more dynamic and bright.
I know that now I have deviated too far from the original design. However, I believe it's the fun being a DIYer to experiement different things in order to make the amp sound right to my ears.
Have fun,
Johnny
hi all,
I also like the akido pre-amp much. After several test.
I like the 12sn7+ 5687 cathode follower with 1:1 interstage output pre-amp.
since the 5687 cathode follower was very low plate resistance.
Use 1:1 will lower more the output impendance.
for example the ratio was 5687=1K.
5687 cathode follower 1KX0.9=900ohm.
use 1:1 will had little lower than 1:1.
900Ohm X 0.9 ( approx Value)= output.
so I love this more.
BTW, Since the pre-amp will pass very small signal & high fidelity
OPT was extreme important. I was testing use 5N SILVER WIRE which was monocrystalline & 4 Layers polyurthame coat wire.
with Nano amorphous core for this pre-amp OPT.
I will let U know the performance later.
thx
thomas
I also like the akido pre-amp much. After several test.
I like the 12sn7+ 5687 cathode follower with 1:1 interstage output pre-amp.
since the 5687 cathode follower was very low plate resistance.
Use 1:1 will lower more the output impendance.
for example the ratio was 5687=1K.
5687 cathode follower 1KX0.9=900ohm.
use 1:1 will had little lower than 1:1.
900Ohm X 0.9 ( approx Value)= output.
so I love this more.
BTW, Since the pre-amp will pass very small signal & high fidelity
OPT was extreme important. I was testing use 5N SILVER WIRE which was monocrystalline & 4 Layers polyurthame coat wire.
with Nano amorphous core for this pre-amp OPT.
I will let U know the performance later.
thx
thomas
An externally hosted image should be here but it was not working when we last tested it.
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.