Hi all again,
I tested yesterday my Aikido on a different setup (not my hedlunds but a freind's one with Dalquist loudspeakers) and I must say that the pre is silent. The only way to hear hum is putting the ear <2cm close to the LF units (not a natural way of enjoy music...). And this result is with cheap tubes, I must invest in best quality tubes, I know!
I'm not sure about improving the noise with the extra RC network Bas suggested me or leave it as is until I get a more decent source selector and pot. The result of the test is:
a) my friend is amazed; b) He wants his Aikido.... so: have you Bas more boards for him?
Regards.
JL
I tested yesterday my Aikido on a different setup (not my hedlunds but a freind's one with Dalquist loudspeakers) and I must say that the pre is silent. The only way to hear hum is putting the ear <2cm close to the LF units (not a natural way of enjoy music...). And this result is with cheap tubes, I must invest in best quality tubes, I know!
I'm not sure about improving the noise with the extra RC network Bas suggested me or leave it as is until I get a more decent source selector and pot. The result of the test is:
a) my friend is amazed; b) He wants his Aikido.... so: have you Bas more boards for him?
Regards.
JL
Think I have run out of octals. But still have novals. Might have to order some more octals...I'll count tonight.
Should the 6x5 heater be biased up just like the signal tubes?
I am aiming for a 300V B+, with a 80V bias from the voltage divider.
And, how do I connect the 80V bias to the 6.3V 3A heater winding, anyway?
I am aiming for a 300V B+, with a 80V bias from the voltage divider.
And, how do I connect the 80V bias to the 6.3V 3A heater winding, anyway?
Ok...the lack of a center tap for the secondaries on the transformer I want to pick up threw me. Thanks for the explanation on how to create the virtual center tap.
Which is truely "better" in terms of overall sonics?
1) Lundahl 1683 w/ 6.3V heaters and a virtual center tap
2) Hammond 6.3V CT transformer
Which is truely "better" in terms of overall sonics?
1) Lundahl 1683 w/ 6.3V heaters and a virtual center tap
2) Hammond 6.3V CT transformer
Theoretically a seperate filament transformer is the best. The reason being less capacitative coupling of noise from one secondary to the other.
In practice it won't be audible I think.
In practice it won't be audible I think.
Got it. I think I'll start with the virtual CT. I can always add separate transformers later...
Now: I've been reading the manual for Broskie's PCB boards, and he has 10-100uF capacitors shunting the heaters to ground.
I can't tell if this is in leu of the bias or not....and if not, what is the purpose?
Now: I've been reading the manual for Broskie's PCB boards, and he has 10-100uF capacitors shunting the heaters to ground.
I can't tell if this is in leu of the bias or not....and if not, what is the purpose?
I only see 0,1uF's on the heaters. (C3/C4)
That is a way to not have to use a upward in voltage referenced heater supply. It works well. Often a bigger value like 0,47uF will work even better. (Once experienced that with a Mu-Follower where the top tube was "floated") One could hear the hum go down through the speakers as I was paralleling the caps.
So if you use that method no resistors and big capacitor combination is needed. And you can leave out the whole R4/R5/C5
combination on our psu boards or your own heater bias section.
That is a way to not have to use a upward in voltage referenced heater supply. It works well. Often a bigger value like 0,47uF will work even better. (Once experienced that with a Mu-Follower where the top tube was "floated") One could hear the hum go down through the speakers as I was paralleling the caps.
So if you use that method no resistors and big capacitor combination is needed. And you can leave out the whole R4/R5/C5
combination on our psu boards or your own heater bias section.
C3/C4 is for the screen in a 9-pin mini board, I think.
C7,8,9,10 are the heater capacitors, if I am not mistaken.
But, now I see up in the text that Broskie suggests leaving them off if AC is fed to the heaters.
C7,8,9,10 are the heater capacitors, if I am not mistaken.
But, now I see up in the text that Broskie suggests leaving them off if AC is fed to the heaters.
Not really. They are for the floating the heaters. But IF you have a screen on pin 9 they won't do harm. And IF you have a screen on 9 the best theoretically would be to use a wire instead.
Ah yes..now I see C7..10 are for a DC supply. (Could not find them on the schematic)
Does the Aikido natural PSRR apply to both differential and common mode noise?
Would a common-mode choke be a noticable advantage in a PSU design?
Would a common-mode choke be a noticable advantage in a PSU design?
One more question on a choke-based PSU:
http://www.vt52.com/diy/tips/tips_chokeload.htm
This implies that a CLCLC PSU with 10H chokes would be too little for a 6SN7 (since the plate impedance is around 7K), and cause high frequency rolloff.
And, Broskie suggests a "high DCR" choke in his manual, which implies that resistence is needed.
Does this mean a CRCRC or CLCRC PSU might actually be a better choice?
http://www.vt52.com/diy/tips/tips_chokeload.htm
This implies that a CLCLC PSU with 10H chokes would be too little for a 6SN7 (since the plate impedance is around 7K), and cause high frequency rolloff.
And, Broskie suggests a "high DCR" choke in his manual, which implies that resistence is needed.
Does this mean a CRCRC or CLCRC PSU might actually be a better choice?
Phew..that's more then one 😉
The choke load you refer to in your link..is not the same as a choke in a psu. If you want to see what a choke will do in your psu download DuncanAmps Psud simulator. (Search in google).
Each has it's own advantages. A resistor is cheap.
A choke has a more resistance the higher the frequencies get. So they say a choke has a higher impedance to AC.
Generally speaking a choke is better than a resistor in a tube amplifier power supply. If I'm not mistaken some of the best sounding amplifiers in the past were built without a choke in the psu. And even today..take the foreplay preamp by bottlehead, it uses RC psu and many people love it to bits.
These days you have solid state devices that can perform the function of a choke. Giving a lot of bang for your buck. But some don't like to use transistors or mosfets in their amplifiers. And for a newbie a choke is waaay easier to implement.
It is worth your while though to play with psud. You could also wait to see what PSU John Broskie comes up with. I for one am pretty curious.
Here is an article by DHT Rob, whicht I translated into english, on PSUD. It goes past the basics into the finer details. But I can see that you are like a sponge at the moment. Very eager to learn! Go boy go!
https://www.diyaudio.com/community/threads/basaudio-net.79841/
https://www.diyaudio.com/community/threads/basaudio-net.79841/
Last edited by a moderator:
Bas: Great Article!! Thanks.
But it brings up more questions in my mind:
1) I'm not sure I 100% agree with the reduction to a single filter stage. This decreases the slope of noise reduction to 6db/octave, whereas 3 or 3 stages would have a noise reduction of 12db and 18db, respectively.
I can see that a 6db filter would have less phase shift...but we are trying to achieve DC, which has no phase.
2) Isn't the 47uF capacitor he arrives at for the first cap too large for a 6x5?
3) Most of his analysis centers around the first 10sec after turn-on. Shouldn't we be more concerned with the steady-state (ie...POST 10sec)?
When playing around with PSUD before, I used the 0-10s portion to make sure I didn't exceed any component maximums....but I focused on the the post-10sec for most evaluations.
But it brings up more questions in my mind:
1) I'm not sure I 100% agree with the reduction to a single filter stage. This decreases the slope of noise reduction to 6db/octave, whereas 3 or 3 stages would have a noise reduction of 12db and 18db, respectively.
I can see that a 6db filter would have less phase shift...but we are trying to achieve DC, which has no phase.
2) Isn't the 47uF capacitor he arrives at for the first cap too large for a 6x5?
3) Most of his analysis centers around the first 10sec after turn-on. Shouldn't we be more concerned with the steady-state (ie...POST 10sec)?
When playing around with PSUD before, I used the 0-10s portion to make sure I didn't exceed any component maximums....but I focused on the the post-10sec for most evaluations.