Looking at building an Aikido linestage. I have decided I'd like to use the noval stereo PCB as I like havinb more options for tubes with the noval version.
1. I see there is mono boards that are sold. What's the advantage of two mono boards as well as two separate power supplies? I'm thinking isolation, cross talk, etc. but not sure.
2. What tube would you go with? I was thinking 12AU7 because I have some on hand, they are relatively cheap and plentiful. I don't want to use a tube that is a) expensive or b) can only be found as NOS.
The Aikido can use:
3. This stage needs to provide 10-20dB of gain. If I wanted to add a switch for 10, 15 and 20dB what would be the best way about doing this?
4. I'd like a PSU that is regulated for both the heaters and B+. I was thinking of the PS-1 PS-1 High & Low Voltage Regulator Kit but I'd like to here other thoughts on it and/or what else is recommended.
5. I am going to include a volume control and source selector control. Was thinking of adding a balance control but are those generally frowned upon?
Thanks!
1. I see there is mono boards that are sold. What's the advantage of two mono boards as well as two separate power supplies? I'm thinking isolation, cross talk, etc. but not sure.
2. What tube would you go with? I was thinking 12AU7 because I have some on hand, they are relatively cheap and plentiful. I don't want to use a tube that is a) expensive or b) can only be found as NOS.
The Aikido can use:
3. This stage needs to provide 10-20dB of gain. If I wanted to add a switch for 10, 15 and 20dB what would be the best way about doing this?
4. I'd like a PSU that is regulated for both the heaters and B+. I was thinking of the PS-1 PS-1 High & Low Voltage Regulator Kit but I'd like to here other thoughts on it and/or what else is recommended.
5. I am going to include a volume control and source selector control. Was thinking of adding a balance control but are those generally frowned upon?
Thanks!
Voltage gain will be mu/2. Choose your valve accordingly.
To get less gain add an attenuator: at the input for best distortion and driving capability, at the output for best noise. Same issue applies for volume control - which is a variable attenuator.
Why worry whether other people will frown on a balance control?
To get less gain add an attenuator: at the input for best distortion and driving capability, at the output for best noise. Same issue applies for volume control - which is a variable attenuator.
Why worry whether other people will frown on a balance control?
Hi DF96, thanks for the info.
I am going to have both a volume control and a switch to add an attenuation network to get 10, 15 and 20dB output. One control will go at the input and the other at the output, guess I will have to decide where each is best suited...
Instead of a balance control I was thinking of just doing dual volume (gain controls). I think I like that idea better.
Regarding the 10, 15 and 20dB attenuation network I have calculated the resistors I would need. Since the output impedance will be <1K I think I could use resistors in the 10K range. Though to get accurate output of 10, 15 and 20dB I think I need to take the output impedance and add that to the attenuation network “tail” (bottom) resistor. Is this the correct thinking?
I am going to have both a volume control and a switch to add an attenuation network to get 10, 15 and 20dB output. One control will go at the input and the other at the output, guess I will have to decide where each is best suited...
Instead of a balance control I was thinking of just doing dual volume (gain controls). I think I like that idea better.
Regarding the 10, 15 and 20dB attenuation network I have calculated the resistors I would need. Since the output impedance will be <1K I think I could use resistors in the 10K range. Though to get accurate output of 10, 15 and 20dB I think I need to take the output impedance and add that to the attenuation network “tail” (bottom) resistor. Is this the correct thinking?
Last edited:
Ah yes, you are correct.
The input of the circuit has a 1M grid leak and the output has a 1M pulldown after the output cap. Does that mean wherever I decided to add the attenuation it must sum to a net 1M to keep the impedances as stock?
That 1M input grid leak seems high. A phono preamp would feed this stage, an input impedance of 100K or so would seem more than adequate.
EDIT: checked numbers again. I think I do want the output impedance added to the shunt resistor and not the series resistor.
The input of the circuit has a 1M grid leak and the output has a 1M pulldown after the output cap. Does that mean wherever I decided to add the attenuation it must sum to a net 1M to keep the impedances as stock?
That 1M input grid leak seems high. A phono preamp would feed this stage, an input impedance of 100K or so would seem more than adequate.
EDIT: checked numbers again. I think I do want the output impedance added to the shunt resistor and not the series resistor.
Last edited:
Grid stopper and grid leak are not the same. 1M would be grid leak, and is fine for a small signal stage. A phono preamp which can drive 100k will drive 1M quite happily.
No, the attenuator does not have to present a 1M resistance. On the contrary, as the 1M resistances are fairly high you can probably ignore them when calculating your attenuator. Issues for an attenuator are the source and and load impedances. If the former is low and the latter is high then you have considerable freedom of design.
No, the attenuator does not have to present a 1M resistance. On the contrary, as the 1M resistances are fairly high you can probably ignore them when calculating your attenuator. Issues for an attenuator are the source and and load impedances. If the former is low and the latter is high then you have considerable freedom of design.
That’s what I am getting at; a phono preamp can drive 100K quite easily. My particular phono stage is op-amp based and has very low output impedance. There is no need for the input impedance to the linestage to be 1M. It seems such high input impedance would only add the potential for noise. A 100K volume control on the input would seem more suitable.
If the attenuation network was at the end of the circuit I would think I would want the overall load of it to be fairly large, no? It would be in parallel with the next stages (poweramp) input impedance. If it is too low it would lower the poweramp’s input impedance.
Only when no input source is connected. You won't listen to music like that, so no problem.ChrisM91 said:
Attenuator impedance is always a compromise between loading the source too much and adding noise. For a simple potential divider the source sees the two resistances in series, and the load sees the two in parallel.
That is not really the issue. The issue is what load impedance does the source see (as high as possible) and what source impedance does the load see (as low as possible).
I've built a couple of Aikidos
I built an HPA using the All-In-One noval board (no longer available) and a scratch built Line stage but using one of Broskie's power supply boards (the PS-1 if memory serves). The HPA uses 6CG7 input tubes and 6DJ8 outputs. The line stage uses 6CG7s for both input and output tubes.
Sound wise I did a lot of tube rolling on the HPA before I settled on NOS GE 6CG7s and NOS Mullard 6DJ8s. The good news is I got results almost as good using Electro-harmonix 6CG7s and NOS Tesla (not JJ) ECC88s from Tubemonger for 30 bucks a pair!
The line stage using solely 6CG7s sounds good but just a little bit too tubey (fat, slightly "slow" bass). I think I'll change it over to 6DJ8 outputs.
As far as boards, I'd go for the stereo board (keep checking the TUBECAD site for availability and order as soon as it's in stock).
The reason I like the stereo board is that you can wire the filament voltage for 12 volts, the two input tubes in series and the two outputs in series. This lessens the current load on the filament rectifier diodes and the voltage regulator. With two mono boards using tubes of different filament current draw E.G. 6CG7 and 6DJ8, you'll have to set the filament voltage at 6 volts with the resultant higher current draw. This might require TO-220 rectifiers with small heat sinks.
Cheers, Steve
I built an HPA using the All-In-One noval board (no longer available) and a scratch built Line stage but using one of Broskie's power supply boards (the PS-1 if memory serves). The HPA uses 6CG7 input tubes and 6DJ8 outputs. The line stage uses 6CG7s for both input and output tubes.
Sound wise I did a lot of tube rolling on the HPA before I settled on NOS GE 6CG7s and NOS Mullard 6DJ8s. The good news is I got results almost as good using Electro-harmonix 6CG7s and NOS Tesla (not JJ) ECC88s from Tubemonger for 30 bucks a pair!
The line stage using solely 6CG7s sounds good but just a little bit too tubey (fat, slightly "slow" bass). I think I'll change it over to 6DJ8 outputs.
As far as boards, I'd go for the stereo board (keep checking the TUBECAD site for availability and order as soon as it's in stock).
The reason I like the stereo board is that you can wire the filament voltage for 12 volts, the two input tubes in series and the two outputs in series. This lessens the current load on the filament rectifier diodes and the voltage regulator. With two mono boards using tubes of different filament current draw E.G. 6CG7 and 6DJ8, you'll have to set the filament voltage at 6 volts with the resultant higher current draw. This might require TO-220 rectifiers with small heat sinks.
Cheers, Steve
At the lowest gain setting that will provide a source impedance of about 68k to whatever comes next. Perhaps a bit high. Two issues here: driving any cable capacitance, and driving any non-linearity in the input impedance of what follows.ChrisM91 said:
There isn't a 'right' answer to attenuator values. Everything is a compromise. Perhaps use values around the geometric mean of the source and load impedances?
I don’t think the 12B4 is compatible with the Akido Linestage PCB unfortunately. I suppose I could use non-PCB mounted sockets and have flying leads going to the board to make it work though.
I was looking at the 5687 tube; it seems it’s just a bit lower gain (amplification factor) than the 12AU7.
Any other tubes you’d suggest?
I was looking at the 5687 tube; it seems it’s just a bit lower gain (amplification factor) than the 12AU7.
Any other tubes you’d suggest?
ChrisM; a slightly different solution?
For a lower gain, as well as a lower B+ voltage, consider building an Aikido based on the 6GM8 tube. The only problem is the 6GM8 (ECC86) is getting a bit rare and pricey. .
I've built Broskie's 12Vac kit but used a 9 volt transformer instead of a 12 volt and the 6GM8 instead of the 6DJ8. Broskie likes to run the 6GM8 at about 12 volts and 2.5 mA per section. To me it sounds a bit cold so I like to crank it up to about 25 volts per section and about 4 or 5 mA per section. The max plate voltage per section is 30 volts. Even with 25 volts and 5 mA the tube is at a small fraction of its max plate dissipation.
All of the 6GM8s that I've run across (even if labeled Zenith, GE or Motorola) have been made in Holland (Philips) or Germany (Valvo or Telefunken). There is even a Russian equivalent out there, 6N27P, but the crooks on eBay want almost as much for these as the European tubes.
Cheers, Steve
For a lower gain, as well as a lower B+ voltage, consider building an Aikido based on the 6GM8 tube. The only problem is the 6GM8 (ECC86) is getting a bit rare and pricey. .
I've built Broskie's 12Vac kit but used a 9 volt transformer instead of a 12 volt and the 6GM8 instead of the 6DJ8. Broskie likes to run the 6GM8 at about 12 volts and 2.5 mA per section. To me it sounds a bit cold so I like to crank it up to about 25 volts per section and about 4 or 5 mA per section. The max plate voltage per section is 30 volts. Even with 25 volts and 5 mA the tube is at a small fraction of its max plate dissipation.
All of the 6GM8s that I've run across (even if labeled Zenith, GE or Motorola) have been made in Holland (Philips) or Germany (Valvo or Telefunken). There is even a Russian equivalent out there, 6N27P, but the crooks on eBay want almost as much for these as the European tubes.
Cheers, Steve
- Status
- This old topic is closed. If you want to reopen this topic, contact a moderator using the "Report Post" button.