Volume pot is not shown.
Thanks for the clarification and input Baudouin0. I will play around with going all ECC88 and see what happens. I suspect it might effect turnover points and gain/cut.
Thanks for the clarification and input Baudouin0. I will play around with going all ECC88 and see what happens. I suspect it might effect turnover points and gain/cut.
I changed it to ECC88 with the same bias as U1 and the performance changed very little. The cut and boost curves stayed the same within less than a dB and the -3dB point when set flat moved from 50KHz to 52KHz. The overall gain of course did not change and distortion simulation results were essentially the same also. So at a minimum the performance did not degrade at all. Given the lower mu I thought the max boost might be effected but that was not the case.
P.S. I ran B+ noise and with 1.9V RMS output and 1mV of power supply ripple the hum is 94dB below the fundamental using the ECC88 throughout.
WAIT!!! I forgot to remove the plate cap. Please stand by...
OK, 140Kz 12AX7 154KHz ECC88. The flat setting as about 0.25dB flatter below 1KHz with the 12AX7 but that is not enough to make it preferable in my opinion.
P.S. I ran B+ noise and with 1.9V RMS output and 1mV of power supply ripple the hum is 94dB below the fundamental using the ECC88 throughout.
WAIT!!! I forgot to remove the plate cap. Please stand by...
OK, 140Kz 12AX7 154KHz ECC88. The flat setting as about 0.25dB flatter below 1KHz with the 12AX7 but that is not enough to make it preferable in my opinion.
Last edited:
Now sim 10mV (more real life) 🙂
Then you could add this as a PSU and see how good it performs for you (in SPICE? 🙂). I use this for my line and phono stages... You could use LM317 if you're careful and use a 30V zener instead of a 105V one.
In my case V1 = 400V, R1=560R, C1=100µF, R2=100R, C2=100µF. Surprisingly quite really for how simple it is.
Use R4 = 910R for ~282V out.
Then you could add this as a PSU and see how good it performs for you (in SPICE? 🙂). I use this for my line and phono stages... You could use LM317 if you're careful and use a 30V zener instead of a 105V one.
In my case V1 = 400V, R1=560R, C1=100µF, R2=100R, C2=100µF. Surprisingly quite really for how simple it is.
Use R4 = 910R for ~282V out.
Last edited:
18mA only? Damn. Also, "real life" to me is the kind of values you'd see in the 1950's... 100R and 47µf but the low current draw helps.
My tone control circuit alone used that... Mine draws 50mA and so does the phono 🙂
My tone control circuit alone used that... Mine draws 50mA and so does the phono 🙂
🙂 The schematic above is a single channel and draws less than 10mA. I could bias it hotter but I am not sure I would gain anything. Even doubling the draw should still give manageable ripple. For the phono amp I will probably use at least one choke in the PS and DC heaters.
I am just the kinda guy that prefers a carb to EFI. 😉 I tend toward using SS as a last resort or if the cost is just completely out of line. PS chokes are reasonable enough that I don't feel too bad about popping $20 or $30 for one. I used a couple on my KT88 power amp and they worked a treat.
In the case of a phono preamp I have a choke that I think is around 14-16H that I salvaged from a Hammond organ so if it works I won't even have to drop any coin on that.
I would love to play around with plate chokes and ITs but now we are talking more like hundreds of dollars and that will have to wait for when I feel a lot more flush.
P.S. Saved your schematic for later use. 🙂
Another concern. Would it be wise/necessary to have an input blocking cap for this kind of "anode follower"? Here I am showing a 25k volume pot and grid leak is through R4. It seems like it might be cheap insurance in case of an input that accidentally has DC on it.
Use a cap before the volume pot and all is good. The volume pot is also acting as grid leak.
Except if you have a bad wiper you'll still get a bunch of crap through the amp.
Better to put the cap (100nF) after the pot and then a 1M leak IMHO.
Better to put the cap (100nF) after the pot and then a 1M leak IMHO.
No DC current flows through wiper R8 or R4. Tone control is at 0V. There is AC current through the wiper because the junction of R8 R4 R7 approximates to virtual earth rather like an op-amp but not quite as good.
Last edited:
I really appreciate all the help. For reference I am strongly considering putting the phono pre in the same case so it would look much like this. In a sense C16 (and therefore R54) are redundant but my fear is that if they are left out you would get a horrible thump when switching to phono. I am assuming that leaving the phono preamp connected to the main line amp and just switching off the input to the phono section is not a good idea.
I've not built a phono pre-amp but the 12AY7 in the first stage is not the lowest of noise. ECC88 may be better (both sections in parallel maybe) but hear they can be microphonic sometimes so you may need to try a few. There's also a 6J52P. You will need regulated heater DC for the input stages.
Of course the first thought is always to use the 12AX7 but it has high Cg-a and can be pretty microphonic. My thought was to try cascaded 12AY7s with local feedback in hopes of getting about the same overall gain with less microphonics and input capacitance plus lower distortion. It may not work but that is the reasoning.
My reasoning on the capacitance is that the Cmiller seen by the cartridge will be the Cg-a of the first 12AY7 (1.3pF v.s. the 12AX7s 1.7pF) times the gain of the first stage only (I am not sure whether the FB from the output of the second stage would reduce this). The Cmiller of the second stage will be driven by the first stage plate and thus buffered from the cartridge.
I planned on using heavily filtered DC heater supply for the phono section (or at least for the gain stages).
My reasoning on the capacitance is that the Cmiller seen by the cartridge will be the Cg-a of the first 12AY7 (1.3pF v.s. the 12AX7s 1.7pF) times the gain of the first stage only (I am not sure whether the FB from the output of the second stage would reduce this). The Cmiller of the second stage will be driven by the first stage plate and thus buffered from the cartridge.
I planned on using heavily filtered DC heater supply for the phono section (or at least for the gain stages).
For phono, I use 6N1, 6N2, or 6N3 depending on how much gain I need for first input. I leave out the P but I use Soviet tubes because they we're a good value 🙂
Are you worried about Cmiller loading the phono preamp input? If so, is that because you're running a moving magnet cartridge, or are you running a moving coil into a step-up transformer?
Edit to add:
NFB will also lower the input capacitance by reducing Miller effect. The feedback reduces the gain of the stage, so will reduce Miller effect along with it.
6J52P makes for a great hi-mu triode but it has extremely high input capacitance.
Edit to add:
NFB will also lower the input capacitance by reducing Miller effect. The feedback reduces the gain of the stage, so will reduce Miller effect along with it.
6J52P makes for a great hi-mu triode but it has extremely high input capacitance.