My intention with the preamp is to have just a single transformer for both the phono stage and line buffer using a single torriod with multi tappings. So guys I need your help to spec this for me so I can get one wound up and ordered.
I suppose the other question is, is there any benefit in having seperate torroid supplies over one large special? I'm thinking space may be at a premium inside the chassis I am thinking ordering for the build. Presently the buffer has back to back 6v +6v 50Va torroids ala Nick design.
Where do you recommend I order the special torroid from?
For the DC heaters I think a 9v winding would be fine. The current valve config requires an amp, or thereabouts, but I think you may find an eventual desire to try two separate valves instead of the ECC88, as an upgrade. We should therefore make sure you have capacity for 4 * 330mA heaters. I would ask for a 9v winding at 2 amps, just for margin.
For the mains supply you need 80mA @ 300v I'd call that 100mA, at least to be safe, perhaps a bit more, but remember that 300v is rectified so you need to work out what input voltage to give the rectifier and if you're using cap or choke input.
For the breadboard you could try a similar set up to the D3A buffer but using two back to back 9v toroids.
I expect you could take the reg supply as a straight tap from the CLC so 22uF - 7H - 100uF and then the two regs, one for each channel.
Just to be sure why not measure the voltage at that point on your D3a buffer; this will be somewhat approximate as the phono draws a bit more current, but should give you a good idea.
What's the rating on those 7H chokes? You may get away with a single TX but you might need more than one choke.
The circuit from the one I posted has not changed, but you do need to be aware of how to calculate the best values for the big resistor in series with the zener. The zener needs to pass 5mA, otherwise the circuit is noisy, to much and it hums, to little and you get noise from the BZX84C15 zener. I would, to be safe, make this resistor at least 3 watt rated but, 5 or 7 would be better. You can use 3 watts if you get a slow start from a vacuum rectifier, otherwise for solid state rectifier put something a bigger one there.
Take your input voltage, whatever that might be, subtract your target voltage plus 15v (the VBE uses rougly 15v) so in your case that 250 + 15 = 265v (that's the voltage the regulator is giving you). Say your input voltage is 315v so thats a 50v drop across that resistor. A quick Ohms law gives you R=V/I or 50/0.005 = 10000, so you need a 10K resistor. If the voltage is 300v then 35/0.005 so 7k (nearest value is 6k8) and so on.
The final bit is the reg needs greater than ~20 volts across it to function, so don't drop the input below 285v, preferably 290v.
You may also want to play with larger caps at C4 and C10, I've made some notes on a new diagram which I will post as soon as I can.
Thanks, that's pretty clear. Just to sure though, the large resistors in series are R4/11?
What sort of difference would you expect the larger cap at C4/10 to make out of interest? Is this the signal return to earth, and a quality cap is needed here?
The value of the cap is multiplied by the HFE of the transistor pair, I tried 4u7 and thought that was a good compromise between cost, size and being able to use a film cap and yes you should use a quality cap. But I've also got one with a lower value on it, 470nF. Just what I have lying around really.
You will still need to de-couple between stages, however. I would do as Stephen suggests, play with various values here. I expect the lowest values you can get away with will actually be best. Or, as you get used to building them, and if you like them, then build 4 regs one per channel per stage.