Hi,
I have recently purchased an ARC SP8 Rev 3 from the '83 in working condition that I Would like to upgrade with some modern parts and ideas. I have already recapped the preamp PSU, upgraded 2 OPAMPS, changed the rectifier diodes and put new valves on it, I first Would like to concentrate on the PSU. Voltage references based on noisy zener diodes is my next step. I Would appreciate it very much to have some feedback on several points I Would like to share with you.
My strategy is to replace 2 zeners and a LM329 with the attached circuit (vRef):
1. 2 B+ voltages (phono and line) are regulated using 1 12BH7A and 1 ECC83 (this one working at 400 V, far over specs ?!) together with a couple of tl2071 OPAMPS. Both cathodes of the ECC83 are paralleled and biassed with a noisy 10V ZENER. I Would like to replace the ZENER with the low noise referred equivalent. To provide enough current to the voltage reference I plan to get additional 6mA from a 24v available regulator.
2. An additional 24V ZENER is used to power a tl071 OPAMP attached to the base of a darlington set acting as a HT voltage regulator. I believe the circuit proposed above can provide lower noise at the OPAMP power supply.
3. A LM329 is also used as a fixed reference of an inverting tl071 input. It is a nice device but also a bit noisy.
I Would appreciate your comments / suggestions.
Kind regards
Toni
Barcelona
I have recently purchased an ARC SP8 Rev 3 from the '83 in working condition that I Would like to upgrade with some modern parts and ideas. I have already recapped the preamp PSU, upgraded 2 OPAMPS, changed the rectifier diodes and put new valves on it, I first Would like to concentrate on the PSU. Voltage references based on noisy zener diodes is my next step. I Would appreciate it very much to have some feedback on several points I Would like to share with you.
My strategy is to replace 2 zeners and a LM329 with the attached circuit (vRef):
1. 2 B+ voltages (phono and line) are regulated using 1 12BH7A and 1 ECC83 (this one working at 400 V, far over specs ?!) together with a couple of tl2071 OPAMPS. Both cathodes of the ECC83 are paralleled and biassed with a noisy 10V ZENER. I Would like to replace the ZENER with the low noise referred equivalent. To provide enough current to the voltage reference I plan to get additional 6mA from a 24v available regulator.
2. An additional 24V ZENER is used to power a tl071 OPAMP attached to the base of a darlington set acting as a HT voltage regulator. I believe the circuit proposed above can provide lower noise at the OPAMP power supply.
3. A LM329 is also used as a fixed reference of an inverting tl071 input. It is a nice device but also a bit noisy.
I Would appreciate your comments / suggestions.
Kind regards
Toni
Barcelona
If you bypass the zeners with caps the noise will be greatly reduced ... LM329 high-performance zener is for instrumentation , I doubt you can hear the difference only in a bad design if noise is somehow amplified
Last edited:
LM329 is very low noise for a solid state reference, i doubt you can get much better. It has the same maximum 20uV guarantee of the LM399 and that part is in a 6.5 digit multimeter..
When it comes to the zener in the cathode lead of the voltage amplifier stage, you could replace that with a simple TL431 with a capacitor over it. its slightly less noisy. but just putting a simple orange drop over the original zener will also cut noise slightly. And you dont have to change the board for the simple addition of a capacitor..
Point 2. a 24V zener stabilised opamp is going to give some noise, how much depends on the zener self noise and zener impedance. However the 071 has a reasonable power supply noise rejection ratio so tackling this will give diminishing returns. You could try another opamp but that also means that the frequency compensation is another.
When it comes to the zener in the cathode lead of the voltage amplifier stage, you could replace that with a simple TL431 with a capacitor over it. its slightly less noisy. but just putting a simple orange drop over the original zener will also cut noise slightly. And you dont have to change the board for the simple addition of a capacitor..
Point 2. a 24V zener stabilised opamp is going to give some noise, how much depends on the zener self noise and zener impedance. However the 071 has a reasonable power supply noise rejection ratio so tackling this will give diminishing returns. You could try another opamp but that also means that the frequency compensation is another.
Thanks guys for your comments !.
zener bypass cap - Do you think 0.1uF would be ok ?
TL431 - Is there any min current requirement to keep the setting voltage ?. The Ecc83 delivers just a few mAs to each catode.
Regards !
T.
zener bypass cap - Do you think 0.1uF would be ok ?
TL431 - Is there any min current requirement to keep the setting voltage ?. The Ecc83 delivers just a few mAs to each catode.
Regards !
T.
TL431 needs 1mA through the anode to kathode to maintain regulation i think. you will also lose some current through the divider resistors. Normally you would make the divider resistors quite small, say 1K per volt so the reference pin current change with temperature doesn’t produce too big of an error. But in this circuit zener voltage is not too critical, so what you could do is take say 4K per volt. so 10K from reference to kathode. and then about 47K from reference pin to anode to replace a 15V zener.
Yes 0,1uF should be ok , I don't know why they didn't do it in the first place