How to improve power supply

robnec

Well-Known Member
Hello. I have build about three or four preamplifiers during last two years. I am tired of expense and trouble of building, basically, the same power supplies. I decided to make separate one that can be used for different units. I bought on auction site Sony single ended amplifier power supply. I will replace capacitors with pio ones, install choke and two regulated heater supplies. I have a few questions
-what should be done in general to make it a great power supply
- any chance for 12 volts heater supply
-what is a purpose of T1 transformer
Robertsony power 2.jpg Sony power.jpg
 
T1 looks like part of a bias/erase oscillator for a tape recorder.

You could free up a 6.3V heater winding by switching to SS rectifiers, which would allow you to make 12.6V by operating it in series with another 6.3V winding.

In general, the highest level of performance results from tightly regulated power supplies, but if you go with a tube for the regulator pass element, then you might have to grab back that 6.3V winding. Suggest you bone up on regulator circuits before proceeding.
 
Hello. Thank you for suggestions. So two 6.3v windings can be put in series for 12.6v? It already has two so it is possible. I would like to keep tube rectification. Would two ao2 tubes for 300v be sufficient for voltage regulation? I intend to make a new box for it so real estate is not a problem. Do you think that pio caps about 30uf have advantage over electrolitics of higher value? Robert
 
Why go to 'expense and trouble' of regulating when you may not notice any audible difference and may not be able to measure any electrical difference.

Is the wiring and layout set up for minimal noise? There is no secondary fuse. Have you checked the design requirements of the 6CA4? The heaters could be elevated.
 
Hello. I use ao2 voltage regulation in all of them and it worked good. My question was a simple, for those who know, practical nature how to make a good power supply. I do it for myself as fun so I do not set any limits as expense and labor. Robert.
 
If you build it with a wide-range series regulator, then it could also serve as a benchtop power supply for new circuit development. However, circuits intended for operation on unregulated power should always be tested with the planned power supply before development is complete. I've spent many years designing analog signal processing circuits and I still find it difficult to achieve an excellent power supply rejection ratio.
 
Do you think that pio caps about 30uf have advantage over electrolitics of higher value? Robert
I am using ASC 386s oil caps which are 100uf x 370vac in my SE amps and they sound really good to me.The big downside is they are quite large. Supposedly their lifespan is in the tens of thousands of hours. Vintage pio caps are even bigger but the newer DC Link caps are relatively small for their ratings. For all my personal builds i use film caps exclusively and they have been totally trouble free.
 
Hello. Gentlemen. I highly appreciate your input but level of arguments is way above my ability. What I am looking for is a direction what has to be installed where. For example put choke such and such here etc.
That is what I am asking for if possible. This power supply has to provide about 300v for maximum four small tubes. Can transformer T1 be removed? Robert
 
Hello. Excellent. That I understand. Where the choke would fit best if I use two ao2 regulators in series? Or shell I rather buy silicon boards ready to use from auction site? Robert
 
Hello. Excellent. That I understand. Where the choke would fit best if I use two ao2 regulators in series? Or shell I rather buy silicon boards ready to use from auction site? Robert
Ok, if i were to mod your ps. , you have C11 to R6 600 ohm to C10. Replace C11 with you PIO or film cap of correct value (you need to stay within the limit of the 6ca4 recommend cap input value). Replace R6 with your choke. There is some "leeway" with the choke inductance values as long as the minimum choke inductance is covered, but the current capability should be what the total current use your ps will require. The choke is supposed to be of a minimum value set by the voltage output from the rectifier/current at maximum output (Ma). As long as the choke inductance exceeds this value then you will not risk your tube rectifier max. current limits. I doubt that the choke you will use is 600 ohms so you will need to make up the difference in another RC filter.
A helpful tool to model these values is PSUDII which is a free download from Duncan Amps iirc.
 
The choke is supposed to be of a minimum value set by the voltage output from the rectifier/current at maximum output (Ma). As long as the choke inductance exceeds this value then you will not risk your tube rectifier max. current limits. I doubt that the choke you will use is 600 ohms so you will need to make up the difference in another RC filter.
There is no minimum inductance value for cap-input filter systems, but to reject 120Hz ripple at least as well as a 600R resistor, you need at least 0.8H.
 
Yeah, minimum inductance is for choke input filter, but not being totally sure of what is acceptable, one could not "go wrong" with at leas the minimum as relates to a choke input filter.
 
I suggest to OP to download "PSUD-II" and start playing with different configurations. For a preamplifier I don't see a lot of sense in using a choke, but your religion may dictate otherwise. I think for a truly quiet supply between 4-5 sections of RC filtering would be needed. Any less and you'll be hearing the hum.
 
Robert, do you have a target B+ voltage level for your preamps?

If there is sufficient voltage margin between what the power supply could provide by simple capacitor input filter, and the preamp requirement, then as suggested RC filtering or a regulator can both drop voltage and lower ripple/noise on the output supply. Were you just powering a known load, or were you wanting to have the option to power from one to many preamps (ie. needing regulation so as to maintain a nominal supply voltage no matter what the load was) - the transformer secondary current rating does seem plentiful.

If you have sufficient voltage margin, then it can be an advantage to add a portion of filtering resistance budget in series with the transformer secondary winding resistance, so as to dampen/smooth the rectifier current pulses. It may seem more productive to just get the largest raw rectified voltage from the largest allowed filter cap connected to the 6CA4, and then filter that down for a low ripple B+, but that may degrade the heater supply voltage waveform (which can't easily be filtered unless you go to a DC heater supply).
 
Hello. Thank you all. I had crazy day at work and I am not at home now. I meant oa2 tubes of course. So any choke above 1 h would work. Great. As I said anything this power supply will meet are four 12a?7 or 6922 type of tubes . I am sure a power transformer will have a huge reserve. Plus 12bh tube will be removed. I will have to learn how to use this simulation program. Robert.
 
Hello. I had zero time to play with anything but work until today. Do the voltage regulation should be after a choke or other way around? I expect around 280 volts to supply a preamplifiers. Is this schematic acceptable? I can easily add one or two filtering stages. Thank you. Robert.sony power (3).jpg
 
Why settle for one? Two separate but equal supplies are rumored to improve channel separation immensely. Never tried it myself. Also, you get bonus points for putting those on a separate chassis for noise isolation. Or so they say ...
 
Voltage regulation circuits, shunt or series, are generally furthest downstream in any power supply design. Never connect gas tubes in parallel with large capacitors -- they will be destroyed. Check the tube's datasheet for maximum parallel capacitance. Understand that series resistance must be present to limit gas tube current. Chokes have little series resistance, so you may be compelled to add some, which defeats the advantage of using a choke. It seems pointless to use a choke in a regulated PSU in any case. What are we trying to accomplish here?
 
Back
Top Bottom