Power supply project

[noizyboy]

Hi, I would like to ask some points of view with regards to a power supply I intend building, for a pair of class AB power amplifier circuits. I have a 1.2 KVA toroidal transformer which has 4-secondary windings at 35v, so I have found I can use two bridge rectifiers per stereo psu.

Question 1 is, Is the 1.2KVA divided by four?

Question 2 is, how much capacitance can I use per rail, i.e. pos/neg for channel A, and pos/neg for channel B ?
(my thoughts are 100,000uF/100,000uF for channel A, and the same for channel B).

Question 3 is, Can I use more?

Question 4 is, do I have to use less?

I would appreciate any points of view.

 

[DF96]

1. Yes, roughly. That does not necessarily mean that you can feed a 600W per channel amplifier, as it depends on duty cycle. Transformer VA rating applies to resistive load not diodes plus caps.
2. depends on peak current rating of rectifiers, and how much you intend to mess up your local mains supply.
3,4. see answer to 2.

 

[noizyboy]

Thank you for your responce, so I have 35A rectifiers and 0.1F 50V caps.

 

[jeronimo2003]

I think that 100,000uF per secondary is a bit to much. If the PSRR of your power amp is that bad look for an other circuit. This amount of decoupling means large current spikes every time the mains line tries to charge them. I learned as a rule of thumb that 1000uF per amp should do the trick. So 10.000uF or 22.000uF per secondary should be enough.

BTW, I also think that you need a good soft start, if you are not sponsoring the fuse manufacturers.

 

[noizyboy]

Thank you very much for your information, I like the idea of the soft start, ehich I will impliment into the supply. I will research the info you have kindly passed on to me and have a better understanding of what I am letting myself in for. So I will need Frankensteins busbars for an on-switch!

 

[jeronimo2003]

Perhaps it's possible to give the link for the amplifier you want to use/build, so we can give you a better advise.

 

[jeronimo2003]

Perhaps it's possible to give the link for the amplifier you want to use/build, so we can give you a better advise.

 

[CBS240]

Thank you very much for your information, I like the idea of the soft start, ehich I will impliment into the supply. I will research the info you have kindly passed on to me and have a better understanding of what I am letting myself in for. So I will need Frankensteins busbars for an on-switch!

Toroidal transformers have very low leakage flux and magnetic memory. Large toroids >400VA require slow start. When power is switched off during the peak of the mains signal the core can remain magnitized in that direction. If power is then switched back on during the same polarity peak it will drive the core into saturation and the fuse will light up. Placing a resistor in series with the primary for a short period of time, 10 cycles or so, then shorting the resistor placing the primary on the line will prevent this.

For my 1.4KVA 120VAC transfo, I use a simple isolated logic circuit that drives a photo-triac to trigger a triac full cycle to place the resistor in series, then a relay closes and places the primary on the line. My transfo is 1.4KVA, 120VAC. It works well and doesn't vaporize fuses.

 

[noizyboy]

Slow start

Thank you CBS240 for your very interesting info on the magnetic flux memory, it is the sort of info I like to get to know about as it will teach me more about audio distortions and the like. My transformer will be running on 240vac in the UK and driving a clone Naim NAP 140 amplifier circuit.

The amp is available from ebay here is the item number for anyone who wishes to view the amp.

A good idea from jeronimo2003, although the amp doen't give any circuit details, I do have the circuit diagram and I also have some built, so I could take meter readings. I don't have a scope at the moment, but I do have access to scopes, network anylisers, and signal generators etc...

I am a test technician for train safety equipment, I don't proffess to be an electronics whizz kid, but I do enjoy good quality audio. So, again, I appreciate all that you guys n gals discuss, along with the info and ideas you bring.

Noizyboy

 

[DF96]

Even without flux memory you can still get an initial DC transient current equal to the peak AC current if you close the switch at the zero crossing of the supply. This then decays according to the L/R time constant, so more R at first means it decays more quickly. It is this DC which causes the 'thump' from a transformer when you switch on.

 

[noizyboy]

All interesting topics for my research. I think with short threads posted, it is a more digestable way to learn, as oposed to trying o decypher some principles books. I think my first area of scrutiny will be the soft start and time constants. Start at the beginning eh!
Thanks,
Noizyboy