bridge rectifier configuration on secondaries

[alibear]

Hi everyone, I am constructing a 36-0-36 supply for a Symasym and am using a transformer with two seperate 25 v secondaries.I want to use two bridge rectifiers, one for each channel, with their individual filter capacitors.
Is it best to connect one secondary to one bridge and the other secondary to the other and join the +ve out of one bridge to the -ve out of the other bridge to give me my 0volts ground, or is it better to take the centre point of the secondary windings straight to my 0v and connect both bridges in parallel to the remaining two secondary connections on the transformer.( as per ROD ELLIOT). I hope I have explained myself clearly.
Also could you tell me which colours of secondary should be regarded as the centre tap if I use the second option above. The data sheet for the transformer reads: RED = V sec
BLACK = 0 V
for one secondary and,
YELLOW = V sec
ORANGE = 0V
for the other secondary.
The transformer is made by NUVOTEM and is RS part No. 257-5203
Thanks for your help
Alan

[sofaspud]

I always see the ground formed straight from the CT. I couldn't give a good reason, though it seems there would be one. If nothing else, why have diode drops there if they can be avoided? The CT on your transformer should be formed by connecting the black and yellow wires.

[wintermute]

It's an interesting question alibear. I've used the two bridges (with +ve out of one tied to -ve out of the other after the regulators for 0V) in my dual rail LM317 power supply, but I've never seen it done on a power amp power supply.

I'm a little confused by your post however, as you seem to be mixing up two different situations.

That is two bridges configured for
A) one bridge for each channel
B) one bridge for each rail.

or are you thinking of using four bridges in total? Ie one for each rail and one pair for each channel?

or when you say channel do you actually mean rail (ie +ve and -ve rails as opposed to left and right channels?)

Tony.

[DF96]

If you put one bridge per secondary then put the outputs in series you have unavoidably got one PSU so only one set of caps is needed.

By joining the secondaries to form a CT sec and then using the bridges in parallel to separate sets of caps you will have a small degree of isolation between two PSUs - not separate but not totally joined either. In either case you need to be careful where the charging pulses end up, and ensure that you don't have either a hum loop or two channels coupled via a dirty ground.

[alibear]

Hi all, thanks for the replies, sorry my post was confusing.
Wintermute, I meant one bridge for the right channel, one for the left.
DF96 again sorry for the confusion, if I use 1 bridge per secondary then obviously I could use one set of filter caps.
So is it better to use one bridge per secondary and join one + to -ve to create 0volts ground with one set of filter caps or is it better to use the centre tap of the two windings as 0volt and use two bridges in parrallel with two lots of filter capacitors as Rob Elliots ESP sound pages,project04 in his power supply projects section.

Thanks
Alan

[wintermute]

Hi Alibear, the simplist way to do it is to use one bridge, one cap bank and the centre tap for your zero volts line.

some advantages of doing this are:

* lower cost
* less chance of problems (eg hum.buz)
* will need less space.

In that config the +ve of your bridge is the +ve rail and the negative of your bridge is the -ve rail and zero volts is your centre tap (formed by the Black and orange wires on your secondaries.

You can of course do this ala Rod Elliots power supply as well (something I did with my chipamp) with a separate bridge and cap bank per channel. The advantage with doing this is theoretically better channel separation, provided it is done properly you shouldn't get any hum problems. The best approach (for ensuring channel separation) is to use two transformers though.

I just remembered that Nelson Pass uses a bridge per channel with the +ve of one tied to the -ve of the other for zero volts . From memory he said the advantage was that it eliminated the problem of buzzing transformer windings under certain circumstances (perhaps dc on the primary).

Tony.