diyAudio - Dual supply using 2 transformers +170-0-170

diyAudio (http://www.diyaudio.com/forums/index.php)

- Power Supplies (http://www.diyaudio.com/forums/power-supplies/)

- - Dual supply using 2 transformers +170-0-170 (http://www.diyaudio.com/forums/power-supplies/215650-dual-supply-using-2-transformers-170-0-170-a.html)

Dual supply using 2 transformers +170-0-170

I am building an amp that requires +170 and -170v rails for b+ on the output tubes. The circuit calls for a typical dual power supply using a 230vCT transformer and a single bridge rectifier.

I have a stock of 115v transformers. Is there anything wrong with me using 2 115v transformers and 2 rectifiers to create this circuit creating a 0v reference to ground by connecting the + out of one bridge to the - of the second bridge creating my +170-0-(-)170?

I have seen some discussions where the solution proposed is to parallel the 2 transformers to create something equivalent to a ct transformer feeding one rectifier. Cost of an additional rectifier is a non issue. What are pros and cons of the 2 configurations. Best sound quality is the ultimate objective.

Quote:

Originally Posted by wlowes (http://www.diyaudio.com/forums/power-supplies/215650-dual-supply-using-2-transformers-170-0-170-a-post3081931.html#post3081931)

I have seen some discussions where the solution proposed is to parallel the 2 transformers to create something equivalent to a ct transformer feeding one rectifier.

That is the best configuration, for a number of reasons, like this one for example:

Quote:

Now about the rail sag:
With a dual bridge, each secondary is assigned to its own supply polarity, eg. the top one cares for the positive and the bottom one for the negative.
But this means that when one of the secondaries has difficulty supplying a current surge, it is alone in coping with it: the other cannot help it and has to stay idle.
By contrast, with a single rectifier bridge, both secondaries are allowed to cooperate for both polarities: they are in effect paralleled, and the voltage sag is halved in this case.
This has to be qualified: when the music signal has a high frequency, the effect is not noticeable because of the filter caps.
But for frequencies lower than the mains frequency (50 or 60Hz), the effect comes into play, and improves the stiffness of the supply

Quote:

Originally Posted by Elvee

Quote:

Is that true? The secondaries are still effectively in series, not parallel; the advantage with a single bridge is one diode drop instead of two. So slightly less sag, but for a quite different reason. Where did you read that?

As I understand it, the reason for using dual bridges when one would suffice is to enable better grounding arrangements for the caps. Having said that, I'm not sure I'm convinced of that anyway.

There is an alternate solution using a dual diode-dual cap voltage doubler, if the current is low, and so , omit one less transformer and its interaction.

Your method is the best way. The dual bridges mean that the transformer secondary is not directly connected to circuit ground. It is only connected when the diodes conduct. That reduces line induced noise. Although that is not normally an issue with vacuum tube amplifiers.

There are a few other advantages also, but that is one of the biggest.

You will want to try switching which primary leads go to the AC hot. That will also affect noise level.

Quote:

Originally Posted by simon7000 (http://www.diyaudio.com/forums/power-supplies/215650-dual-supply-using-2-transformers-170-0-170-a-post3082380.html#post3082380)

I´m not sure if you are referring to my post, but using a full wave doubler, secondary is at HF ground via the output caps, and if half wave, it is directly to gnd. I used it some time ago with SS rectifiers and 2 * 270µF for a regenerative AM receiver. But it induces high AC ripple in the regenerative detector, so I rewind my trafo for an 6X4 dual diode tube rectifier. For this reason I explicitly told "when load current is low".

Thank you. Lots to consider. I should have mentioned that I am using fairly overspec toroids at 5A ea. Each is driving a pair of 6C33C so not light current.