Combining two positive regulated suppiles for obtaining a a dual one

[Atilla]

Let's say you've a transformer with two secondaries, not a center-tap. Assume a low-power circuit in which current draw is not going to be an issue for any component provided. Preamp or active crossover or similar.

If I had two PSUs - rectifier bridge, cap, regulator, wired to a single, non-center tapped output of a transformer - what would be the cons of having two of those, joining the GND of the first one with the plus of the second, compared to having the proper, inverted layout, with a negative regulator on the second PSU?

I've seen the join-two-positive idea on Tangent's site. Like so - http://tangentsoft.net/elec/tread/pguide.html#dual . It doesn't give me any alarm bells based on my knowledge of electronic components, but I'm sure there's got to be a reason there are negative voltage regulators and designs that use them.

Where is the catch, if any? And consider I've got a box full of positive regulators and no negative ones.

I've always been curious about that, but only recently started a design that would need such a supply.

[sawreyrw]

If you have the parts shown on the reference document, it will work fine. The 2 secondaries must be isolated.

[woody]

Ok. this is from memory but I seem to rember a brief comment about this by Walter Jung back about 1979- 1980. The comment was something like... The hi speed transient responce of a stacked pair of positive regulators will not be as good as using
a positive and negative regulator but this effect is way above the
audio band. I am not claming that is an exact quote heck at my
age it's hard to rember to buy the right brand beer (joke).

[Atilla]

Alright, as long as there isn't any obvious catch that I'm missing, I'm happy with using the above mentioned setup.

[AndrewT]

have a read of jbau's thread on 317/337 regulators.

He claims one specific make of 337 outperforms all the 317 versions.

[jakobsladderz]

Against:
*not everyone has two isolated windings as opposed to a single center-tapped winding.
*You'll need twice as many diodes to make a full wave rectifier (since you're basically making two independent supplies).
*The voltage drop (read as power wastage) through the diodes will be twice that of a positive/negative reg version.
The counter arguments would be:
*better matching of the positive and negative rails (since you're using the same devices, the transient response should be close to identical.
*The voltage not dropped in the diodes would be dropped in the regulator anyway.

If you've already got the transformer then there's no problem either way from what I can see.

Cheers.

[Atilla]

Yes, obviously this won't apply for center-tapped transformers. I mostly have access to dual-secondary ones however, so I often consider dual rectifier bridges anyway.

Thanks a lot for all the feedback!

[Speedskater]

Originally posted by woody
Ok. this is from memory but I seem to rember a brief comment about this by Walter Jung back about 1979- 1980. The comment was something like... The hi speed transient responce of a stacked pair of positive regulators will not be as good as using
a positive and negative regulator but this effect is way above the
audio band. I am not claming that is an exact quote heck at my
age it's hard to rember to buy the right brand beer (joke).

Somewhere in these old Walt Jung articles:

5/07
Sources 101: Audio Current Regulator Tests for High Performance.
Part 2: Precise High Current/Voltage Operation
http://www.audioxpress.com/magsdirx/...a/jung2779.pdf

4/07
Sources 101: Audio Current Regulator Tests for High Performance.
Part 1: Basics of Operation
http://www.audioxpress.com/magsdirx/...a/jung2778.pdf

I think that he writes:
That 2 positive regulators are better than a positive and a negative regulator.
That he LM317 is better than the LM337 regulator.

[AndrewT]

Hi Atilla,
can you separate the Regulated zero volts line from the unregulated zero volts line?

Try powering up each dual polarity half on it's own.

Try powering up both dual polarity halves simultaneously.
Now connect the two zero volts together. Does the bulb come on?

I have drawn this out and tried to analyse what you have and I cannot see the short circuit route for all those diodes confusing my little brain.

But don't give up yet, there are solutions. I know one, there are probably others.

[willy]

I am using a power supply exactly as you described in your first post. I needed a center tap, so I wired 2 12V Xformers together.
So I have 12-0-12 power supply. (secondary voltage of course after rectification) However, the Xformers have to be the same type with same voltage output. Before regulation my rails were about 1/2 V different.

Not recommended I guess - but it worked.