HUM problems one transformer two (or mor) psu's

[kscharf]

Looking back over some earlier threads I see some people claim that they had to use one transformer per psu or they had hum problems. With one transformer per psu or one psu for multiple channels (amps) there would be no problems.

Ground loops were the claimed cause. However, looking at the hookups with one transfomer driving more than one psu showed a common thing, TWO bridge rectifiers were used to develop postive and negative supply voltages. There are TWO ways to hook up a toroidal power transformer. Most of these transformers have two independant secondary windings or 18-25 volts each. These are used with two bridge rectifiers to make the required +/- 25-35 volts to power the '3886T chips. BUT, we can also wire the secondaries in series and use a SINGLE bridge rectifier in a dual full wave configuration. Now the transformer return wire can be grounded, the transformer windings are no longer floating.

Used in this configuration with multiple recitifiers we would isolate the power lines from each amplifier, but have a common return line, which would hopefully prevent any ground loops. It also saves us one bridge rectifier unit per psu (though the ones we use must now have twice the piv rating as they are used in the full wave configuration with twice the voltage across each diode).

My motivation in this research is that it's cheaper to buy one big high va transformer than several smaller ones! Has anybody tried wiring their psu's as full wave instead of bridge using a single transformer?

[AMV8]

Hi

The reason I use two bridge rectifiers per power amplifier ( that is one for the positive rail supply and another for the negative rail supply ) is that normally this eliminates hum. I think you will also find threads from others making the same point.

Don

[kscharf]

The equipment made by the company I work for (Niles Audio) uses the center-taped secondary with a single bridge (full wave x 2) because it helped to eliminate ground loops. In our case we had multiple windings on the transformer, each amp module (two channels or one stereo channel per module) had its' own winding and psu.

However, as all the center taps go to the same ground point, a single winding should in theory be just as good (Niles went with separate windings with the idea of keeping crosstalk between modules down).

[ianpengelly]

Hi kscharf,

Whilst I think I know what you are suggesting, it would be very useful if you could post a diagram of your suggestion such that others can easily comprehend it.

I would like to draw your attention to the fact that a lot of the people that have suffered from this problem have been using the Chipamp.com circuit boards with the CarlosFM rectifier PCBs, which aren't easily changeable from a bridge rectifier to a full wave.

I'll let some of the more informed experts discuss the pros and cons of a full wave vs bridge rectifier.

Cheers

Ian

[kscharf]

Hope the system doesn't 'reformat this'

-------------------------
( / \
) / \
( / ~ \
) / \
( + - / - +\----------------------+
(----------+ | \ / | +
) | | \ / |
( | | \ ~/ ----
) | | \ / ----
(------------------------- |
| | |
------------------------------------------+ gnd (0v)
| |
| ------
| ------
| |
|----------------------------------+ -

This is a "bridge rectifier" unit (contains 4 diodes) used as a dual positive and negative full wave rectifier. Not that the center tap of the transformer secondary is common to both
the postive and negative supplies. Only a single center tapped transformer secondary (or two in series) and a single rectifier unit are required.

[kscharf]

Oh well, it did!
It looked ok on my end, but after sending it the system
put the crlf's where it wanted to.

I'm going to have to draw this and send a gif or jpg or something.

[ianpengelly]

Best of luck getting a sketch done... hopefully it won't make a mess of it like it did with your previous attempt!