Parallel bridge rectifiers

Splitting up heaters for less cross talk between left and right channels...

Is there anything wrong with paralleling up a couple bridge rectifiers like shown in the pic below to split the load (heat) a bit between two rectifiers instead of one?

Or should I just use one big chunky bridge...?

it should work OK/

JR

I cannot see anything wrong with it but chunky bridges are cheap enough. Given the regulators you are using it does not need to be all that chunky.

Cheers

Ian

bluebird said:

Is there anything wrong with paralleling up a couple bridge rectifiers like shown in the pic below to split the load

Click to expand...

The diodes on left side of the bridges (from transformer to the common point/gnd) are in parallel, so the load will be splitted  equally only if diodes are perfectly matched (same dI/dV).  In the worst case one diode is loaded 100%, another 0%, so your bridges should be anyway rated for higher current than the sum of both channels.

Use something like 4A 50v bridges and 3A 12v regulators, and you won't need to service the power supply for a long time.

moamps said:

The diodes on left side of the bridges (from transformer to the common point/gnd) are in parallel, so the load will be splitted  equally only if diodes are perfectly matched (same dI/dV).  In the worst case one diode is loaded 100%, another 0%, so your bridges should be anyway rated for higher current than the sum of both channels.

Click to expand...

The bridges are feeding two separate loads so no matching is required.

JR

You already have massive caps and even regulators. There will be zero cross-talk with a shared rectifier bridge.

In my experience there is zero cross-talk in a setup of your size with just single regulator. Problems are elsewhere if you do. These are just heaters after all.

I guess splitting the regulators work from heat distribution point of view, each handling only half the current.

JohnRoberts said:

The bridges are feeding two separate loads so no matching is required.

Click to expand...

Look once again.

moamps said:

Look once again.

Click to expand...

no matter how many times I look they are two separate loads labelled left and right.

JR

Loads are separated but bridges are not.

Kingston said:

You already have massive caps and even regulators. There will be zero cross-talk with a shared rectifier bridge.

Click to expand...

Yeah I hear ya, its just the bridges are on a PCB. Combined 3 amps makes quite a bit of heat even if a single bridge is rated for 5 amps or so.  Dividing it up into two 5 amp bridges allows me to not have to make space for some sort of heat sink on one..

I also have the hi voltage for the tubes divided up this way. Its probably all overkill, but it is what it is.

The problem is the toroid I'm using is mechanically buzzing in circuit.  I've tested a separate (same model)  toroid out of circuit with plain resistive loads matching the current draw and the thing is mechanically quiet as a mouse. So I'm just trying to figure out if this is a faulty transformer or not. Next step is pulling the problem transformer out of circuit and testing it with resistive loads as well.

bluebird said:

Dividing it up into two 5 amp bridges allows me to not have to make space for some sort of heat sink on one.

Click to expand...

Or bend it and bolt to chassis. I've done one instance where the rectifier was external from the PCB because it had to be bolted to chassis one way or other.

moamps said:

Loads are separated but bridges are not.

Click to expand...

The outputs of the bridges connected to loads are indeed separate... no sharing between bridges involved.

No I will not look at it again... you look at it again.  ;D

JR

I have done this in quite a few projects with PSU's built on the pcb where I want them stereo.  For example 2x 1176 in one box.  Never had a problem.

JohnRoberts said:

The outputs of the bridges connected to loads are indeed separate... no sharing between bridges involved.

Click to expand...

Will right channel work now?

Really looks like half the bridge is in parallel and half is not,  4 out of the eight diodes are sharing the load. 4 are not. So maybe your both right...if we're talking about sharing the directional current flow that is.

I'm pretty sure this is not whats making the transformer buzz so we can be done with it now.

Thanks for all the answers

Hey I just wanted to throw this out there. One time I was powering 8 tube heaters and no matter what I did the secondary of the transformer got hot. It was too many filaments in parallel. I had good luck by using a transformer with two secondaries and running two completely different rectifier bridges with only 3,300uF 50 caps, each one to only 4 tubes. Some 0.1ohm R's to slow inrush current helps too. There's a fair amount of inrush there..

All the best,

moamps said:

Will right channel work now?

Click to expand...

actually yes... but half the diodes in the left bridge are doing 2x the work.

JR

"I'm pretty sure this is not whats making the transformer buzz so we can be done with it now." That's something different. if your tfmr is buzzing it could be you have a short somewhere or your tfmr is overloaded.  Is it getting hot? If not try tightening the retaining bolts.

Andy.

JohnRoberts said:

actually yes... but half the diodes in the left bridge are doing 2x the work.

Click to expand...

Moamps is right, there is a paralleling action going on for the negative half diodes in the two bridges.  Both loads would need to be isolated (pos and neg rails) for diode currents to be separated.

Symmetric wiring and traces and bridges would be needed.

I'd be more  concerned about the stress on the regulator at start up if those are valve heaters - and the regulator heatsinking will be a concern just for continuous current if there is a fair bit of transformer winding headroom to cover mains variation.