diyAudio (
-   Class D (
-   -   Full-bridge vs half-bridge Class D amplifier? (

eem2am 26th October 2011 09:22 PM

Full-bridge vs half-bridge Class D amplifier?
I would have thought that if you are supplying a Class D amplifier with an SMPS.....then it would be far better to make the Class D amplifier a Full Bridge Class D amplifier?

..because it makes the supplying SMPS much simpler in that it only has a single rail output.

Also, unless the audio power is very very low , a full-bridge class D amplifier is cheaper than a half-bridge class D amplifier.....since with a full-bridge class D, your using the full rail voltage....not just half of it.?

ssanmor 27th October 2011 10:35 AM

The design and feedback of a half bridge is easier, and making a SMPS symmetric output is very simple, almost no additional components needed.
For high power (>700W or so), full bridge is preferable for a Class-D amplifier (or using 2 complete half-bridge amplifiers in bridge-tied-load configuration).

eem2am 27th October 2011 11:03 AM

but surely an smps of ~300W with +/- outputs will be half/full bridge or forward...and need expensive coupled output inductor?

ssanmor 27th October 2011 11:33 AM

Yes, I agree, it is highly recommended (but not so expensive)

gmarsh 27th October 2011 11:42 AM

Hang on, aren't you a SMPS designer? You should know this is easy.

Coupled output inductor isn't expensive - you don't have to change the core or the amount of copper. If your inductor is made by winding several strands of wire in parallel on the same core, as it typically is at a ~300W power level, just give half of the strands to one winding, half to the other winding, and change how the wire terminates to the PCB so you can hook up both windings. Done, probably didn't cost you anything.

If your power supply pushes the main transformer one way (flyback, single transistor forward, etc) and has a single output winding, you'll have to add a second winding. But since the power level remains the same, it's just a matter of "giving half the strands to the second rail" as with the inductor. This change is probably free too.

Double the rectifiers, output capacitors, etc. for the new output rail and change the output connector to one with a spot for the new output rail. You're spending money here, but it's not going to be much.

By making those changes, you've made the class D amplifier a whole lot simpler since it can be a half bridge. Multiply your simpler amplifier across potentially several audio channels and you've saved a ton of money.

ssanmor 27th October 2011 12:04 PM

Totally agree with gmarsh.

eem2am 28th October 2011 08:02 AM

I am sorry but a coupled inductor is a custom wound component.

-Somebody has to sit there and fidget with magnet wire and wind it nice and evenly to reduce leakage.

Two separate output inductors is better, because you can likely get them as off the shelf parts.

Guitar amplifier sales volumes are always low.

-custom wound components are to be avoided at all costs.

(it youre doing just DIY then admitted its fine though)

ssanmor 28th October 2011 08:51 AM

Just the opposite...
When you produce commercial SMPS you usually design the magnetic components to be very optimized and hence standard components are almost never used. So you will need your inductors to be custom wound anyway, and the cost of a coupled choke is almost the same as the equivalent power single choke.
For a DIYer it is sometimes more dificult to wind inductors (or have them wound) properly so they usually stick to standard pieces.

gmarsh 28th October 2011 11:24 AM

The main transformer has to be a custom wound component too. Most companies just buy the main transformer, output inductor, PFC inductor, auxiliary supply transformer and whatnot as a single "magnetics package" from the same shop.

Plus, leakage inductance in a coupled inductor isn't an issue. Think about it - if you're using two separate 47uH output inductors, which is a perfectly acceptable thing to do, you've effectively built a dual-winding, 0uH inductor with 47uH of leakage inductance per winding. So you can screw up the construction of the coupled inductor beyond belief and it'll still be an improvement over separate inductors.

ssanmor 28th October 2011 11:56 AM

I agree again

All times are GMT. The time now is 06:57 PM.

Search Engine Optimisation provided by DragonByte SEO (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 18.75%
vBulletin Optimisation provided by vB Optimise (Pro) - vBulletin Mods & Addons Copyright © 2018 DragonByte Technologies Ltd.
Copyright ©1999-2018 diyAudio