diyAudio (
-   Class D (
-   -   full or half bridge balanced or not? (

oldroborg 16th December 2005 08:08 PM

full or half bridge balanced or not?
I'm getting into a mess trying to get a full bridge classd working with single rail PSU.
I can get half a bridge working well with no shoot through (i shamelessly used the high side driver from a schematic by jaka rakman & it had alot more control over the high side than my primative attempts :rolleyes: ) had diode pump too, so didn't need extra high side supply.
However when i simulate two half bridges & link them with the anticipated reactive load for a speaker & filter inductors it goes wrong, my mosfets won't switch off quickly if they stay on for long enough anyway (gate drive gets messed up) or the gate drive just goes plain bananas :D
Would i be 20,000x better off using a balanced supply & going half bridge? I'm thinking of using two auto-transformers to produce a balanced supply (a big transformer is hard to source & prolly end up expensive, 4000VA 220:110 convertors are pretty cheap), also get the added benefit of being able to run off two phases rather than 1 phase+ground so i can balance the power usage of the amp better.
I reckon if i'm having these problems before i even get real world pcb & lead impedances in there i'll never do much except modulate the smoke output of the amp with the sub input :(
I need to go back to class-d 101 or chicken out & use gate driver IC's :angel:


poobah 17th December 2005 01:53 AM

My experience with bootstrapping (diode pump) upper gate drives has always been erratic. The bootstrapping is cool when you have very predictable waveforms and you know you can reliably get the chicken/egg cycle going.

Depending on your configuration, your diode pumps may be getting their charge from the lower FET on the OPPOSITE 1/2 bridge... see the problem?

You got a schematic?

Gate drive IC's are your friends and so are nice solid floating gate supplies. The best way to get problems with FETs is to use weanie gate drive. No pride lost in using drivers... that's what they're for...


mzzj 17th December 2005 09:39 AM

You dont need my opinion, you should know already that full-bridge is my way :)

Beware of simulators, sometimes its more tricky to get circuit working on spice than in real-word. forcing small timestepping helps in switchercad, 10nS works reasonably well with 100k modulator freq.

Gate drive transformers offer enough good performance for subwoofer amps up to 50-100khz even when driving real big mosfets.

oldroborg 18th December 2005 10:08 PM

@poobah: Yep, it seems the boot strapping is getting messed up when in the the full bridge configuration. The schematic is very similar to one i posted in an earlier thread (not got the circuit here.)
I've tried many configuations & all have problems (in ltspice anyway) the best i've got so far is to introduce fairly long switching delays & twiddle component values to alieviate dv/dt turn problems (something i wouldn't have with a gate driver :smash:
@mzzj: your full bridge bca topo amp is cool i'm playing around simulating one up now!
I'm anticipating running no higher than 50Khz as i'm intersted in frequencies no higher than 100hz (high cut of 24dB/octave at 60hz) Bascially do the motor controler trick of staying just above audible frequencies

All times are GMT. The time now is 01:27 AM.

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