Audio mod ZVS? / High Voltage / Forums

Forums

4hv.org :: Forums :: High Voltage

« Previous topic | Next topic »

Audio mod ZVS?

Move Thread

LAN_403

U4R1A

Sat Mar 23 2013, 07:17PM

Registered Member #3505 Joined: Sun Dec 12 2010, 06:03AM
Location: Albuquerque NM
Posts: 108

I got bored and threw together a little ZVS driver. I've seen online in several places that people have audio modulated these guys in varies ways. Of the ones that I have seen there was one that did not require a separate amp and implemented the audio by biasing a mosfet in series with the (+) rail. I've tried to replicate this in a few different ways but I'm getting zero audio output. Would anyone happen to have a schematic laying around that they would like to share or some pointers? Thanks in advance.

-Justin

Sigurthr

Sat Mar 23 2013, 09:47PM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

The way you are attempting with the mosfet in series with the DC rail basically makes the mosfet a Class-A amplifier. Having the mosfet on the + DC rail requires that your audio source is isolated from any common power source that the ZVS might be on, as the mosfet's Source pin will be at the DC bus voltage (minus Vdrop across the Rds). Class-A amps at best waste half the power going in to them as heat.

Aside from that the audio signal you are pumping in to the gate has to have a relatively large Vp-p swing to move the Rds of the mosfet between just before full ON and conduction threshold. Mosfets require about 10V between Gate and Source to fully turn on and generally have a threshold near 4V. So, you would want an audio signal of around +- 2.5V (5Vp-p) and a voltage divider set up to bias the Gate around 7V. This means on full modulation upswing you would see 9.5V between Gate and Source and on full modulation downswing you would see 4.5V between Gate and Source. You're probably not getting any audio out because you either don't have the gate biased properly or you are feeding it the 100mVp-p signal most line level or headphone out jacks deliver.

It's a pretty simple circuit actually, but terribly inefficent and requires a ton of cooling on the Class-A amp. My advice is to place the mosfet in series with the negative line of the ZVS's input (Mosfet Drain to ZVS Neg DC input, Mosfet Source to power supply negative). This way the audio source doesn't need to be isolated.

Microwatt

Sun Mar 24 2013, 01:41AM

Registered Member #3282 Joined: Wed Oct 06 2010, 05:01PM
Location:
Posts: 224

i think i have an idea on how to do it. you need a VCO chip with the gates of the mosfet. the audio feeds into the vco and varies the frequency vs the audio voltage.

Ben Solon

Mon Mar 25 2013, 01:04PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

There are two ways to do it: am and fm. With fm you might as well us a oscillator based driver, because as soon as you change the frequency in the zvs without changing the lc section zvs becomes pointless and inefficient. Which means the only way to do it becomes am. The class a amp method is really no different than changing the series resistance to the zvs to drop a voltage preportioal to the audio signal. I would recommend something along the lines of a class d amp for bus modulation at this power level. The amp through the transformer is all fine and well, but you can only push a transformer so far, a bridge of Fets or igbts and a wire are much more rugged.

U4R1A

Tue Mar 26 2013, 03:36AM

Registered Member #3505 Joined: Sun Dec 12 2010, 06:03AM
Location: Albuquerque NM
Posts: 108

Thanks for the suggestions. I got it working as a class A amp but the power loss makes the design highly inefficient. Ben I may end up trying a class D but what about a class AB? Seems like a good cheap compromise. It there any major issues with the AB amp?

Dr. Dark Current

Tue Mar 26 2013, 08:28AM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

The negative rail of the ZVS driver can simply be PWM-ed by a MOSFET, but you need to put a fast diode from the FET's drain to the positive supply rail and bypass the supply with a cap close to the transistor-diode combo. You might then need to use a dedicated supply for the gates, or PWM the positive rail, but this needs isolated signal to the gate of the FET.

Ben Solon

Tue Mar 26 2013, 12:32PM

Registered Member #3900 Joined: Thu May 19 2011, 08:28PM
Location:
Posts: 600

U4R1A wrote ...

Thanks for the suggestions. I got it working as a class A amp but the power loss makes the design highly inefficient. Ben I may end up trying a class D but what about a class AB? Seems like a good cheap compromise. It there any major issues with the AB amp?

yes, it would work, but in the end you don't get much more efficiency. you're still relying on bjt's dropping power over themselves to modulate the remainder of the power to the zvs. the main benefit of ab over a and b is greatly reduced zero cross distortion.

you need a low impedance analog voltage to supply the rail of the zvs, which imo can only be achieved by class d due to the saturation of the switching devices. as long as they're linear you're eating huge chunks of power just to get your modulated bus.

Dr. Dark Current:

unfiltered pwm won't work, because it would interrupt the oscillation of the mazzilli driver itself. Once you pass over from PWM to filtered pwm, you're left with class d or synchronous buck. chopping up the power supplied to the circuit won't help, you need a low impedance analog voltage.

Dr. Dark Current

Tue Mar 26 2013, 01:21PM

Registered Member #152 Joined: Sun Feb 12 2006, 03:36PM
Location: Czech Rep.
Posts: 3384

ben123324: It doesn't interrupt the oscillation, the series inductor smooths out the current (it must be sufficiently big of course, which it should be anyway for efficient operation of the oscillator). Of course you need to maintain the supply for the gates, so either PWM the positive rail or make a dedicated isolated supply for the gates.

Sigurthr

Tue Mar 26 2013, 03:48PM

Registered Member #4463 Joined: Wed Apr 18 2012, 08:08AM
Location: MI's Upper Peninsula
Posts: 597

I still think the amplifier and series transformer is the simplest and most fool proof way to go. Rewind a mot and grab a cheap $40 motorcycle amp.

Here is mine in operation:

note: still uploading the video at time of edit, so it may take a while to show up.

Moderator(s): Chris Russell, Noelle, Alex, Tesladownunder, Dave Marshall, Dave Billington, Bjørn, Steve Conner, Wolfram, Kizmo, Mads Barnkob