DC Isolation for 9V Battery - Design Help Needed

[Rekinom]

HI all, first time poster.

What I'm looking to do is isolate a 9VDC load. Specifically, I've got a 9v battery supply, and need to isolate one of the loads(200ma max). My best guess is to somehow turn it into an AC voltage, run it through a 1:1 toroidal transformer, then rectify it back to the original 9VDC.

This will be used in an audio circuit, and as such a lower noise potential would be ideal.

And to curb questions relating to the project, the reason for doing this is that I have two audio effect circuits, both digital each with theirown clocks, since each one seems to be tying it's ground to both the audio out and the common for the power input, I would like to isolate one of them on the power side. It currently appears that the clocks are combining creating a high frequency whine(in the 5000hz range).

After spending a lot of time on google, I came across this forum, and thought perhaps someone may be able to help me with the design.

I grasp the basic principles, but after digging through Mouser and Jameco, I'm at a loss at to which specific components to use.
I think I have a firm grasp on the isolation and rectifying aspect, but am unsure of the DC to AC conversion on the front end. Using a transistor of some sort? Does it matter what voltage the AC signal is? Will the frequency ultimately matter? Are there some components that will produce less rf or em interference?

Any help would be greatly appreciated.

Thanks!

 

[berkeman]

HI all, first time poster.

What I'm looking to do is isolate a 9VDC load. Specifically, I've got a 9v battery supply, and need to isolate one of the loads(200ma max). My best guess is to somehow turn it into an AC voltage, run it through a 1:1 toroidal transformer, then rectify it back to the original 9VDC.

This will be used in an audio circuit, and as such a lower noise potential would be ideal.

And to curb questions relating to the project, the reason for doing this is that I have two audio effect circuits, both digital each with theirown clocks, since each one seems to be tying it's ground to both the audio out and the common for the power input, I would like to isolate one of them on the power side. It currently appears that the clocks are combining creating a high frequency whine(in the 5000hz range).

After spending a lot of time on google, I came across this forum, and thought perhaps someone may be able to help me with the design.

I grasp the basic principles, but after digging through Mouser and Jameco, I'm at a loss at to which specific components to use.
I think I have a firm grasp on the isolation and rectifying aspect, but am unsure of the DC to AC conversion on the front end. Using a transistor of some sort? Does it matter what voltage the AC signal is? Will the frequency ultimately matter? Are there some components that will produce less rf or em interference?

Any help would be greatly appreciated.

Thanks!

Welcome to the PF.

The National Semiconductor SimpleSwitcher series should give you what you need:

http://www.national.com/analog/power/simple_switcher

There's a ton of design resources available, and it even looks like there may be some modules available now as well.

They generally run at 57kHz or so, which shouldn't interfere with your audio. If it is a problem, you can use an output voltage of 10-12V, and post regulate with a linear regulator to eliminate the output ripple of the DC-DC converter.

 

[Rekinom]

Welcome to the PF.

The National Semiconductor SimpleSwitcher series should give you what you need:

http://www.national.com/analog/power/simple_switcher

There's a ton of design resources available, and it even looks like there may be some modules available now as well.

They generally run at 57kHz or so, which shouldn't interfere with your audio. If it is a problem, you can use an output voltage of 10-12V, and post regulate with a linear regulator to eliminate the output ripple of the DC-DC converter.

Wow, that's perfect, thank you!

 

[stevenb]

HI all, first time poster.

What I'm looking to do is isolate a 9VDC load. Specifically, I've got a 9v battery supply, and need to isolate one of the loads(200ma max). My best guess is to somehow turn it into an AC voltage, run it through a 1:1 toroidal transformer, then rectify it back to the original 9VDC.

This will be used in an audio circuit, and as such a lower noise potential would be ideal.

And to curb questions relating to the project, the reason for doing this is that I have two audio effect circuits, both digital each with theirown clocks, since each one seems to be tying it's ground to both the audio out and the common for the power input, I would like to isolate one of them on the power side. It currently appears that the clocks are combining creating a high frequency whine(in the 5000hz range).

After spending a lot of time on google, I came across this forum, and thought perhaps someone may be able to help me with the design.

I grasp the basic principles, but after digging through Mouser and Jameco, I'm at a loss at to which specific components to use.
I think I have a firm grasp on the isolation and rectifying aspect, but am unsure of the DC to AC conversion on the front end. Using a transistor of some sort? Does it matter what voltage the AC signal is? Will the frequency ultimately matter? Are there some components that will produce less rf or em interference?

Any help would be greatly appreciated.

Thanks!

The above advice to use a ready made product is a good suggestion.

For understanding purposes, and also in case you are adventurous enough to build your own, I've given a link to information about the flyback buck/boost topology which should be a good design approach based on your need to isolate and to have output voltage nearly equal to input voltage. You'll see that the basic circuit topology is very simple. However, the complexity comes in when you try to do the switching control. You need to implement a controller and develop the control algorithm for regulation, limits and proper operation.

http://nptel.iitm.ac.in/courses/Webcourse-contents/IIT%20Kharagpur/Power%20Electronics/PDF/L-22(DP)(PE)%20((EE)NPTEL).pdf

 

[maimonides]

Before you start designing and building: Have you verified wether isolated power supplies will solve your problem? From what you write, you only need a 9V battery to do this test.

 

[Perfection]

The National Semiconductor SimpleSwitcher series

Fixed. ;)

 

[berkeman]

Fixed. ;)

Thanks Perfection! I edited my post thanks to your heads-up.

 

[Perfection]

You got rid of my awesome joke :(