EEVblog Electronics Community Forum

Electronics => Projects, Designs, and Technical Stuff => Topic started by: TheCrazyRichard on April 05, 2013, 01:09:39 pm

Title: A random benchtop PSU idea
Post by: TheCrazyRichard on April 05, 2013, 01:09:39 pm
Hi!

I've been bored in the school again, and then I suddenly got another crazy idea.

What about combining a switchmode PSU as a first stage, with a linear PSU behind it?
Since the switchmode ones are high efficiency, but also contain a lot of ripple, why not add a linear regulator there to get rid of the ripple there?

So it might still be something like a mid-efficiency PSU with a smooth output, right?


What do you think about it?


I might actually implement it to my PSU when I get to building it...

--Richard
Title: Re: A random benchtop PSU idea
Post by: kripton2035 on April 05, 2013, 01:23:55 pm
it has already been done here fr example
a switchmode psu standard followed by a regulator module.
http://www.tuxgraphics.org/electronics/201005/bench-power-supply-v3.shtml (http://www.tuxgraphics.org/electronics/201005/bench-power-supply-v3.shtml)
Title: Re: A random benchtop PSU idea
Post by: Ghydda on April 05, 2013, 02:02:05 pm
There is one caveat to that solution though.
The open loop bandwidth of the linear regulator must be ludicrously high in order for the regulator to have a fighting chance of suppressing the ripple-noise.
Title: Re: A random benchtop PSU idea
Post by: smashedProton on April 05, 2013, 02:45:44 pm
You can make the smps follow a couple of volts above the ldo.  That Is a tracking prereregulator.  Watch daves video
Title: Re: A random benchtop PSU idea
Post by: smashedProton on April 05, 2013, 02:45:52 pm
You can make the smps follow a couple of volts above the ldo.  That Is a tracking prereregulator.  Watch daves video
Title: Re: A random benchtop PSU idea
Post by: C on April 05, 2013, 03:57:33 pm
If noise is important, You could do a "gatling gun" type of smps where each noise pulse becomes smaller and a higher frequency.

Why not use some active noise cancellation?
For a lot of RFI filters you see a transformer where one winding is in the positive side and one winding is negative side to cancel noise.
That transformer should also be a time delay.
so add some anti-noise to the output of that transformer.

So what is good and bad points of idea?
C