quietPower, a low noise switch mode power supply

in power supply by DP | 3 comments

Brian from teho Labs┬áhas been working on creating a power supply – the quietPower:

quietPower is a power supply under development at teho Labs to provide a low noise bipolar output. This is essential for creating high quality measurement systems.

quietPower is still under development, when it is finished full schematic and a parts list will be made available.

Via the forum.

This entry was posted in power supply.

Comments

  1. John says:

    Great work. I’m trying to use the LT3439 in a similar design and was quite impressed with your write up. I like your trick using the multi-winding inductor to get the turns ratio you need. You mentioned:

    “PCB layout with respect to magnetic leakage from the transformer turned out to be extremely important to minimize output noise. ”

    How exactly did you change your layout to accomplish this? Is it a matter of keeping the traces as far from the windings as possible or do you need to do something more complex?

    • Brian says:

      I cannot remember where it is but there is an application note. I think it is from LT so look for it where they compare the magnetic leakage for different standard types of transformers. They use a little magnetic pickup (which I don’t have the equivalent of). Anyway it does show you the typical leakage pattern for the normal E type and other transformers. With this knowledge you must then avoid traces that would pick up these fields.

      On the first board I showed in the video I thought I was being good by using two side loading and placing the switcher under the transformer, it turned out this was a bad idea because of the magnetic coupling. I didn’t do any fancy calculations or anything. I just treated the board like it was running at a GHz instead of 50 KHz. Normally you just wouldn’t care about a transduction of -60 to -100 dB but here you do! (The massive wavelength to trace size difference helps but 10 orders of magnitude is a lot).

      The other mistake I made was the diode selection. I used diodes that looked to be in the same series for the rectifier but just had a higher forward current rating. This didn’t work out well. In the end I simulated different diodes in LTSpice (I think). In any case I can’t remember if the one I selected was in the library but I made sure the snap back didn’t produce some massive ns scale transient, that would pollute everything.

      • John says:

        Brian,

        Thanks for the thorough reply! Sounds like a lot of work. When you’re finally satisfied with the design what do you plan on doing with it? I would love to see the layout some day. Maybe you could sell the board as a solder-on module with castellated vias so you could it drop onto other PCB’s.

Leave a Comment

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>

Notify me of followup comments via e-mail. You can also subscribe without commenting.