[sdiy] TB-303 smps interference

[Richie Burnett]

It's common to get leakage of the magnetic field from a flyback transformer used in a switched mode power supply. A flyback transformer requires a small air gap in the magnetic path, but that causes the field lines to spread outside the transformer itself.

The metal shield you made is what smps designers call a "flux band". The shorted turn around the outside of the transformer kills off the leakage field beyond the transformer. You often see these ribbons of copper foil soldered right around the outside of transformers in high quality switched mode power supplies to achieve compliance with radiated emissions standards.

-Richie,


Sent from my Sony Ericsson Xperia ray

"cheater00 ." <cheater00 at gmail.com> wrote:

>On Sat, Nov 23, 2013 at 5:39 PM, cheater00 . <cheater00 at gmail.com> wrote:
>> Changing the VCO pitch doesn't change anything. However, if I start
>> the sequencer, the waveforms change in frequency a lot! You can see
>> some slides. I'm thinking this must be related to the oscillator
>> pitch.
>
>This idea was incorrect. Using a highly precise measurement instrument:
>
>http://imgur.com/a/idtGJ#46
>
>I have been able to recover the volume envelope from the output:
>
>http://imgur.com/a/idtGJ#54
>
>Then, using the external trigger output in my timebase, I have made it
>trigger (that is, display one "screenful" of the waveform) only on the
>highest peaks of that recovered envelope. (to do this I have changed
>the triggering mode to normal on the 7B53A plugin)
>
>By using the variable storage mode of my oscilloscope mainframe I have
>overlaid all of those triggered waveforms on top of eachother. The
>frequency was stable. There were no slides present. Note that the
>fuzziness is just a general property of the storage mode on very dense
>graphs:
>
>http://imgur.com/a/idtGJ#58
>
>(if the waveforms didn't align /perfectly/, the graph would be very messy)
>
>This shows that the frequency of the ripple I am experiencing is
>related to load (I don't think it's wrong to assume that higher volume
>output = higher load for the power supply).
>
>The next questions to answer are: I wonder what this frequency
>bleeding into the output actually is? Can this be the switching
>frequency of the smps? If so, why would it change frequency according
>to current draw? Can I improve ripple rejection by swapping out any
>decoupling capacitors on the power rails?
>
>Cheers,
>D.
>_______________________________________________
>Synth-diy mailing list
>Synth-diy at dropmix.xs4all.nl
>http://dropmix.xs4all.nl/mailman/listinfo/synth-diy