3
\$\begingroup\$

Computer science & math major here.

I'm designing a linear power supply to introduce myself to electronics. To keep things cool and cheap, I am trying to limit the amount of power dissipated in the LDOs when delivering high currents (1 amp) at low voltage.

My plan is to use a relay (or maybe a pair of traics) to switch which tap of my center tapped transformer feeds my rectifier and filtering caps to reduce the voltage drop across my LDOs.

Is this a totally misguided idea and need I be concerned about transients caused by the change in current of the secondary coil of my transformer when I switch from the center tap (12 VAC) to the outer tap (24 VAC)?

I've read some articles and done some modeling of the behavior of a transformer feeding into a full-wave bridge rectifier and then into filtering cap as a pre regulator for my LDO (image not mine).

hotlink

Because, as I understand it, kickback transients are proportional to the current flowing across the inductor; I've started to design an "interrupter" using an opto (to detect 0 points of the AC wave) and a couple of logic gates to force the transition to happen when the capacitors are supplying current. Is this enough precaution to protect my puny circuit from high voltage transformer transients?

Imgur link to pdf output from kicad for those interested of schematic without "interrupter".

Wolfram alpha output of "interrupter" logic also for those interested. Where U is high when user needs to use the outer tap, O is the opto (high when AC is near 0v), and C is high when the outer tap is in use.

[edit 1] changed a word.

\$\endgroup\$
  • \$\begingroup\$ I have some bench power supplies that do add/remove transformer windings, however not in such big steps. \$\endgroup\$ – PlasmaHH May 24 '17 at 20:51
  • \$\begingroup\$ @PlasmaHH - how do they change taps under load? \$\endgroup\$ – Steven B Wright May 24 '17 at 20:53
  • \$\begingroup\$ I have never looked, but likely in a make-before-break fashion \$\endgroup\$ – PlasmaHH May 24 '17 at 21:03
  • \$\begingroup\$ @PlasmaHH What's the difference between changing taps under load and switching the full load on or off? Those are both large current changes. \$\endgroup\$ – Paul Uszak May 24 '17 at 21:47
  • 2
    \$\begingroup\$ @PaulUszak: but not so large impedance changes. When you switch the load off, you have a huge voltage spike due to EMF because the current can go nowhere. If you switch in/out some winding you still have a good path for the current to go. This is btw. also done at power distribution substations when the voltage changes outside the limits they switch in/out some windings. Google image for "tap changer" \$\endgroup\$ – PlasmaHH May 24 '17 at 21:52
1
\$\begingroup\$

Is this any good?

schematic

from a:-

PSU

It's circuit diagrams for the Korad (US) /Tenma (UK) series of bench PSUs. Thousands of people have them. They switch the windings as you change output voltage, which you can actually hear quite easily. With two SPDT relays, it might be able to switch in four tap settings. They also automatically switch off the output as you change level. This might be the key to your question.

\$\endgroup\$
  • \$\begingroup\$ I might be missing something but I don't see any switching going on there, looks like they are using a center tapped 48V transformer to make get a pre-regulated +-24V and then tosses that power through a variety of LDOs. \$\endgroup\$ – Steven B Wright May 24 '17 at 21:35
  • \$\begingroup\$ I will however see if I can find a schematic for the original Korad design. Or maybe you could point me to where in the schematics you've provided this switching is happening. Thanks for answering by the way. \$\endgroup\$ – Steven B Wright May 24 '17 at 21:37
  • 1
    \$\begingroup\$ @StevenBWright Sheet 4 - Output board. It shows the 4 switched taps. You have to read the sheets together as it's a large and well spaced circuit. If you do the binary math, there could be 4 output levels. \$\endgroup\$ – Paul Uszak May 24 '17 at 21:40
  • \$\begingroup\$ @StevenBWright Grid square 2B \$\endgroup\$ – Paul Uszak May 24 '17 at 21:51
  • \$\begingroup\$ I''m looking at it now, trying to find where CN9 goes once it leaves the output board. I'm surprised that they can just hook the relay switched transformer output directly onto those transistors. \$\endgroup\$ – Steven B Wright May 24 '17 at 22:05

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.