Lab Power Supply 0-30V 0-3A /CV/CC/

[ZeTeX]

(Previous thread that went off-topic: https://www.eevblog.com/forum/projects/pass-transistor-driver-problems/msg903960/#msg903960)


So it's summer break and I'm working the my PSU again, the schematics are pretty much ready! I ordered some parts and I'm waiting for them now.

Anyways at the end I decided to not use switching pre regulator and just have 2 voltages from my duel secondary transformer (0-12V-24V 160VA, pretty overkill really but that's what I have), so I tried to design tap switching circuit without using relay or any mechanical components as they wear out, pretty hard to get in my local electronic store and most important - slow.
So, can I improve the tap switching circuity but using less compoents \ something easier? I'm pretty sure there is a way around U5 without inverting the output from the zener.
*The components that I use in LTspice simulation are not the components that I'm going end up using. 

[Kleinstein]

For using the 2. nd transformer tap, there are two more simple solutions:
1. Use a robust transistor (e.g. like 3N3055) that can stand a high base current. At low output voltages this transistor works as an emitter follower with the low rail at the collector. At higher output voltages the transistor can not work as an emitter follower, but just the base emitter diode is active, so the normal driving stage does all the work. For protection a diode is needed in series with the collector.

2. Use a circuit like a class H audio amplifier: Similar to you circuit, but using a N_MOSFET as source follower or an NPN as a emitter follower, with a constant voltage at the base of gate. So no need for U5. Often just something like 5 V used for the supply of the OPs is about right.

The downside of both methods is that the power dissipation will shift between the two output stages, as there is not hard switching, but linear operation.

[ZeTeX]

For using the 2. nd transformer tap, there are two more simple solutions:
1. Use a robust transistor (e.g. like 3N3055) that can stand a high base current. At low output voltages this transistor works as an emitter follower with the low rail at the collector. At higher output voltages the transistor can not work as an emitter follower, but just the base emitter diode is active, so the normal driving stage does all the work. For protection a diode is needed in series with the collector.

2. Use a circuit like a class H audio amplifier: Similar to you circuit, but using a N_MOSFET as source follower or an NPN as a emitter follower, with a constant voltage at the base of gate. So no need for U5. Often just something like 5 V used for the supply of the OPs is about right.

The downside of both methods is that the power dissipation will shift between the two output stages, as there is not hard switching, but linear operation.

Is there a way to make it so there will be almost no power dissipation at all?

[StillTrying]

Is there a way to make it so there will be almost no power dissipation at all?

This method done by timsu in
 https://www.eevblog.com/forum/projects/linear-power-supply-based-on-hpagilent-e3610a/msg942813/#msg942813
seems to work. Because you don't need the mosfet to switch massively fast you can get the currents down to a few mA.
I simulated it at the time...

[ZeTeX]

Is there a way to make it so there will be almost no power dissipation at all?

This method done by timsu in
 https://www.eevblog.com/forum/projects/linear-power-supply-based-on-hpagilent-e3610a/msg942813/#msg942813
seems to work. Because you don't need the mosfet to switch massively fast you can get the currents down to a few mA.
I simulated it at the time...

Great! Took it to my design:

Seems to be working fine.