A serious power supply

[nick.sek]

Does any one have a link to a really solid schematic for a variable linear power supply? As I want to build my own or a least get a very serious kit. I'm looking for something with relays to engage the higher voltages as well as constant current abilities. If any one could post a link I would be extremely grateful.

Thanks

Nick

[BravoV]

Though still in early stage, just watch closely this thread here -> https://www.eevblog.com/forum/projects-designs-and-technical-stuff/general-purpose-power-supply-design-7488/

[alm]

Define 'higher voltages'.

This ELV design (good opportunity to practice your German reading skills ) uses a voltage doubler to reduce dissipation at lower voltages. Another design is tap switching. They also sell kits, no idea what shipping would be like to whatever country you live in. Both are solid designs with variable voltage and current limits. HP published a power supply design handbook (AN90B), not sure if it discussed tap switching, but it discussed many other power supply principles.

Amspire's design is also very interesting, I almost forgot about it since that thread wasn't updated for a whole week . Less conventional, and aimed at lower currents. I wouldn't start building yet if you don't have a solid understanding of this topic, and it will be a while (if ever) for kits to become available.

[free_electron]

why relays for switching? the really good ones use a thyristor controlled bridge to do this ... if you want to see a really good schematic go get the servicemanual for the E3610 power supply of Agilent. they have it onine. that is a proven design.

[david77]

Have a look in this ongoing thread:
https://www.eevblog.com/forum/general-chat/lm350-power-supply/msg109454/#new

[TerminalJack505]

why relays for switching? the really good ones use a thyristor controlled bridge to do this ... if you want to see a really good schematic go get the servicemanual for the E3610 power supply of Agilent. they have it onine. that is a proven design.

One reason is that you are going to lose at least 2V through each of those bridges.

[free_electron]

You lose 2 volts in your classic rectifier anyway. Replace two diodes with two thyristors and your losses remain the same.

If loss is so important then use mosfet to do the switching. almost zero loss. this is what the higher power agilents do.
besides, the thyristor trick has another advantage ( besides the clickless switching ) you can play with the phase angle of iginiting the thyristors and use this as a tracking preregulator... this reduces the heat generated in the pass transistors substantially as you have to 'burn-off' les voltage. it is especially efficient at high current - low voltage settings.

[TerminalJack505]

Ah, I assumed there would be a bridge for each tap.  I'll have to check out the design.

[free_electron]

nope. they use the thyristors in the positive chain. the 'bridge' is made with 2 diodes and 4 thyristors. there are variations where they use 2 diodes and 2 triacs.
transformer has a secondary with a center tap.
One end has 2 diodes , center tap has a triac , and the other end has a triac too. they just apply fase regulation to the triac on the center tap for low range. if you go higher they stop the triac on the center tap and start doing fase chopping on the triac at the end.

[Zero999]

One advantage of a relay is it can be used with a twin secondary transformer to switch from series to parallel so double the current can be drawn, at lower supply voltages. This of course means the power dissipation would be higher but it provides more flexibility.

[AlphZeta]

I just built one myself, including the relay on the input side. This power supply uses an LM338 and can deliver 5A from 1.25 to 20+ volts. If you are interested, you can check out my build log here:

http://www.kerrywong.com/2012/04/30/5a-lab-power-supply-with-digital-readout/

[nick.sek]

Thanks everybody for the input. I'll let you guys know where I stand in a couple of weeks on designs and building!