Power supply over-voltage protection

[rocknrollnick]

Hi all! im currently designing a high current, fixed voltage power supply mainly to run a HAM radio, but also for other applications that need a 13.8v high current supply, but i need a bit of advice for part of the circuit, i need a simple but effective way to shut down the power supply if a fault causes it to produce more than around 15v.
still early planning stages of the build but basic specs are these:
230VAC input (UK single phase)
13.8VDC, 30A (maybe 25A) output
linear power supply whih is based around an old microwave oven transformer

everything else can vary
thanks in advance!

[jeroen79]

You can put an SCR crowbar and a fuse on the output.

[pelule]

Most simple is a CrowBar with a Thyristor, but with 25A/30 @ 13.8V need a strong one.
Example/Ideas: http://www.elektronik-kompendium.de/public/schaerer/crowbar.htm

[NiHaoMike]

Do you have some old PC PSUs lying around? It's pretty easy to tweak many of them to output 13.8V.

[rocknrollnick]

thanks everyone! ill read up on the crowbar circuits, as for a PC PSU its something that i thought about, but people from my HAM radio club have told me they are usually too noisy for radio use, especially cheaper ones, they all recommended building/buying a linear power supply for radio stuff, plus i have to do an electronics project for my next course anyway

[KhronX]

If PC PSU's are too noisy, it might be worth tacking on a capacitor multiplier circuit on the output, to clean it up It wouldn't need to drop much voltage, so it shouldn't need too much heatsinking, despite being virtually a linear regulator.

http://sound.whsites.net/project15.htm

[mmagin]

If PC PSU's are too noisy, it might be worth tacking on a capacitor multiplier circuit on the output, to clean it up It wouldn't need to drop much voltage, so it shouldn't need too much heatsinking, despite being virtually a linear regulator.

http://sound.whsites.net/project15.htm

That probably works great after a rectifier operating on 50/60 Hz AC, but I'm not so sure it would cure the many-KHz ripple in a cheap PC power supply.  Additionally, the interference to a radio from an such isn't just the ripple across the output, there's common mode stuff conducted out the input and output.  (Yes, various country-specific requirements place a limit on that, but a lot of the limits are pretty generous to the manufacturer.)

[T3sl4co1l]

FYI, a zener diode into an SCR is okay, but doesn't work satisfactorily for slowly rising voltages.  A Schmitt trigger should be used, so that the SCR is turned on suddenly.  This can be arranged by using a TL431, a PNP transistor, and some resistors.

Tim

[eblc1388]

 T3sl4co1l has provided you with a great idea. Here is the LTspice simulation.

You would also need secondary protection to isolate the faulty power supply from the mains once the SCR has fired or else the power supply might end up smoking. Usually this is done via shunt tripping the power supply incoming MCB with additional circuitry.

Since you are planning to use a microwave oven transformer, I'm not quite sure the primary fuse will successfully fused as these type of transformers usually have a rather large leakage reactant. 

[T3sl4co1l]

T3sl4co1l has provided you with a great idea. Here is the LTspice simulation.

You would also need secondary protection to isolate the faulty power supply from the mains once the SCR has fired or else the power supply might end up smoking. Usually this is done via shunt tripping the power supply incoming MCB with additional circuitry.

Since you are planning to use a microwave oven transformer, I'm not quite sure the primary fuse will successfully fused as these type of transformers usually have a rather large leakage reactant.

Thanks for drawing that up!

Tim