thinking on an adjustable power supply max 200amps

[M. András]

hi i would like to get some advice on this, idea is that it would be powering and testing some high current motors, power leds etc. maybe multiple rails each up to 50 volts, i thinking on a huge transformer with pwm regulation or a complete switchmode supply, but the problem is i cant find a controller ic neither of these ideas with high enough ratings, can are these be extended with external devices to work on these levels or somehow bypassed, these can work on few amps, voltage range of these would be good for me, i cant find any controllers from any of the manufacturers unfortunately for this idea ( extra thing would  be an adjustable current limit too)

[amspire]

There are a massive number of regulator chips designed to work with external transistors. You should have no trouble finding hundreds.

But 200A at 50V - that is 10,000 watts. Unless you really understand the problems of power electronic design, such a design is a near impossible task.

It is very good if your power supply can achieve 90% efficiency. That means you will be dissipating 1KW of heat in transistor resistance losses and switching losses.

When you have this much energy involved in switching transients, you are taking big time vaporizing of components as you are sorting out the bugs.  This is a big budget project for power electronics design experts.

Where is the 10KW+ coming from anyway? 3 phase mains?

Richard.

[Psi]

have a look at the circuit and design for some of the DIY electric vehicle motor controllers out there.
They normally PWM from 72v to 144v at a few hundred amps.

[NiHaoMike]

What about repurpose a welding machine?

[M. András]

it would be the theoritical maximum power of the different invidual rails, at now simply looking for ideas, advices, if its  possible within reasonable limits so it can change in a wide range, at now with lack of instruments, enough knowledge for these applications (due to missing years from school) and budget, and with the limits of the mains power fuses by the supplier im limited to approx 4KW power on each of the 3 phase (3x16A 230v lines) there is no dedicated true 3 phase connector in the whole building dont know why they didnt made one 30 years ago

as far as i know a welding machine has really poor effiency

[Zero999]

Why not simply but one?

You could also probably parallel several off the self power supplies as long as they support it.

As far as an SMPS is concerned, you need to use a resonant topology, otherwise it'll be inefficient but designing it yourself, is virtually impossible without the skills and test equipment. The easiest way to do it would be to use a large transformer (preferably three phase or even size phase for a better power factor and lower ripple), a variac to vary the primary voltage and a rectifier and smoothing capacitors to give you DC but it'll be very heavy, bulky with poor regulation and ripple.

[hacklordsniper]

I was recently offered 0-50 V 50 A smps adjustable power supplies dirty cheap if my memory serves me well. If this could suit your flavor i will dig the manufacturer data.

Designing a 10 kw power supply is pretty tough job, not that it cant be done.

[M. András]

mains transformer wouldnt be a problem for me, but the regulation  like pwm, once again without equipment at least without a decent scope, i could get the equation from the datasheets of the controllers etc to calculate most of the things, maybe after linear regulation as a second stage regulation, directly with linear regulation after 100watts it would be a nice heater on lower voltage levels, without a multiple tapped transformer. (size does not matter cos i will build the casing myself)

ripple as far as i know can be get rid of with using some L-C filters or multiple ones, i dont think high power leds over 50 watts would be complaining about a few hundred millivolts ripple, motors definitely wont complain about it

about buying an complete prebuilt supply i would rather build it myself and learning from it even more

(about knowledge, i left the school 2 years before graduation and after that it would took 2 years more to get an certificated electro-technician, cos i needed money but when i can i will continue it, i wanted to do electronics since i was 3 years old)

[IanB]

A rule of thumb in power applications is that the current should be no more than half the supply voltage. So for example you might have 24 V and 10 A, or 120 V and 50 A, or 1000 V and 500 A.

Rules are made to be broken of course, and you will of course find exceptions to the above, but those exceptions are taken with care and with great knowledge.

Based on that rule of thumb, a 50 V and 200 A power supply is likely to be very complicated and difficult to build. It is not a project I would find myself considering.

[sonicj]

my 52V >200A motor tester. 

[IanB]

my 52V >200A motor tester. 

Ah, you cheated! 

[ejeffrey]

A rule of thumb in power applications is that the current should be no more than half the supply voltage. So for example you might have 24 V and 10 A, or 120 V and 50 A, or 1000 V and 500 A.

That isn't a bad rule of thumb for a sweet spot if you just need to put out a lot of power and can choose your operating voltage/current -- like powering resistive heaters where you can choose the element geometry to suit, but it isn't crazy to go well beyond that if an application calls for it.  I have a 120 amp @ 5 volt magnet power supply.  I didn't build it myself, the design isn't that complicated.  It is definitely not a beginner project, but once you have cut your teeth on power design so you are confident you can do all the current and power rating calculations correctly on a smaller scale, it isn't crazy.  It is also certainly not the easiest way to get 600 watts if that were your only goal.

[Zero999]

There are the obvious safety reasons for keeping the voltage low, for example you probably don't want to use 240VDC in a 10kW go-kart, it's safer to stick to 48V and suffer the high current.

[M. András]

my 52V >200A motor tester. 

thats a cheat indeed