2.5kV 10kW DC Source Design

[ACHU9539]

I Need a DC source 10kw power and 2.5 kv output voltage .I have a transformer 415/1800.this transformer is specially made .it has 3 seperate sigle phase output windings in output side.my idea is to connect this 1800v to rectifier module and 4500v 1500uf capacitor .

my question is how to protect this source from short circuite?any protection logics available for this higher voltage with low cost. This setup is for temparory.please give me threads for making it
how to protect capcitor and diode from short circuite?
how to improve this idea?

 

[KlausST]

Hi,

Maybe a dumb answer... but I´d simply use a suitable fuse at the output.

Klaus

 

[BradtheRad]

Your secondary windings are wye or star connected. Their common join is ground.
This needs to have a ground placed at the load somewhere, or you get no current flow.
In simulation I find I must create a ground at the midpoint of the load. Then I get current flow.

 

[FvM]

This needs to have a ground placed at the load somewhere, or you get no current flow.
In simulation I find I must create a ground at the midpoint of the load. Then I get current flow.

You get current flow if the load is placed across the output capacitor and a 3-phase primary supply is connected. Star point should be floating.

 

[D.A.(Tony)Stewart]

I Need a DC source 10kw power and 2.5 kv output voltage .I have a transformer 415/1800.this transformer is specially made .it has 3 seperate sigle phase output windings in output side.my idea is to connect this 1800v to rectifier module and 4500v 1500uf capacitor .

my question is how to protect this source from short circuite?any protection logics available for this higher voltage with low cost. This setup is for temparory.please give me threads for making it
how to protect capcitor and diode from short circuite?
how to improve this idea?

The biggest problem is your surge power on current driving the low ESR of 20 mF with HV needs a inrush current limiter than bypass time delay relay to shunt that. The energy could be limited by switched series PWM with rising duty cycle to minimize the dI step during dt duration such that dI=dV*dt/L using NMOS on low side of 3mF.

That is 45 kJ or 45 kW in 1 second or 4.5kW in 10 seconds. But low loss if using PWM, so I would choose a 5 second rising exponential PWM ramp for 10kW inrush using a 50 mV current sense for 2A full scale and feedback to regulate PWM towards 100% . It can also be used for Isc protection shutdown on output. Thus 25 mohm current shunt with short loop to Vgain amp to regulate with a Vref.

 

[FvM]

Only 4.7 kJ according my math. Precharge circuit is suggested nevertheless, but it can implemented with power resistors and contactor. Post #1 suggests that a variac is driving the transformer, it can do manual precharge.

Regarding fuses, HV DC fuses are bulky and expensive, thus I think primary 3-phase MCB is the best solution.

 

[Easy peasy]

As to Brad's comment about grounding the common point - you do not need to do this in the real world - he is getting confounded by the simulator he is using - there is a clear current path ...

--- Updated Dec 2, 2021 ---

2 x Relays and resistor inrush limiting is definitely needed here - 1stly due to plain inrush current and 2ndly due to the over-voltage spike that will result from the leakage energy in the Tx ....

 

[c_mitra]

" In theory, theory and practice are the same; in practice, the theory is less like practice than the theory suggests ..."

In theory, Murphy says, Man cannot stand on two legs. We need a prop.

 

[Easy peasy]

Murphy never designed any power electronics ...

 

[BradtheRad]

Yes, in the case of 3-phase (replying to replies to my post #3), it operates even if the common and load are not grounded together. My simulations show this. So as I composed my post, I re-examined the OP to see if whether it mentions 3-phase from the transformer.

I only found the phrase:
3 seperate single phase

Single phase implies not 3-phase. And single phase is how the transformer operates according to his winding arrangement in the schematic. Therefore he has no current flow until he installs a common ground.

If there's a method to pursuade one and only one flux field to build and collapse so as to create 3 evenly spaced waveforms, I don't know what it is.

 

[c_mitra]

Yes, in the case of 3-phase (replying to replies to my post #3), it operates even if the common and load are not grounded together. My simulations show this. So as I composed my post, I re-examined the OP to see if whether it mentions 3-phase from the transformer.

I only found the phrase:
3 seperate single phase

Single phase implies not 3-phase. And single phase is how the transformer operates according to his winding arrangement in the schematic. Therefore he has no current flow until he installs a common ground.

If there's a method to persuade one and only one flux field to build and collapse so as to create 3 evenly spaced waveforms, I don't know what it is.

I think the OP has some confusion about the 3-phase, because what he shows in the diagram looks like a 3-phase supply. He has a 3-phase autotransformer that feeds three phases to another transformer and he uses a rectifier set designed for a 3-phase.

I am not a transformer man but if you want three phase outputs you need to wind three separate pairs on three separate legs- that may look more like 3 separate single phase. I guess the problem begins there,

 

[Easy peasy]

3 separate single phases at 120 degrees apart to eatch other is 3 phase, the OPs sketch is essentially correct ....

 

[FvM]

3 separate single phases at 120 degrees apart to eatch other is 3 phase, the OPs sketch is essentially correct ....

That's how I read the post #1 schematic, but the exact transformer circuit isn't given, can be e.g. wye-wye or delta-wye. The OP should clarify.