Continue to Site

Welcome to EDAboard.com

Welcome to our site! EDAboard.com is an international Electronics Discussion Forum focused on EDA software, circuits, schematics, books, theory, papers, asic, pld, 8051, DSP, Network, RF, Analog Design, PCB, Service Manuals... and a whole lot more! To participate you need to register. Registration is free. Click here to register now.

Register Log in

[SOLVED] Master's project: Advice 500 kW rectifier

Status
Not open for further replies.
Eta

Eta

Newbie level 4
Joined
Dec 28, 2017
Messages
5
Helped
0
Reputation
0
Reaction score
0
Trophy points
1
Activity points
65
Dear All,

I understand, if probably most of you will shake their head reading 500 kW, but please bear with my post.

I was looking for quite a challenge for a project for my master's degree, and approached an eccentric professor, who immediately told me he needed a 500 kW 3-phase rectifier nobody is interested in doing. I am fascinated by it.

I have built a lot of power electronic circuits, but never more than 10 kW. The professor insists that 'there is not much of a difference'. However, I am confuzed about the controling, and overall how different such a high power converter is. I understood, that I will have to probably interleave it? IGBT or GaN MOSFET?

I would really appreciate honest advice, on whether this is even possible, and maybe some suggestions on resources I could further use. I have 3 months to complete this project, and would still like to get reasonable amount of sleep ;)
I have plenty of 'theoretical' knowledge, but feel a bit insecure with the practicality of the whole project.

Thank you very much.
 

You will agree that the specification is somewhat incomplete, as it does not mention to what voltage will be made such rectification, which would supposedly be for 220Vac / 380Vac consumption voltage? Or for some industrial application in the kilo volt range? You are aware that it is not child's play :) Without having clear requirements of robustness/safety/cost in terms of innovation to be achieved, could it be said that any solution would be accepted by the academic committee?
 

Oh yes, I am remiss about the specification, I apologize.

This inverter should be tied to the mains, I am not quite sure what the highest voltage level available at the university is, but I am sure this application should not go into high voltage.

This should be a prototype which should provide a working example while fulfilling the requirements. Safety and cost are not really part of the picture.

It should be a learning experience. I understand that my committee is not like a consumer that will have high expectations for me, but nonetheless I would perfer this not to turn into a nightmare. :)
 

Eta said:
This inverter should be tied to the mains, I am not quite sure what the highest voltage level available at the university is, but I am sure this application should not go into high voltage.
Click to expand...

Now I'm confused, you initially mentioned a 'rectifier' and now you are referring to an 'inverter'.
 

Oh, you are a careful reader. I did infact type too fast. Rectifier is what I meant.

(A rectifier and inverter is pretty much the same thing in hardware to my understanding - doesn't power flow simply decide what it is called?)
 

A rectifier and inverter is pretty much the same thing in hardware to my understanding - doesn't power flow simply decide what it is called?
Click to expand...
That's true if the rectifier is designed as active front-end (AFE), featuring high power factor and bidirectional operation. Any specifications?
 

What do you plan to do with the 500kW of power you rectify?

Fancy GaN devices have no place in such high power, low speed applications. You'd be looking at silicon rectifiers/thyristors. Many of them, in puck packages. You're looking at tens of thousands of dollars in components....
 

I would bet you have no strong background on this matter; listen, a system of such magnitude is not that simple, particularly considering the deadline you said, otherwise you would think twice before considering choosing that to work with. Take a look on some concepts such as Input Power Factor Correction, as well as Synchronous Rectification and you'll have a better sight of the amount of work that is expected to spent.
 

It should eventually be used for charging batteries.

I understand the high cost, but there are many opportunities for funding research.

Thanks for your voices though. I have definitely comprehended that I have not enough information about this project and will need to spend more time on constraining it.

As a young grad student with no particular work experience I cannot deny your claim, but I will try to grow in this field as it is of particular interest.
 

Hi,

You need to specify voltage, (switching-)frequency, current and current waveform.
Is it full wave controlled, phase angle controlled or something else?

Klaus
 

A 230/400V 180kW AFE can be run with a 1200V/450A IGBT Sixpack module. Respectively 1200A dual modules could be used for a 500 kW mains input. But in a lack of detail specifications it's not clear what's the right topology for your project.
 

Eta said:
It should eventually be used for charging batteries.

I understand the high cost, but there are many opportunities for funding research.

Thanks for your voices though. I have definitely comprehended that I have not enough information about this project and will need to spend more time on constraining it.

As a young grad student with no particular work experience I cannot deny your claim, but I will try to grow in this field as it is of particular interest.
Click to expand...

Okay I'm guessing this is for some sort of grid storage application...

Even regardless of your experience level, 3 months is an insane timeline for this. 3 months is maybe enough to construct a proper lab for performing such work, which is necessary if you want to avoid burning down a building on your first arc flash.
 

3 months is maybe enough to construct a proper lab for performing such work, which is necessary if you want to avoid burning down a building on your first arc flash.
Click to expand...
Or as a different approach, fix the circuit topology, design a schematic, simulate it in suitable tool like Plecs or Matlab power electronics toolbox.
 

  • Like
Reactions: FvM

    FvM

    FvM

    Points: 2
    Helpful Answer Positive Rating
The most straightforward approach is a 12 pulse rectifier off 5 transformers 11kV - 480V phase to phase - this could be designed in two weeks with the build time for the rest ...
 

500kW of power will not be available in the lab wall; you need to work close to a substation. The final output voltage is important: for example if you are producing Al by electrolysis, that will produce about 50Kg of Al per hour. But electrolysis will need low voltage and high current (10kA is common in many industries).

For charging batteries, you will need again low voltage at high current. You can of course put them in series but perhaps you can go only so far...

A single silicon rectifier will not be able to handle 10kA and you will need several in parallel. Mercury rectifiers are common at very high currents but they will also need massive cooling.

At 500kW (and say 480V) you will need to handle 1000A; most likely the supply will be 11kV (three phase) or higher. And you will need to provide control electronics.

HVDC transmissions are getting popular but 11kV is not really high voltage in that context.

These are rather turnkey projects; each project is specific for a purpose.
 

Thanks all for your replies, I respect your efforts in offering help.

I will definitely have to see the professor and figure out exactly where he wants to go with this project.

An approach through simulation and lower power to get the controling interface stable and existent sounds more reasonable at this time to me.

I will keep in touch and open a new thread as soon as I have acquired more information, as well as condensed my question into a more streightforward issue which is not as ambiguous as this one.

On this behalf I best believe this thread may be closed.
Eta
 

Hi,

For a chemical industry we design a part of the electronics.
It is up to 3000V, 6000A RMS but limited to about 1MW.

Klaus
 

For a chemical industry we design a part of the electronics...
Click to expand...

High voltage but "low" current, but in the range of around 500kW, comes the electrostatic precipitations that are often fitted to smoke stacks to remove particles. My memory is thoroughly rusted but they too remove about kgs of stuff from the **smoke** per hour or so...
 

Status
Not open for further replies.

Similar threads

Part and Inventory Search

Welcome to EDABoard.com

Sponsor

Back
Top