IGBT >10kw dc to ac inverter

Thread Starter

magudaman

Joined Feb 27, 2012
39
I want to take 400vdc and convert it to 260vac at 12kw continuously. I found some really interesting IGBT inverter controller boards from China that looks really interesting. Basically high voltage dc going through two igbt into inductor. Has anyone messed around with these or similar boards? Seems like quiet a few different versions out there.


US $70.19 2% Off | Pure Sine Wave High-power IGBT Module Control 2000A Driver Board
https://a.aliexpress.com/_mLEFwAe

Or

US $57.00 5% Off | High Power Pure Sine Wave Inverter Drive Board (10-100KW) IGBT Module Drive Board
https://a.aliexpress.com/_mOVpTLy


diagram of the high power side:
0CCBD22D-98E1-4A3F-A1AC-AE65CD48364B.jpeg
 

Thread Starter

magudaman

Joined Feb 27, 2012
39
I've never worked with these,
That's 46 Amps, is your DC supply able to support more than this...?
Depending on how the voltage regulation works on these boards I can feed the circuit with as much as 450vdc 34a, however it looks like some are limited to 420vdc max. This gives me ~15kw of source power. Should be plenty to handle switching losses of my 12kw load.
 

Thread Starter

magudaman

Joined Feb 27, 2012
39
Figured I will update this thread with what I’m doing with this project and more about IGBT driver board that I got.
I received my board and my IGBTs and both look great but have yet to be tested electrically. Here are some photos of my gear!

IMG_7528.jpeg

IMG_7529.jpeg

IMG_7532.jpeg

IMG_7533.jpeg

IMG_7534.jpeg
 

Thread Starter

magudaman

Joined Feb 27, 2012
39
On a with the project!

Before I order any more parts I needed to make sure my 15kw of dc power supplies that will feed to IGBT inverter was going to work for my project. I am using 3x 150vdc 5kw units in series and Pioneer Magnetics wouldn’t get back to about the isolation voltage rating of their supplies. So I load tested each one at about 4500w and then decided to just go for it.

single supply under test of around 4500w
IMG_7449.jpeg
I temporarily put in them in series with back-feed protection diodes and turned turned it on. No explosion. I reconfigured my load for 450v and connected it up to supplies. It works!!

3 supplies in series under load:
IMG_7469.jpeg
 

Thread Starter

magudaman

Joined Feb 27, 2012
39
Where's the 2.5mH inductor? A 2.5mH 52A inductor is not going to be small!
I hadn’t ordered it yet. I wanted to make sure I could get 450vdc before ordering all the parts.
But they are here now too!

I was having trouble finding single inductors at 2.5mh so I decided to do lots in parallel. I didn’t want to do an AliExpress product since they didn’t seem to provide all the ratings.

For inductors I am going to try to run 12 parallel 370uh 5 amp units. At full operating current these only have 200uh so 2.4mh at 60 amps but I will be only pulling around 50a, so hopefully closer to 2.5mh or 2.6mh. Drawing calls for 2.5mh.

IMG_7536.jpeg

Bulk capacitors call for 10,000uf but since I’m feeding from a dc power supply I’m going to try to run the supply with significantly less to start. Pre built IGBT inverters of similar design rated at 10kw have something 1300uf. I am going to be running 4x 500vdc 470uf. I might end up adding two more for a total 2820uf. I guess we will see what the scope shows.

IMG_7531.jpeg

Lastly is the 15uf ac capacitor, which is huge thin film cap
IMG_7535.jpeg

I’m hoping to make some progress on the initial wiring this weekend and will hopefully have some updates! I sure hope all these parts don’t just turn into ewaste on power up.
 

Thread Starter

magudaman

Joined Feb 27, 2012
39
Twelve 370uH inductors in parallel gives 30uH not 2.5mH.
IMG_7537.jpeg
You are correct, I believe I was thinking series inductance but doing that I won’t have the current capacity. Well I guess I will spec and order the correct setup. For now I can get running with 5a capacity and work out some of the other bugs. At least it was $27 for all 12 and not more :confused:
 

schmitt trigger

Joined Jul 12, 2010
903
PLEASE keeps us posted as your project advances. I really hope you can finish it successfully.

I looked at the provided links, and have a concern with the
https://a.aliexpress.com/_mLEFwAe
As it has a programming and display connector CN2, but there is no description of the communication protocol, neither a register description and its functions, etc.
Does the product include additional information to what is presented on the web page?
 

Irving

Joined Jan 30, 2016
3,894
I too am interested in this...

I am planning a simple 5kW battery grid tie inverter/charger. My wheelchairs (active & backup) each have a home built 4.8kW of LiFePO4 battery (200Ah x 24v) - it will pay me to sell to the grid at peak demand and recharge during low demand/high renewable availability when I get paid to use power, or when I have surplus solar (4kW array gives me >22kW output on a good summer's day) - I can even recharge the active chair from the backup. Eventually I want to build a 15+kW house battery for which I will build 4 identical parallel 5kW inverter/chargers
 

Ian0

Joined Aug 7, 2020
9,835
4kW array will give you an average of 16kWh per day through the summer in Britain. Most of our customers have 24 off 300W panels (7.2kW) and our best figure so far this year is 36kWh in one day from our customer on the Mull of Kintyre.
 

Irving

Joined Jan 30, 2016
3,894
At t
4kW array will give you an average of 16kWh per day through the summer in Britain. Most of our customers have 24 off 300W panels (7.2kW) and our best figure so far this year is 36kWh in one day from our customer on the Mull of Kintyre.
At the risk of going OT, my array of 14 panels is not optimal; all are angled 15° off flat roof to meet planning regs on visibility from road, 9 of them face 140°, the rest +/- 90° from that. The best output everwas 23.2kWh, but 22kWh isn't unusual in summer. Average Apr->Sept 2.08Mw or 11.4kWh/day
 

Thread Starter

magudaman

Joined Feb 27, 2012
39
I made a bit of progress yesterday with a full temporary setup but I wanted to post some of the details I found that are critical to testing and design.

First off the all the electric schematic posted so far and on most AliExpress retailers are for earlier versions of the board. The photo below is the for the v4 board. There are some significant changes such as current is now measured from a Hall effect sensor and the voltage sense resistors are now 150k ohm each.
IMG_8231.png
I also found a Chinese forum write up on setup and testing of what appears what an early version of the board. I converted and made it into a pdf and also attached that here. It had a couple good setup and testing steps which I tried to follow, however my board doesn’t seem to be providing IGBT control output.

Ok back to my setup. I got the control board wired to the IGBTs and all the sense lines connected. I’m connected very haphazardly to the cap bank and the small 5a inductors. I powered up the board and tried to measure the IGBT gate signals but can catch anything with my cheapy scope. I was hoping to verify I saw good signals as the PDF I uploaded indicated. I’m guessing the unit is not turning on the output since the high side dc input voltage is too low however I thought it would run since the one sales page says: “The IGBT module drive control board: the default startup time is 7 seconds, the dead zone is 2uS, and there is no output 150V undervoltage protection, which is convenient for beginners to install and use”
But it also says the alternative program is:
“Another program: start time 7 seconds, dead zone 2uS, output short circuit and 150V undervoltage protection, start adjustable constant current output, suitable for mass production and electronic master use.”
no idea how to switch between these two programs.
So basically without it starting up, I guess my next step is to just power the high side up at say 300vdc or so. Unfortunately I currently only have a 7ma 200-1000vdc variable supply and of course my 15kw supply. I need to figure out what I can use for initial power up to make sure it doesn’t blow up. Ideally I would have 0-500v 1 amp variable supply which appears to run about $250 or so. Or maybe I run my 15kw supply with a series resistor to current limit. Well I need to get this thing online before October so hopefully progress will be fast!
 

Attachments

Thread Starter

magudaman

Joined Feb 27, 2012
39
Well crap. Got the capacitors charged up to 350v+ and the unit immediately goes into fault. With such limited information from the vendors I’m pretty stuck. The only other thing for me left to try to is provide some voltage on the current sense lead.

IMG_8437.jpeg

Any good idea from group here?
 
Top