Sinusoidal inverter

[vielle568]

I've a 5KW sinusoidal inverter that converts 240VDC to 220VAC and recently it has ceased to function. There was a smell as if some component had burnt out but I could find no charred remains in the unit. Sadly I have no schemetic and the manufacturer (Chinese) won't supply any details.
Can someone please explain how this thing works and what I should do to try and fix it? It isn't the old form of inverter with a motor driving a generator; it's more recent than that. There's a huge transformer in the chassis and some relays but as far as I can see all the other components are solid state. There are also some PCBs that appear to control the LCD display panel. Normally the inverter has a start up cycle; it takes about a minute to come up to speed before the output power is obtained. Now all I get when I press 'start' is the sound of a relay switching followed by a very unstable and feeble sounding alarm signal.
Any help would be gratefully received. Thanks.
Vielle568

 

[crutschow]

Generally high powered sinusoidal inverters use pulse-width modulation to chop the DC to generate the AC, which then drives the transformer primary. The transformer likely provides isolation between the DC and the AC. But knowing what failed in such a circuit would be difficult without a schematic.

Since you smelled a burned component you might try checking the high powered semiconductors for a possible failure.

A sharp picture of the circuits might be helpful.

 

[RCinFLA]

Look on the bottom of the power MOSFET's as the leads come out of the epoxy case. You may see some grey smoke residue.

The way it works is a high frequency PWM chopping on primary side of transformer. The leakage inductance of the large low frequency transformer along with a AC capacitor across secondary winding filters the high frequency PWM out, leaving a relatively clean sinewave. There may be an additional L-C filtering on output to further improve sinewave output.

The MOSFET power driver is an H-bridge. There are two common design problems with this configuration. One is not leaving enough dead time in the MOSFET switching to prevent pull up and pull down MOSFET's from being on at same time. Second, which is tougher to solve, is not allowing PWM offset to gradually build up a DC offset current on the transformer.

Both result in power MOSFET's burn out.

 

[vielle568]

5KW Sinusoidal Inverter

Thanks everyone for the replies to my questions about the sinusoidal inverter.

I've taken a couple of photos of the unit so that that layout will be visible. As I said I've no schematic to go by on this one so I've not much information to offer. The two photos show the different sides of the unit; it's about three feet square to give you an idea of size. The transformer is mounted on the base of the unit and there's a huge fuse in the center of one side (no, it's not burnt out) but other than that I'm not sure of the function of the other large components.

I do understand your explanation of how the unit functions and what the probable cause of the problem is, but I'm more familiar with small PCB mounted components and not these large solid state power devices. When you talk of a MOSFET I think of something like a transistor but I know that I am in the wrong realm!

If someone could point out what's what from the photos I'd be most grateful. Thank you.

Vielle568

 

[BeeBop]

Your H-bridge is on the upper right in the first photo. Those look like IGBTs (Insulated Gate Bipolar Transistors) between the fans. Below them looks like an inductor (filter) and a contactor (relay) on the left, next to the green power resistor. On the right of that contactor looks like a rectifier, and below those a line filter. On the upper left are five large electrolytic capacitors and a smaller transformer.

The boards control much more than an LCD; they are running the inverter. I'm guessing the microcontroller is the 40 pin DIP on the board in the upper right in the second picture.

Is this made by Toshiba? Oh, never mind that - I see in your first post it is Chinese made. Few of the manufacturers of these things are of any help.

 

[audioguru]

It appears to be very well made.

 

[legal]

modified square wave inverter

hi guys have been designing a couple of inverters using the sg3525 with the feedback function.now am tryin to incoperate an igbt dual pack module how do i go about it?

 

[RCinFLA]

Looks like just a push-pull design with two insulated gate bipolar transistors. Not sure with the third smaller one at upper right is for.

The two caps with brown wires are AC output filter caps. The three caps in parallel at far left are filter on DC power supply.

At least check the DC fuse on the large flat wire above the middle-left fan.

 

[vielle568]

5KW Sinusoidal Inverter

Hello Again,

Thanks everyone for explaining about the inverter; I get the idea of how it works and what is probably causing the failure.

I’ve been looking at the H bridge and I’ve been able to get the part numbers from the power transistors (or whatever they are):
2MBI150N-060 150A 600V 73007 Japan FE

Both power transistors in the bridge have the same part number and maker but for some reason the barcodes are slightly different.

Anyway, I’ve found a couple of suppliers but before I replace the parts is there any way that I can check that the components in place at the moment are actually faulty? As you can see from the photo there’s no obvious sign of any burning and I’m not sure how to determine what has caused the failure. You can blow the photo up quite big and I don't think you'll see any damage around the terminals, but there again I might be looking in the wrong place.

The black component to the right of the H bridge seems to be a three input bridge rectifier. It takes the input coming from the battery bank and must filter out any ripples that are left from the generator that is used to charge the batteries (a wind turbine). I’ve checked out the fat fuse on the left of the H bridge and it seems to be OK.

Any suggestions on how I can check out the power transistors?

Vielle568

 

[RCinFLA]

They are insulated gate bipolar modules with two devices in each package. This explains the strange hookup and the two modules form an H-bridge.

These puppies are likely quite expensive and I would not buy replacements without some trouble shooting to be sure they are cause of failure.

 

[BeeBop]

I agree with RC; not only are they expensive, but also quite rugged. I would do some probing and measuring before ruling anything out.

Are there signals on the gates? Is the micro-controller running?

You haven't mentioned batteries; what kind of battery packs are there, and can you confirm that they are being charged?

You did say there is an alarm. This could mean the unit is finding something it does not like: low battery, over or under current or voltage, thermal issues....

 

[vielle568]

The inverter is connected to the output of a battery bank that consists of twenty 12V 150AH duty cycle batteries. There's a system controller that regulates the charge coming in from the wind turbine and solar panels and the battery level is generally in the range of 240V to 260V. At present, the batteries are still charged and are at 254V. If the wind turbine is let to run I can confirm that a charge is monitored on the system controller display panel. The charge level depends on the wind speed but it can go up to several amps on a breezy day. A typical charge level would be between 1A to 2A. The input from the solar panels in low as I only have ten 30W panels where as the wind generator is 3KW.

Alarms are fitted in the system controller to indicate over voltage and under voltage. When over voltage occurs a relay trips and switches and shunts the power to a dump load. Low voltage triggers an alarm; if it goes down below 225V it will cause the inverter to stop working.

Normally the inverter is set to run at 240V or above. The alarm in the inverter starts beeping if the volatge level starts to get low. If the level is too low the alarm will be continuous and the inverter will cease to function.

 

[RCinFLA]

It sounds like this a hybrid battery based grid tie inverter used to push solar/wind generated power into the grid. These systems must syncronize to grid, matching phase and voltage to allow the inverter to run in parallel. With batteries the inverter should run and supply power when the grid goes down.

With solar/wind power pushed to grid or used for UPS uninterrupted power source the inverter runs continously either to push power to grid or supply power when grid goes down. These type of inverters are usually bidirectional where they can supply AC power or charge batteries from grid power.

The inverter will not engage with grid if it is unable to syncronize with grid for some reason. If inverter is capable of supplying AC power without grid, first check is to see if inverter is operating with grid absent. If so then the problem is syncronization with grid.

 

[rfoperation]

thanks ..But this circuit schematic diagram?

 

[rfoperation]

Dear, vielle 568 I want make a pic controlled sine wave inverter.But,I don't know professional circuit schematics..Help me please..Thanks.

 

[rfoperation]

This project schematic diagram and pic source codes.Where is this download links..Thanks..