• Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Grid Tie Inverter Schematic

The circuit is NOT usable to feed power to the Vac, What you have here is a synchronized rectifier but the opposite IGBT antiparallel dione turns on too-- causing a short.

You can not use square waves to feed power to the grid, you need to have a sine wave equal to the grid sinewave and phase synchronized -- in reality you need a DC/AC inverter with a Pure Sine wave with variable level to be just higher just enough to place the current direction toward the GRID .
The basic circuit is quite critical because it needs to detect the Grid drops to automatically disconnect from the Grid -- this to avoid electrocuting the workers trying to bring the Grid power up.

ALSO you need to detect the current to insure that the system is in phase with the Grid to have a very high power factor .
The companies that are manufacturing Grid tied Inverters try to keep their schematic design quite SECRET.


Most Helpful Member
The circuit is NOT usable to feed power to the Vac, What you have here is a synchronized rectifier but the opposite IGBT antiparallel dione turns on too-- causing a short.
In reference to which circuit/circuits being there are numerous ones posted in this thread. :confused:

Some work just fine and have been successfully built and implemented by countless people and others have never worked and will never work. :eek:

Be more specific and take a look at how old this thread is and when its last activity was, you're about a year late for the party! :p
Last edited:
Dear nando37
I`m writing about http://www.electro-tech-online.com/threads/grid-tie-inverter-schematic-2-0.105324/
This project is very valuable, and what is most important, works. Works even without PWM section, not described by tcmtech :(
I`m not sure about relation between power from battery to power detect by the kWh meter (so called efficiency of this GTI), but soon i`ll be able to verify that.

This project is designed to work with very small power (<2kW), and it is not important how bad Power Factor is.
This circuit disconnects from mains in case of power failure because of basic way it works - kind of anti islanding.
General in my country it is forbidden by low to connect GTI to the grid, so things i`m doing, may be illegal!
Therefore i`m using 5kVA UPS system, with resistive load at the output, and kWh meter between this system and GTI. Load is to simulate typical situation in grid = to get excessive power from GTI without damage to UPS.

Last edited:


Most Helpful Member
Thanks Dragon! Don't worry about the negative posts. I have gotten rather used to them by now. TO be honest I rather expect it being these are very much pirate GTI type designs with minimal control and refinement. :p

If you are wondering how I come up with the estimated average efficiency numbers I actually run my test designs right through an actual KWh meter I bought from my local utility company while back feeding the grid and yes I will be the first to admit choice of parts and circuit setup can bias the maximum efficiency point considerably as well! :eek:

What I do is I set the input power at a specific value with an adjustable power supply as the DC power source and just let the unit run for a specific period of time and read the meters results! Its hard to argue with with one of those when its set up that way! ;)

Its also a easy way to determine a units absolute peak capacity, its maximum continuous capacity, and its minimum set point where it should connecect or disconnect from the grid to keep from drawing power when the input power source is too low.
transistorized Half h-bridge driver for mosfets and IGBT

i really love your inverter schematic ,its very helpful.i used the attached circuit for my inverter.
pls i need some calculation on transformer winding, choice is copper wire and nos of mosfet to be used.
Greetings all,

This is my first post, what an interesting forum!!

I have come across this website: http://www.neo-aerodynamic.com/GridTieInverter.html

I was hoping for some constructive comments on this very simple circuit and its explanation. It seems to me that is allows the input DC to add to the incomming AC, via the cap. Obviously the DC only "gets through" due to the alternating current through the cap. It is an interesting circuit. Indeed it would have anti island built in due to the cap between the DC and the rest of the circuit.

On a different level: We here in OZ are scammed when it comes to PV systems. We go into dedt to purchase the system, we get screwed on our valuable "green energy" by only getting a small feed in tarrif (0.16-0.24c) - our feed in is metered separately to our actual used electrical metering. Our energy is then on-sold to others as "green energy" at a much highter tarrif. To make it even worse, our actual household energy is charged at the standard feed in rates. What a CON.

Im keen on a plug and play set up that cuts down MY actual usage. Awesome thread !!



Most Helpful Member
I was hoping for some constructive comments on this very simple circuit and its explanation. It seems to me that is allows the input DC to add to the incomming AC, via the cap. Obviously the DC only "gets through" due to the alternating current through the cap. It is an interesting circuit. Indeed it would have anti island built in due to the cap between the DC and the rest of the circuit.
See post 40 and up on this thread. http://www.electro-tech-online.com/threads/grid-tie-inverter-schematic-2-0.105324/ Its been covered before.

As far as I have ever been able to design and actually prove to be workable the two transformer two transistor two resistor and one diode version is as simple as grid tie inverters can get. Granted its not the most reliable, safe, or efficient but it does do the job.

View attachment 62512
Thanks for that tcmtech,

Indeed, I missed post 40 on that thread. I am not sure that its been "covered" though and I will take a closer look at this idea. I think that the Cap is the real key to this circuit. A concern would likely be the high voltage that might appear at the DC input if that input had a low internal resistance. Since it looks to be a simple build, I will get to work ASAP and post my results.

Cheers and keep up the great posts!
Hi all. First port.

Here in Brazil every house has an direct electrik schower, from 4 to 8 kW each, and some are with dimmer, also a lot of peaks on the grids.
So, many GTIs and assyncronopus generators with some kWs, are bether than worse for us grids, and they also can compensate some existing harmonics...

@Tcmtech it is possible to make the GTI without the main transformer? Fedding 230VDC direct to 220VAC?
hello there, is this blog still on?

Hi guys,

am new to this forum. Pretty much read most of the inputs regarding gti's here on this site. Tried a few ideas, all worked as expected. Wonder if there are any circuit ideas out there to diy-build an H5 or HERIC gti like the ones built by German companies SMA/SUNWAYS. They apparently have 96-99% efficiency. It looks simple as the H bridge is extended by one or two IGBTs, but how to drive them is another story. Appreciate ideas, thoughts or references. Rgds


Most Helpful Member
First for those who are curious as to what an H5 or HETIC GTI is, View attachment High Efficiency GTI for PV Systems..pdf .

Now for the real question. :p

Looks simple and has high working efficiency is usually anything but simple to build for most DIY type fabrication projects and for what work and costs go into highly complex designs they generally don't always add up in the end to be totally justified for the little added energy savings that can be had.

Picking up 10% or less extra GTI efficiency, which can easily be the equivalent cost of what adding 20 - 50% more output capacity to the AE or RE type energy source it's connected to, doesn't make sense most of the time. It might make for a good numbers bragging point but for a person who is truly trying to get by on producing their own power for the least cost in many cases adding more capacity to the source is easier, cheaper, and better cost justified that trying to squeeze a few percent better efficiency out of one device or component.

Typically once you hit around 80 - 90% efficiency on a DIY project such as this every time you cut the remaining losses in half the complexity and design costs double while the chance for failed builds and unexpected breakdowns goes up proportionally as well. On top of that in many ways your personal resources for the added and more complex fabrication steps and related work itself needs to be expanded on and that gets expensive really fast. :eek:

Making that step up from having one or two basic multi meters and a soldering gun to having a function generator, a fair oscilloscope, and one or more variable or adjustable power supplies that work above a few tens of watts output, plus having a micro controller programming station quickly puts the up front construction tooling and costs well out of the basic DIYers reach.

I may not be an expert on all things DIY grid tie but going by what little information and public accessible DIY construction plans and blueprints there are on the internet there may be a good reason why so few else are willing to develop more complex designs than mine and share them.

Making even the basic designs look easy and buildable is suprizingly hard and what you get in return is more fights than praise for doing it. :p
Last edited:
Thanks for your reply, tcmtech. I fully agree with you. I have been looking for these h5/heric diagrams for a couple of weeks now and only adding 1 or 2 addnl switches and adjusting timings on the simulator drove me almost mad. I think it's a fine balancing act and lots of headaches for a few panels. Following this blog, I was hoping that someone had tried it out before. I ordered parts to build your gti, using mosfets. You mentioned that you might later add pwm. I played a bit with pwm, the 555 does not produce triangle waves, more sawtooth I ordered an IC8038 to try it out. But there are also ready built pwm-ICs out there. Is there one you'd recommend? I'd like to download/order your cd that you mentioned earlier. Btw, many thanks for all your posts, I learned a lot here. Rgds


Most Helpful Member
I never finished the CD and probably wont. Since the start of this thread some 4 years ago a lot has happened in my life.

Married, regular job, wife made me tear down and burn my work shed, new house project, stuff like that. :p

What info I have is here on this site spread out over a number of threads.
As far as adding PWM to my GTI circuits honestly its not really worth the work in most cases. You don't gain much other than the main transformer is quieter and a little less line side filtering is needed. Mostly it ends up as more of a headache to build set up and tune and control if you don't know what you are trying to set it for. :eek:

To be honest if you have read all the threads and my responses you have probably noticed that the vast majority of the people who do ask questions don't have the proper knowledge and understandings or resources to build what I have presented so far. :eek:

More power to those of you who have and did though! :D
Hi tcmtech,

finished building your gti on a breadboard. works great and it looks like its doing more than 80 pc eff.

i did a few mods, used 2x 741 opamps, ditched the zeners as they failed (prob. low watt). the zero cross dead band is working great. confirmed on oscilloscope. (igbts used were hgtg20n60a4d 600v/20a , will try mosfets as well)
initially i had a few shorts, which burned a few 7812s. the shorts originated from ir2113. i fixed this with ref. to the datasheet.
looking at the power handling circuit, Vss is connected to Vcc which causes a short, when Sd is grounded through To(I).
basically i removed the connections between Vss an To(D) on both drivers. And also, did not ground To(I).

It worked without a glitch for many hours, with and without load.

Please let me know if I missed sth here, or recommend any other change.

Will do bit of experimenting before putting it all together. pictures will follow soon.

back to heric, i find this concept really challenging. I run a few sims on iCircuit/mac and it looks promising. i bought an arduino mega for this. will brush up my programming skills to see if heric is doable.



Most Helpful Member
Yea a little schematical oops on my part.:eek:

Or a hidden easter egg to set the bar high enough to limit this project to those who are smart enough to read data sheets and not just go by what some second rate draftsman draws up and puts on line. :p

Glad to hear you have one running! :D
Hi tcmtech,

have added some pictures of my mess with the circuit pdf. works well on the board. the 555/zr431 protection was set to 27V and it does the job.

also, the output is earthed (through center-tapped 2x100nF caps), which creates a nice rectangular trace.

I think an LC tank would reduce the output power quite a bit. I ran a few sims where I uploaded a square wave onto a sine wave. The image looked slightly distorted but the power wasn't much affected. What are your thoughts on this?

I haven't uploaded the output to the grid yet. Will need to add some circuit breakers for the worst case.

I really like those Mosfets, much faster than the IGBTs, and they don't even get warm (at 35W).



Hi Selimy

Nice square curves...
I think that they have the advantage of a self regulated tipe, tath is, with less voltage at the imput, the power is less, and with the growing of the imput voltage, the power increase, as a good regulator for a wind or solar generator.
Also your schemes are with complet informations. Very nice, and very good for those wo not are so familiar with some knowleges that others mint to every have to have.
In my opinion, a square or rectangular trace of power under 3 kW, are optimal for all grids, they also compensate much harmonic picks from other devices conected.
In my last experiences I made 4 dc relays as output devices, and I achieved very nice rectangular waves, that I modified with a potenciometer in the feder comand circuit. But, I also must put your mencioned circuit breakers at some places, because wen I put it on the grid, it blows the fuses...


Most Helpful Member
Nice work!

As I have mentioned in most of my GTI threads these designs are very open to modification so what ever components work for you is fine with me. ;)

Relating to the LC tank circuits you are right they are not really necessary but if I don't include them the harmonics nazis have a fit and as Fendal mentioned at lower power levels it's really just an option. :p

Given the amounts of line noise already present in most typical utility power systems from all the SMPS powered devices any GTI undr a few KW isn't going to make anything any worse. If anything given the fact that they are a power returning device not a power using device they may very well have some canceling effects.

So what are you planning to use as your power source?
Thank you, Fendel
now that you mention, it falls all into places. Good to have this for varying input power. When I increased the voltage, the frequency or the phase never shifted. Only the amplitude went up. Up to now I always thought that PWM was a must. I noticed many problems with pwm while simulating and paying with. The biggest issue was the turn on/off rise/fall times, feedback of output to modify input and most of all the time lag between the source and the output sine wave (up to 20 degs offset). In other words, a massive headache. I believe pwm would be needed if we go transformerless, lets say, H5 or HERIC.
You lost me with your 4 DC relays. How do you make it work? I tried relays at low freqs to emulate HERIC on iCircuit and it looked very good.

Hi Selimy and tcmtech

I use 4 solid state relays 40 A, 5 to 220Vdc, command 3 to 32 Vdc, and I want conect them direct on the 220 Vac grid... without transformers...
gti str fr.JPG
gti esquema str.JPG
gti str r.JPG

Latest threads

EE World Online Articles