# Theoretical: Connecting square wave inverter to grid

#### MCH170

Joined May 13, 2021
22
This is a purely hypothetical question. Lets say I have a solar panel and I make an "inverter" that produces a 340V (680Vpp) square wave and can perfectly sync with the grid. Since a 240V grid has a peak voltage of 339V my inverter's output is always higher than the grid's voltage, so if I was to connect my inverter to the grid, power would flow from the solar panel to the grid offsetting some of my home's power usage.

Questions:
Would that actually work and result in a lower energy bill?
Would the fact that my power output is minimal compared the size of the power grid be enough to stop my inverter's bad output from causing damage to any connected devices?

Obviously, no electricity company would ever allow you to connect such a device to the power grid and connecting anything without permission is illegal, I know.

#### BobTPH

Joined Jun 5, 2013
9,255
My guess is it would blow up your inverter.

#### Pyrex

Joined Feb 16, 2022
296
Hi,
it's not a good idea. It will be a huge current pulses from the inveter .
Let's say, the grid sine wave at certain moment is 100V , and the inverter produces 340V. So, the difference is 240V. A typical grid impedance is about 1 Ohm. So, 240 Amperes will flow.

It's much better to add a filter to convert square wave to sine wave. Series LC tank tuned to 50 Hz can do that, or another type of filter

#### Ian0

Joined Aug 7, 2020
10,236
Hi,
it's not a good idea. It will be a huge current pulses from the inveter .
Let's say, the grid sine wave at certain moment is 100V , and the inverter produces 340V. So, the difference is 240V. A typical grid impedance is about 1 Ohm. So, 240 Amperes will flow.

It's much better to add a filter to convert square wave to sine wave. Series LC tank tuned to 50 Hz can do that, or another type of filter
I agree it's not a good idea, but it won't be quite such a disaster because huge current pulses can't flow - they are limited by how much sunshine falls on the solar panel.

#### MCH170

Joined May 13, 2021
22
I agree it's not a good idea, but it won't be quite such a disaster because huge current pulses can't flow - they are limited by how much sunshine falls on the solar panel.
That was my assumption as well.

#### MCH170

Joined May 13, 2021
22
My guess is it would blow up your inverter.
Why though? Current never flows into it, only out.

#### BobTPH

Joined Jun 5, 2013
9,255
Why though? Current never flows into it, only out.
What happens when the sun is blocked and the open circuit voltage falls to 50V while the grid is at 370V?

#### nsaspook

Joined Aug 27, 2009
13,538
Why though? Current never flows into it, only out.
Until the magic smoke starts, then, all bets are off the table. The truth is, current will flow into (it's AC so current is always changing direction but that's not very important here) it because electrical energy will see your poor square-wave, unsynchronized inverter as a load and then likely trip the breaker before the house burns down from the resulting electrical fire inside the inverter.

#### ronsimpson

Joined Oct 7, 2019
3,200
Trying to slam a square wave into a sign wave is not good. I don't know anything about your inverter but:
The inverter should have a current limit. At some point in time the inverter is trying to push 339V into a grid which is at 100V as it is a sign wave. The inverter should go into current limit and fold back to 100V. The inverter will follow the sign wave because it cannot lift the grid. The inverter will pop out of current limit at the peaks of the grid voltage then go back into current limit for much of the time.

#### ronsimpson

Joined Oct 7, 2019
3,200
The inverters I worked on, that are designed to push into the power line, produce a sine wave of current. The voltage is set by the power line.
I have seen inverters designed to light up a load, they make a sine or square wave (rectangle) that work with most loads. They will not be happy with a power company as a load.
There are many different designs.

#### Pyrex

Joined Feb 16, 2022
296
I agree it's not a good idea, but it won't be quite such a disaster because huge current pulses can't flow - they are limited by how much sunshine falls on the solar panel.
It depends on the power of the inverter. A 100kW inverter can deliver 140 Amperes ,if there's enought sunlight

#### Ian0

Joined Aug 7, 2020
10,236
It depends on the power of the inverter. A 100kW inverter can deliver 140 Amperes ,if there's enought sunlight
If he had 500 square metres of solar panels! (Or some of those Ebay solar panels which have a "peak" rating of ten times the normal output power)

#### Ian0

Joined Aug 7, 2020
10,236
The harmonics current regulations for grid-tie inverters are not that strict, so if it was a square-wave inverter with a simple current limit (even the limitation of a solar panel acting as a current source) it would probably still be legal.
However, don't forget that Power=Volts x Amps x PowerFactor, so if the current is non-sinusoidal the power factor will be poor so the power supplied to the grid would be much lower.

#### crutschow

Joined Mar 14, 2008
34,818
I have seen inverters designed to light up a load, they make a sine or square wave (rectangle) that work with most loads.
If you design a quasi-sinewave inverter with the the dead-time equal to the positive and negative peak times, then the RMS value will be near 1/√2 (.707) of the peak value (example below), so it will work reasonably well for most loads that use the RMS peak value, such as rectifier power supplies, and those that work from the RMS value, such as heaters and motors.
But it still would likely be problematic to connect that to the sinewave line voltage.

#### Ramussons

Joined May 3, 2013
1,417
This is a purely hypothetical question. Lets say I have a solar panel and I make an "inverter" that produces a 340V (680Vpp) square wave and can perfectly sync with the grid. Since a 240V grid has a peak voltage of 339V my inverter's output is always higher than the grid's voltage, so if I was to connect my inverter to the grid, power would flow from the solar panel to the grid offsetting some of my home's power usage.

Questions:
Would that actually work and result in a lower energy bill?
Would the fact that my power output is minimal compared the size of the power grid be enough to stop my inverter's bad output from causing damage to any connected devices?

Obviously, no electricity company would ever allow you to connect such a device to the power grid and connecting anything without permission is illegal, I know.
You will have an explosion. You are trying to parallel a Sine Wave generator with a Square Wave Generator.
Even if the Frequencies are the same and the Phases synchronised, there will be periods wnen your Invertor is feeding a virtual short.
You will realise this when you plot the Square and Sine wave voltages.

#### Jerry-Hat-Trick

Joined Aug 31, 2022
588
Grid Tie Inverters are actually not that complicated and there is a huge amount on information available for free on the web - some semiconductor manufacturers actually publish schematics and I believe offer Evaluation Boards. For a start, maybe take a look at https://solar.smps.us/grid-tie-inverter-schematic.html . The key to turning a square wave into a sine wave is to chop the signal with a varying mark/space ratio determined by the grid voltage. And if you try to draw too much current from the solar panels the voltage output will drop to below that necessary to match the mains voltage. For that, the system needs to incorporate a maximum power point tracker - MPPT - which typically adjusts the power drawn up and down a little bit to home on in the optimum level, which can vary quite quickly as the panels go in and out of being shaded.

Ideally the DC voltage output of the solar panels in series is sufficient to avod the need for a boost converter although small power grid tie inverters for individual panels do incorporate boost converters and this can actually be more efficient overall if a string on series panels has some panels with less than optimal performance.

Typically, if your solar system is supplying more power than you are using, that power goes back into the grid. And if you are using more power than the solar system can provide then the grid will provide the difference.

It's an intereting subject. Years ago I went on a course in Germany run by SMA intended mainly for distributors/resellers. They were hugely successful as Germany had a very good incentive system which is why Germany has (or at least had) the largest solar power network per capita in the world. Probably both SMA and Germany have since been overtaken, I don't actually know.