Fake 300VCD supply to test powering for faulty solar string inverter

Thread Starter

Analog Trust

Joined May 3, 2018
5
Hello everybody,
I'm Fabio and I'm writing from the south of IT about, as said in the title, powering a solar string inverter with a 300VDC supply.
Somebody asked to try a fix for this one (since I've yet found a solution to repair some similar inverters).
They told me that there's no response when the apparatus has fed with its panel string, but I'm not sure of this. I don't have the original solar panels to test it safely.
I've checked the majority of things could be replaced but everything seems to be OK, so I'm expecting a faulty board and/or EEPROM.
I don't have a bench test supply and the inverter powers on with max 1000V @ 12A so I'm wondering the best way to turn it on and receive a response from serial port is:
- commonly rectify the main AC lab supply with a diode bridge 1N4007 and capacitor (150-220mF 400V) to get out 310VDC (checking on scope the correct voltage and supply nature before connecting the load)
- a 1/4W 820k resistor paralleled with the capacitor will ensure its discharge when the circuit will be disconnected from main power source.
- the rectifier assembly must be located in a well isolated box, no live wires cause the cap output comes directly into the original inverter input connectors - according to the manual, these connectors seems to be enough to limit input current.
- I can provide an automatic magnetothermal switch rated 400V 8A before the rectifier bridge and an isolator switch w/fuse between the cap and the inverter to provide "current limiting" (it wouldn't exceed 10A).
- for sure I will not connect the output grid back into the mains until I'm not sure that the apparatus could provide enough isolation. But seems to be unnecessary because the inverter will give me a grid fault error.
Knowing how much could be unsafe to run rectification on a 250VAC source and supposing to pay so much attention when going to apply it, I'm asking you for further suggestions and consideration about, and specifically: current limiting tips on the circuit, the possibility to get output grid re-connection and any kind you're thinking I've not yet took into account :)
Thanks in advance
 
Last edited:

Hymie

Joined Mar 30, 2018
713
Hello everybody,
I'm Fabio and I'm writing from the south of IT about, as said in the title, powering a solar string inverter with a 300VDC supply.
Somebody asked to try a fix for this one (since I've yet found a solution to repair some similar inverters).
They told me that there's no response when the apparatus has fed with its panel string, but I'm not sure of this. I don't have the original solar panels to test it safely.
I've checked the majority of things could be replaced but everything seems to be OK, so I'm expecting a faulty board and/or EEPROM.
I don't have a bench test supply and the inverter powers on with max 1000V @ 12A so I'm wondering the best way to turn it on and receive a response from serial port is:
- commonly rectify the main AC lab supply with a diode bridge 1N4007 and capacitor (150-220mF 400V) to get out 310VDC (checking on scope the correct voltage and supply nature before connecting the load)
- a 1/4W 820k resistor paralleled with the capacitor will ensure its discharge when the circuit will be disconnected from main power source.
- the rectifier assembly must be located in a well isolated box, no live wires cause the cap output comes directly into the original inverter input connectors - according to the manual, these connectors seems to be enough to limit input current.
- I can provide an automatic magnetothermal switch rated 400V 8A before the rectifier bridge and an isolator switch w/fuse between the cap and the inverter to provide "current limiting" (it wouldn't exceed 10A).
- for sure I will not connect the output grid back into the mains until I'm not sure that the apparatus could provide enough isolation. But seems to be unnecessary because the inverter will give me a grid fault error.
Knowing how much could be unsafe to run rectification on a 250VAC source and supposing to pay so much attention when going to apply it, I'm asking you for further suggestions and consideration about, and specifically: current limiting tips on the circuit, the possibility to get output grid re-connection and any kind you're thinking I've not yet took into account :)
Thanks in advance
I’m not an expert on solar panel inverters – but offer the following information/advice that might help.

Solar panel inverters connected to the mains supply (feeding power to the grid), are not simply conventional inverter circuits generating a mains voltage at 50Hz – but include complex control circuits to ensure correct operation.

The inverter ac output has to track the mains supply frequency precisely. Imagine a solar panel inverter with an output fixed at exactly 50Hz. If the mains supply of the national grid happened to be 50.1Hz, within a very short time the inverter output and national grid would be 180 degrees out of phase.

On top of that, the precise output voltage and resultant current fed to the national grid has to be controlled.

A further design feature is that the inverter has some form of anti-islanding. In that, I mean that it can detect that it has become isolated from the national grid and is incorrectly assuming that its own output voltage is the grid supply.

Given all the above, I would not be surprised to find that the output of a solar panel inverter will not operate unless its circuitry is at least detecting the presence of the grid supply (with which its output is to be matched).

So even if you had a correctly operating unit – without a connection to the mains supply, it may not function, giving the impression that it is broke.
 

Thread Starter

Analog Trust

Joined May 3, 2018
5
My doubt started when those persons brought me the inverter said that pluggin the input string gave absolutely no response (display not lighting up). They came from a faraway place, so no way to test it with the original equipment to be sure of this. Powering it to check both display response and eventually serial port debugging will be enough for me, helping me to understand further steps. I don't care if it will give me a fault for lacking grid connection or grid impedance too high.
Thank you Hymie
 

Hymie

Joined Mar 30, 2018
713
Having pointed out possible problems with what you are proposing to do – I would expect that if the dc input to the inverter is within the specification for the device, that it would show some indication by at least powering up.

I’m willing to bet someone has done what you are proposing on a working system – thereby using the national grid to power their solar power dc inverter; buying the energy at £0.15/kWh and selling at a feed-in tariff at £0.45/kWh.

So using rectified mains as a dc source should fool the inverter into thinking it is being supplied by the solar panels.

My only other comment would be that with your proposed capacitor of 220uF would result in a peak to peak ripple voltage of 45V (at 1A), which is quite large on 310Vdc. So you might need to increase the capacitor value to reduce the ripple voltage – but by using a scope you will see how much ripple voltage you have on the system (and some indication of the current being drawn).
 

DVM

Joined May 5, 2018
3
Hello everybody,
I'm Fabio and I'm writing from the south of IT about, as said in the title, powering a solar string inverter with a 300VDC supply.
Somebody asked to try a fix for this one (since I've yet found a solution to repair some similar inverters).
They told me that there's no response when the apparatus has fed with its panel string, but I'm not sure of this. I don't have the original solar panels to test it safely.
I've checked the majority of things could be replaced but everything seems to be OK, so I'm expecting a faulty board and/or EEPROM.
I don't have a bench test supply and the inverter powers on with max 1000V @ 12A so I'm wondering the best way to turn it on and receive a response from serial port is:
- commonly rectify the main AC lab supply with a diode bridge 1N4007 and capacitor (150-220mF 400V) to get out 310VDC (checking on scope the correct voltage and supply nature before connecting the load)
- a 1/4W 820k resistor paralleled with the capacitor will ensure its discharge when the circuit will be disconnected from main power source.
- the rectifier assembly must be located in a well isolated box, no live wires cause the cap output comes directly into the original inverter input connectors - according to the manual, these connectors seems to be enough to limit input current.
- I can provide an automatic magnetothermal switch rated 400V 8A before the rectifier bridge and an isolator switch w/fuse between the cap and the inverter to provide "current limiting" (it wouldn't exceed 10A).
- for sure I will not connect the output grid back into the mains until I'm not sure that the apparatus could provide enough isolation. But seems to be unnecessary because the inverter will give me a grid fault error.
Knowing how much could be unsafe to run rectification on a 250VAC source and supposing to pay so much attention when going to apply it, I'm asking you for further suggestions and consideration about, and specifically: current limiting tips on the circuit, the possibility to get output grid re-connection and any kind you're thinking I've not yet took into account :)
Thanks in advance
 

DVM

Joined May 5, 2018
3
HI FABIO
You could rectify the grid voltage with a bridge rectifier as you proposed.
The value will be your grid RMS value times 1.41.
Your problem will be the current of your supply.
Your inverter needs to be connected to the grid to deliver power and ask this power from
the panels.
Without doing this a fault will appear.
Solar panels have a max current supply ( mostly around 8- 9 Amps for 120 Watts panels)
Your power supply has no current limiter -
And to work safe you will need a safety transformer between you and the panels, somewhere
in the circuit.
The current limiter has to be provided, once you need to test a little bit longer.
 

Thread Starter

Analog Trust

Joined May 3, 2018
5
Thank you Hymie and DVM. As solution I'm then going to put a 1:1 trasformer before the diode rectifier, then going to improve capacitor leveling up to the ripple voltage I need.
 

Thread Starter

Analog Trust

Joined May 3, 2018
5
Great suggestion about the power supply, DECELL, thank you.
Phase rotations in the line insulation can be hazardous and this is the reason which I want to power the unit with disconnected grid.
But reading the article I can assume that PLL sync must be yet engineered internally to the inverter?
You're right, I didn't said we are speaking about a Sunny Boy 3000 equipped with transformer and "Sunny Island mode" (GRID OFF).
As I said, more than other parts I've checked its transfo, inductor and grid line filter and everything seems to be OK.
Now I need for debugging.
 

Thread Starter

Analog Trust

Joined May 3, 2018
5
Today, a probably faulty relay switch (slightly conductive) was found on the main board of the unit.
I'm still needing to check a powering of the unit before proceeding with replacement
 
Top