Solar Panel QRM from SolarRiver inverter

[G4AON]

A recently installed PV solar panel inverter is causing QRM on HF through 6m. The installation is approx 100 yards from my house and comprises PV panels on a roof connected by a run of roughly 60 feet of cable to an inverter located in a garage. The inverter is a 4KW SolarRiver 4000TL-D by Samil Power (http://www.samilpower.com).

The interference presents itself as multiple carriers spaced 20 KHz apart across 80m, 10m and 6m. The signal level is -80 dBm on 80m (around S7 to S8), it is less strong on 10m and 6m. The carriers are relatively clean and vary in amplitude over a period of around one second. They are within a few Hz of 3500, 3520, 3540, etc.

I have tried fitting several TDK ferrite clip on chokes to both the mains and PV panel side of the inverter, unfortunately without making much difference. When the owner turned off the PV isolators the interference stopped, it is also not present after sunset.

Has anyone had experience of fixing the interference from one of these units?

73 Dave

[G4AON]

Forgot to add the type of clip on Ferrite core, they are TDK ZCAT3035-1330

73 Dave

[W1VT]

http://www.mouser.com/ds/2/400/e9a15_zcat-13869.pdf
If  you look at page 4 of the data sheet, the example filter has very little effect at 30MHz--it is more effective at higher frequencies.

Typically, to be effective at low frequencies, many turns on a toroid are used--the impedance typically varies as the square of the number of turns.

Zack Lau W1VT

[G4AON]

Thanks for the link Jack

The wiring has no slack to enable me to even form one loop through a ferrite. Virtually all the ferrite "clip on" chokes here seem to be made by TDK and the ones I have are probably the best of a mediocre bunch in terms of being effective in the lower HF region.

I have seen metal encased LC filters offered by some solar panel supply companies, but they appeared to be intended for use inside a cabinet, but ultimately may be the most effective form of filter.

I will e-mail Samil Power to ask if they have any suggestions, at least making them aware of the issue may prompt them to consider HF interference in future designs. I won't hold my breath though.

73 Dave

[WX7G]

Ferrite cores alone can offer up to 30 dB of attenuation. Note that a core must be placed around both conductors, not one. Looking at the datasheet, to obtain 3k of common-mode impedance at 3.5 MHz will require three dozen cores. You might need to filter both the solar panel side and the output side.

If you want to fix this in one try an off-the-shelf EMI filter should be installed at the input and the output. They will need to be installed at the controller and grounded to the controller case/gnd with short wires. If you chose not to use the filters recommended by the manufacturer I can recommend some.

   WX7G, NARTE Certified EMC Engineer