MP II in Parralel - funny noise, do I need to worry?

Q&A and troubleshooting
, ,
1

I looking for experience sharing and expertise about those strange noise you could hear from the following video.

reading the forums, it’s look like this installation is not a isolate case, but I could not find a clear answer about diagnostic and solution.

few more tips: 2 x MpII 5000/70 in parallel , Italian grid, time to time but regularly we got L1 overload but no alarm. those overloads don’t show up on the different graphs where voltage, amp, frequency look within the reasonable limits (small loads). at the same time AC in , the grid as no issue neither.

This installation work fine except those funny noises, do we need to worry or find a remedy ?

(this installation was done with 2 others houses with the same exact configuration, the other 2 never got this symptoms).

2

This grunting sound can be caused by peak loads, even if these peak loads fall within the maximum power of the multi-voltage heater.

An MP has some difficulty with repeated peak loads, but as far as I understand this would not cause any immediate problems.

I once connected an electric paint burner to a MP as load during a test. This paint burner apparently regulates the filament temperature with short on/off pulses, and this caused the MP to make the grunting sound.

If you have a Victron smart shunt, you can see these peaks in the current consumption graphs.

You could investigate whether certain appliances in your home generate short but high peak loads. Perhaps a water heater or an air conditioner/heater?
Turning these off might stop the grunting noise.

3

This noise typically crops up in two situations:

  1. When consumption exceeds the installation’s maximum power due to a high reactive load (all blocks in the installation will be noisy). The remedy is to scale the system up.

  2. When the total MP2 power is higher than the reactive load, but the load distribution among the blocks is unbalanced you get noise from one block out of two, or two out of three, e.t.c. In practice, this stems from incorrect impedance on the AC IN/OUT lines, influenced by cable length and cross-sectional area.

You can see the full picture by using two portable dual-channel oscilloscopes.

UPD: DSP power analyser won’t help you in this case, the amplitude cycle is too very short. Most DSP-based power analysers average over several cycles, so anything shorter than a second or two simply vanishes in their filters.