Solar PV Hints, Tips & Troubleshooting

Got a strange one going on…..

Recently installed a Solis S5-EH1P(3-6)K-L 5kw.

System seems to be performing as expected however, when decently generating, e.g. above 3kw, there seems to be a click or a sudden tick noise that happens every now and then. Edit (noticed it does this when its on battery also, e.g. night time 3kw load.)

It appears to happen more frequently when im generating above 5kw, every 5-10 minutes.

Its coming from inside the invertor. Any ideas what this could be?

The fan is very noticeable, so i dont think its that. If its a relay, its slapping the contacts very hard.

It almost sounds like those lighters that has the electric spark. Actually, it sounds like an electrical ark.

They make a bang/clunk rather than a click.

yup, this is like a loud spark.

AFAIK Dyness are warrantied to 90% dept of discharge but just double check the model of your batteries on their website to be certain.

Is the time clock on car charger, solis inverter and ESB meter the same?

The inverter doesn't appear to support NTP (internet source time) for it's timekeeping, so you'll need to check that time clock reasonably often. That in adding in buffers of a few mins so that the car stops charging a few mins before the change-over of the smart meter to the new tariff, etc.

Is the inverter on ? got batteries?

Just seen some unexplained behaviour with the inverter the last couple of days. This is the solar production graph from yesterday. It seems that production is slow until something kicks in around 10:30 and then again at 13:30.

This seems to tally up the power being generated from the two strings but here again you can see big jumps in production at 10:30 on string 1 and 13:30 on string 2.

There is no shading on the panels which would be causing the sudden jumps in production.

Below is a graph from March where the production ramp up is much smoother without the jumps.

This rapid jump has been a feature over the last week or so but there is nothing to explain why this is the case. I have shutdown and restarted the whole system but am still seeing the same patterns following it.

Can anyone give some pointers as to why this may be happening?

Hmm, from the charts I'd agree with you that it looks like a shading performance profile. But the chart he included from March (which would have the same general sun orientation as now Sept) doesn't seem to have it. So I'm inclinded to think that it's not from that point.

Unless there's been some recent addition which causes shading, perhaps the neighbour has installed a new satelite dish (or something you get the idea)

@viewofthehill what's the size and orientation of both your arrays?

The two arrays are 4.56Kw SSE and 3.8Kw WSW at right angles. The SSE array is on a single storey with a hedge about 10 metres from it. The WSW is on a two storey. The hedge is abou 6m high.

Just had a look back at this time last year. Apologies for the graphs but the app doesn't store data from more than 6 months ago. The chart below is an example from the almost the same day last year (21 September). As you can see on the graph there is the same ramp up as in the March graph. The only thing that has changed in the last year is that the hedge is now probably 2m lower than it was originally. Past the hedge the nearest object that would cast a shadow is a hedge about 60m from that.

For comparison purposes I have the same graph for 18 September this year

Looking at the two you can see it starts producing at a similar time but there seems to be a big jump at 9:30 and again at 12:30 (time on the graphs are out due to system time vs. Summer time). PV1 is the SSE array and PV2 is the WSW array

I had a look this morning and it seems that the voltage is causing the problem

The voltage on both strings appear to rise to 100 and then stay there for a time period before going to 300/350 for the strings.

Voltage graph from last September. It seems the voltage rise was when the panels started producing.

Normally voltage climbs very early on when the panel starts up, it should get up to between the panel's stated Voc and it's Vmpp, then the current starts climbing after that and voltage stays relatively stable. If all of the panels are in full sun when this condition occurs (that being the voltage hovering around 100v DC) then I'd wonder if the panels are going into a bypass state, but normally this would be triggered by by a low-light condition. Are the panels clean of dirt and definitely not in shade when that condition occurs?

Not a clue. I'd forward those images of the voltage and power for both strings onto the installer and see what they make of them. But keep us posted please.