Solar panels with battery over discharging.

[ericmark]

Theory the battery should not discharge any more once it hits 10% where there is grid power and 3% when grid power is lost, this means I will always have some power to run the central heating if grid power is lost, and should prevent battery damage due to over discharge. This is what I expect to see, the purple line state of charge being flat as it drops to 10%

However most days I see this, where the state of charge drops to zero for a short time.

And often does not return to 10% for quite some time, so the back up which should be there is gone. And already had one battery fail. This was replaced FOC and the firm which installed the system do seem to be trying to resolve the problem. However is this problem just with my system, or is it wide spread, the chart is from the PC, the phone app does show it, but harder to find and see what is going on, so may be people are simply not aware it is going on? One can't hover a mouse pointer over the chart on a phone to show the window seen on PC version giving the data.

Also it should not be charging a battery when also importing from the grid, as the day goes on and the usage increases, and the solar increases the scale will change to accommodate the higher figures, so again harder to see what has happened.

But is this normal and most don't notice what is going on, or have I got a rouge system?

 

[Why Not Indeed]

What do you want it to do? There can't always be enough sun to charge the battery, so at some point you are going to be charging it from the grid [or to put it another way, shifting some of your grid import from peak to off-peak].
From that first graph I guess the battery was charged whilst the sun was shining to avoid exporting to the grid. Then you had a shower at 1300 and some grid import was avoided by draining the battery. The battery carried on charging after that, to about 95%, then when the PV output dropped off, the battery was drawn down. It all seems fairly sensible to me.
Not sure what's going on in the second graph with the SOC. Going from 9% to 0% instantly when the battery metric [which I assume is power in/out of the battery] is at 0 makes no sense. Is it "balancing the cells" or something? [Bit of EV terminology there, maybe completely irrelevant].

 

[ericmark]

The installer also agrees not doing what it should, it should not discharge to under 10% unless the grid fails, and even then should stop at 3%.

What seems odd is that I am the only one with this problem. They seem to think it is a software glitch, but simply can't see this just being with my inverter.

 

[JohnW2]

deleted ... pressed send button too soon!

 

[JohnW2]

Theory the battery should not discharge any more once it hits 10% where there is grid power and 3% when grid power is lost, this means I will always have some power to run the central heating if grid power is lost, and should prevent battery damage due to over discharge. This is what I expect to see, the purple line state of charge being flat as it drops to 10%

Have you any idea how the 'state of charge' (SOC) is determined, estimated or measured, and how accurate/reliable the figures are? Your second graph seems to indicate that the figures are fairly 'course', only being able to change by increments of 1%.

However most days I see this, where the state of charge drops to zero for a short time. .............. And often does not return to 10% for quite some time, so the back up which should be there is gone. And already had one battery fail.

I am very suspicious from those apparent very sudden drops in SOC, from 9% to 4% at about 02:30 and then from 4% to 0% a bit before 04:00. There were no changes in demand at those times and, throughout that period, grid supply exactly mirrored consumption, as one would expect. I therefore strongly suspect that those apparent changes were 'not real' - unless there is something fundamentally wrong with the (replaced) batter which causes it to suffer from very sudden spontaneous falls in SOC.

However, 'the system' presumably did take those apparent changes seriously (as 'true') since, when SOC apparently fell to zero, it seems that it immediately started charging the battery from the grid, and continued charging it (for 20 minutes or so) until SOC rose to 6%, after which SOC remained at 5% until a little after solar power started charging it at around t about 09:00,when, as expected, SOC started to rise (albeit still only 7% at about 10:00.

Is that 'emergency charging' of the battery from the grid (when SOC falls to zero, or near-zero) something you know is meant to happen?

Kind Regards, John

 

[ericmark]

The instructions are very vague, I got a hand over pack, some 57 pages but that includes the EIC etc. So does not tell me that much.

A hunt for the battery LV-L32-1P has given no results.

 

[JohnW2]

The instructions are very vague, I got a hand over pack, some 57 pages but that includes the EIC etc. So does not tell me that much. A hunt for the battery LV-L32-1P has given no results.

Fair enough - but, as I said, I strongly suspect that those apparent sudden and very rapid appreciable drops in 'state of charge' are not 'real' - i.e. they are probably due to some fault/error in whatever it is that is assessing/.measuring that 'state of charge'.

Even ancient, dying batteries do not usually behave like that.

If I were in your position, I think I would probably try to find a way to continuously monitor/record the voltage across the battery terminals, and then compare that with the 'state of charge' graphs you are seeing.

Kind Regards, John

 

[ericmark]

The battery is encased with no option to use volt meter or clamp on, and brand new so when the battery failed it was covered by warranty.

I was hoping to find out if others have the same or not.

 

[JohnW2]

The battery is encased with no option to use volt meter or clamp on,

There surely must be some means of access to something connected to the battery terminals?

Kind Regards, John

 

[hllns]

The instructions are very vague, I got a hand over pack, some 57 pages but that includes the EIC etc. So does not tell me that much.

A hunt for the battery LV-L32-1P has given no results.

One of these?

GREENLINX

www.luxpower.co.uk

CANbus to the inverter then RS485 linking the batteries, the batteries should be doing the monitoring but perhaps the inverter isn't quite "talking the talk" with them.

 

[hllns]

If it is one of those it does sound like those devices also have their own separate monitoring that can connect to WiFi too though, the bumph on the website isn't incredibly detailed about how you set it up but any information would be from the horse's mouth rather than the inverters interpretation of it.

 

[ericmark]

Yes that's the battery, and think possibly you are right. Thanks for address, I will now look further.

Having looked further it says "Step 6：Set the inverter to charge and discharge the battery for 5 minutes" in the set up, so wonder if the battery is doing some sort of set up?

 

[hllns]

Looking at your other screenshots it looks like you have the same vendor for the inverter (Lux?) so you wouldn't expect incompatibility unless it's down to mismatched firmware/software.

Given the software bells and whistles is there definitely not a logging/status output in an advanced part of the web interface somewhere?

They also claim that the batteries can be discharged fully in normal operation so in theory them being discharged shouldn't have too much of an adverse effect on performance.

 

[ericmark]

The saga continues. British Gas arrived to fit my smart meter, seems I need some special smart meter to work with solar panels or EV chargers, I have solar panels and I need a classic setting up first, and he simply walked away without fitting it. Ringing British gas they say he has listed missing equipment for reason for not fitting, and the earliest appointment is the 19th January next year.

He was telling me the job came in one his phone, and it has to come in on his tablet to fit a smart meter when I have solar panels.

Think maybe time to leave British Gas, was on the phone 45 minutes to get a new appointment. What ever the TV adds say clearly they don't want to fit smart meters. So any excess power I am not being paid for. So some one is getting electric for free, to the tune of 187.3 kWh to date fed into the grid and not been paid for.

 

[hllns]

If you have a battery system that's lasting you through to the early hours you may find it better to go with a smart tariff designed around this sort of setup, they have a cheap overnight rate for battery top-up and pay you extra for export during peak hours if the batteries can spare it.