My 4.5kWh setup

reklamos · 06-08-2019 11:30pm #1

All,

I decided to create this thread and share my experience with my PV setup. I hope it will be useful to others.
It has been exactly one month since everything was installed. The install took whole day(crew of three). The longest time it took to do all the electrical part running the cables from fuse box to inverter. I cannot complain about the work performed so far. BER assessment was also done. The only thing that is outstanding is the SEAI grant.
Here are the details of the setup and some stats for one month:
Solis Hybrid inverter 5kWh
12x375W panels(4.5kWh total), roof facing south so all panels are on one side.
2x3.5kWh batteries
Hot water diverter

Total Stats:
Produced: 511.0kWh
Consumed: 579.0kWh
Fed to Grid: 14.0kWh
Bought: 86.0kWh
Batt Charge: 187.0kWh
Batt discharge: 179.0kWh

Best day: 27.0kWh
Worst day: 7.1kWh
Peak production: 5.5kWh
Self use rate: 61.3%

We have to admit that the summer has been pretty good so far. Most of the days I had battery charged and tank full of hot water. During sunny days the battery is full before 11am. Because there is always someone at home, we try to use as much solar power as possible during the day. This is very important part, you need to adjust to solar productions and have heavy electricity users, when there is plenty of sun. So timed dishwashers, washing machines etc. come in handy unless you can control them remotely.
The battery for me come in handy in the evening and during night time. I do have couple fish tanks and servers with switches running 24x7, so there is constant power demand. If the battery gets full charge during the day and there is enough sun till 5pm, then battery lasts full night and does not drop below 20%, until they start charging again in the morning.

So far I'm liking my setup but not everything is perfect.
1. Although my roof is facing south, it is pretty steep. It does not help that on the west side of the roof there is chimney that casts moving shadow during sunny evenings. I think I do lose some energy because of that but cannot quantify how much. The solution would be to split the panels as the inverter is dual mppt but at the moment all panels are connected to single DC input. Or install optimizers on 4 panels. At the moment I'll leave it like it is and may revisit it later.
2. The inverter does produce a lot of heat during sunny hours and it does not have active cooling. On good days the temperature was climbing to 65 degrees and the heatsink was very hot to touch. It does not help that inverter is in the attic that is facing south, so during sunny days the ambient temperature is warm. To improve cooling I made two shrouds that attach to the top of the heatsink and have two 12V computer fans running. The fans have temperature sensors that are attached to the heatsink of inverter and change the speed based on temperature. Since this was implemented the temperature of inverter never went above 42 degrees, which is good enough for me.
3. Inverter does send stats to the cloud on port 10000. There is a webpage and an app that you can use to see what is going on. It sends stats every 6 minutes so it is not instant and the stats are aggregated. The pages and the app leave a lot to be desired. Although it provides a lot of information the products does not look to be complete or user friendly. Sometimes it stops showing some stats(like production, usage etc). Sometimes the webpages changes to Chinese. I do not really like the provided app or web page, the good thing that on the inverter there is an option to send data to custom server and this is what I'll use. I'm working on Grafana dashboard.
4. I noticed that inverter constantly tried to connect to some Chinese NTP server which was not responding but once I redirected it to known good NTP server it stopped doing that.
5. I'm investigating how the battery is charged/discharged. I do not think that this up and down charge/discharge is good for the life span of the battery. I'd prefer it to have more steady charging cycle. The Advanced settings menu is not locked on inverter so I'm pocking around to see if there is anything can be done about it. If anyone has recommendation, please let me know.

unkel · 07-08-2019 12:56am

Nice system, seems to be matched very well to your needs. Well wear!

I wouldn't worry too much about your points 1. and 2. It is what it is and making changes will cost money. Your parts have a warranty. On points 3. and 4. you've already sorted them. Point 5. I would contact your installer and ask them what exact settings did they use. The state of charge of the battery that the system shows is not the real state of charge anyway. There is protection built in at the minimum and maximum levels. If you already know the settings that were programmed, will you share them here? Also what batteries did you get? Pylontech US3000 by any chance?

You have left out the most pertinent detail though. How much did your install set you back in total?

reklamos · 07-08-2019 8:04pm

Total was 12,979 - the grant
Yes, the batteries are 2x Pylontech US3000
The inverter has a profile for US2000 so this is what is used for battery settings. I can get the settings if needed from inverter.

unkel · 08-08-2019 11:15pm

That was not a bad price. The benefit of having two of them (and the larger US3000 Pylontech batteries at that) is not just in more storage, it is also in max charge and discharge rate. It means your inverter can charge and discharge the batteries at 37A times 2 as you have the batteries in parallel. That means your inverter can discharge at 3.7kW. So if it is dark and your panels produce nothing but you have juice in the battery and someone switches the 3kW kettle on, it will still all come out of your battery

Some people with a cheaper setup with just a single Pylontech US2000 don't realise that it can only provide 25A, or 1.3kW. So on a sunny day with a 4kwp setup most of your production will go to the grid (or to the diverter). And if someone puts a 3kW kettle on in the evening, almost all of that will have to be bought from the grid

[Deleted User] · 09-08-2019 10:10pm

What is the measured system round trip efficiency?

Energy from solar through inverter, into battery charger, into battery, outtov battery, into inverter to end load?

or System Energy In versus Energy Out?

reklamos · 10-08-2019 2:21pm

Sir Liamalot wrote: »

What is the measured system round trip efficiency?

Energy from solar through inverter, into battery charger, into battery, outtov battery, into inverter to end load?

or System Energy In versus Energy Out?

The round trip efficiency of the battery is 94% but there are other components involved in the chain so that would drop more but I cannot measure by how much. My guesstimate is 80-90%

reklamos · 10-08-2019 2:30pm

Looks like today is going to be the worst day so far in terms of solar power generation and tomorrow is not looking good either.
I started using https://www.yr.no to try to predict how much power will be generated and this allow to plan the usage better.

[Deleted User] · 10-08-2019 5:53pm

reklamos wrote: »

The round trip efficiency of the battery is 94%

Indeed I was asking what the measured efficiency from your system is?
Not the contrived datasheet figure. There's a rather large difference usually.
I'm curious for a real world example so I can see what normal operating conditions viability is at.

reklamos · 10-08-2019 6:40pm

Sir Liamalot wrote: »

Indeed I was asking what the measured efficiency from your system is?
Not the contrived datasheet figure. There's a rather large difference usually.
I'm curious for a real world example so I can see what normal operating conditions viability is at.

I agree that we cannot fully trust the specs sheet but it comes pretty close. Again the only figures I have are coming from inverter. If I look at total charge(187kWh) and discharge figures(179kWh), the figure comes at ~95%.

The batteries are only over month old and time will tell how they will hold.

[Deleted User] · 11-08-2019 2:28pm

reklamos wrote: »

I agree that we cannot fully trust the specs sheet but it comes pretty close.

I measure them on my systems and find a gulf of difference between Standard Test Conditions (STC) and Normal Operating Conditions (NOC).

STC are climate controlled and optimised load.

reklamos wrote: »

Again the only figures I have are coming from inverter.

Unlikely this is displaying PV DC input so. Probably Inverter Output Only.
The only way to know is to put a clamp-on DC meter on the PV incoming or hardwire an interrupting shunted DC meter.
I've not seen an inverter meter show idle self-consumption at dawn & dusk. Usually, they read 0W when they mean < 0W

reklamos wrote: »

If I look at total charge(187kWh) and discharge figures(179kWh), the figure comes at ~95%.

This is only part of the data.
Charge suggests to me Charger Output so inverter and charger losses not included.
Discharge suggests to me Inverter input.
So if this is how it is 95% is battery efficiency and not system efficiency.

The figure I am most interested in is (Inverter to Load Output / PV DC Input) x 100 [averaged over week]

It's a rather critical coefficient when calculating payback.

Did I tell ya the wan about Tesla PoowerWalls having a 150W continuous heater load in the Winter?

Great job with adding active cooling. You will get an exponential increase in power compared to the self-consumption of the cooling system.

reklamos wrote: »

The batteries are only over month old and time will tell how they will hold.

I would say a very long time. Batteries are quite resilient. It is usually the Charger and BMS that is the Achilles heel of systems.

unkel · 11-08-2019 7:10pm

Sir Liamalot wrote: »

Did I tell ya the wan about Tesla PoowerWalls having a 150W continuous heater load in the Winter?

Jaysus. I presume only when it is installed outdoors and only when it is cold (sub zero temps) That's 3.6kWh lost on a very cold day. That's more than even a large solar array would produce on most winter days.

How did you find this out?

[Deleted User] · 12-08-2019 5:25am

My bad. It's a 300W heater.

Link.

unkel · 12-08-2019 9:09pm

PW sounds buggy from quickly reading a few pages of your link. Also did a quick Google and it appears Tesla do not install the PW in living areas. So not inside your house. Sounds like if you don't have a garage and you live in an area where it gets cold, your PW is going to be very inefficient to the extent of being worse than not having one at times :eek:

reklamos · 07-09-2019 4:27pm

Small update as another month has gone.
I think it was good but I did not hit the levels of July.
The best day produced 25.5kWh and the worst day 4.3kWh

reklamos · 09-09-2019 2:35pm

Big day today, after 2 month and 3 days, I generated my first MHh.

garo · 09-09-2019 10:42pm

Well done! I am 26 days in with a 4.8kW system and 383kWh generated.

rolion · 10-09-2019 8:00am

Very good rewarding figures.

However,"don't shot the mesenger" but the good days are gone...until next March you will have modest returns with more days just generating to keep the cables warm.

reklamos · 18-09-2019 10:07am

My first full electricity bill has arrived since PV was installed.
In 2 months I've imported 187 units, I looked at my bill for previous year same period and it was 987 units. Average daily import is ~3kWh down from ~16kWh
If electricity prices, generation and usage to stay the same, the whole install should payoff in ~10 years. But we all know that prices will go up and generation will fluctuate depending on season/weather. So I estimate it should payoff in ~13-15 years. It may seem a lot of time, but the feeling when I looked at the bill this morning was very pleasant

and I hope more pleasant surprises in those 13-15 years.

There are still things that can be improved in the house that would help me to drop consumption a little bit so stay tuned.

unkel · 18-09-2019 12:52pm

reklamos wrote: »

I estimate it should payoff in ~13-15 years.

If you factor in a 5% or so increase in the electricity price per year and a modest FIT that is likely coming and you got a good deal on the install, then that looks like a realistic figure to me.

At 15 years, that is still a net 7% return on your investment. A lot more than you get with your money on deposit / in any savings account

And of course you are doing the world a big favour being pretty much electricity zero emission with your install

air · 18-09-2019 12:56pm

The percentage comparison vs a deposit account isn't very fair though as there is no capital left at the end of the PVs lifetime.
On the plus side the savings are like tax free returns

unkel · 18-09-2019 1:39pm

air wrote: »

The percentage comparison vs a deposit account isn't very fair though as there is no capital left at the end of the PVs lifetime.

That's true, but the return is pretty decent. Let's say your install lasts 20 years (panels last longer, battery shorter) just for simplicity a €6k investment:

Keep your €6k. On a deposit account you get 2% interest (at best), but 30% DIRT is withheld. So net you get 1.4%. With compound interest in 20 years this will grow your €6k to €7,8k

or lose your €6k into a PV install you get a yearly saving of 7% or €420 per year. Put this saving on deposit and with compound net interest of 1.4% after 20 years this is worth €9.3k

That's 20% more! While you obviously also have greatly contributed to the environment. Worthy investment in my book, unless of course you move house before the 20 years are up...

air · 18-09-2019 1:58pm

Your theory is relying on introducing an additional €420 / year (€8400 total) of capital that has to be earned and deposited to earn the interest.
So I don't think the argument stacks up.

[Deleted User] · 18-09-2019 3:26pm

I'm of the it's a faster ROI if I'm just willing to hang the PV at 15° and give any abundance to my neighbours frame of mind.
So much less hardware expense.

PV is super reliable. Decades of service.

garo · 18-09-2019 3:52pm

air wrote: »

Your theory is relying on introducing an additional €420 / year (€8400 total) of capital that has to be earned and deposited to earn the interest.
So I don't think the argument stacks up.

While I agree the 7% RoI figure is not sound due to the principal not being repaid at the end of the year, this 420€ is indeed real earned savings. You pay 6,000 out of your pocket at the start but then you save 420 every year in electricity bills. That 420 saved can be deposited in a bank and earn interest. So the 9.3k vs 7.8k comparison after 20 years is valid.

garo · 18-09-2019 3:55pm

PS: Finally got the CT sensor connected to the right wire today. It was connected to one out of two wires coming out of the main fuse so was ignoring any power drawn on half the circuits. Washer dryer microwave etc. Nice to see the app now showing real power draw and not some made up ridiculous number. Expect that daily import will also go down from the 5.5 units it was averaging.

rolion · 18-09-2019 5:45pm

reklamos wrote: »

My first full electricity bill has arrived since PV was installed.
In 2 months I've imported 187 units, I looked at my bill for previous year same period and it was 987 units. Average daily import is ~3kWh down from ~16kWh

There are still things that can be improved in the house that would help me to drop consumption a little bit so stay tuned.

As being your first bill with PVs let me "congratulate" you .
May i suggest to wait for 12 months and then you get the figures corect as from todays until end of March the system will not perform as good during summer hot bright days.

air · 18-09-2019 6:16pm

garo wrote: »

That 420 saved can be deposited in a bank and earn interest.

It can't, you're completely misinterpreting the financials.
I can see where you are coming from from a cashflow perspective but from an ROI perspective you're completely wrong.

Take the following simplified scenario:
Electricity supply cost is €400 per year
PV system of some nominal capacity costs €6000 and will cover 100% of consumption. It's usable lifetime is 20 years. Zero residual value.

Option 1. You buy PV to supply your energy
Pay 6000 up front, saving of 400/ year

After 20 years you have received 8000 back in savings (NB. Zero cash is created for reinvestment)
Compound annual rate is 1.45% and that is all. There is no cash generated for reinvestment elsewhere, you've simply eliminated a liability.

Option 2. You keep 6000 in the bank and pay 400/annum for electricity.
Invest your 6000 at 1.45% CAR after tax

After 20 years
8,0000!!!

The difference is you're tied in for 20 years with the PV and any unforeseen maintenance or repairs reduce the return.

[Deleted User] · 18-09-2019 8:48pm

High frequency inverters lasting 20 years?

All my days I have not seen one do that.

Seen a few transformer jobs last that long alright. Installers and manufacturers don't like them though, expensive to ship, expensive to build, heavy to mount to walls every day.

reklamos · 18-09-2019 9:13pm

rolion wrote: »

As being your first bill with PVs let me "congratulate" you .
May i suggest to wait for 12 months and then you get the figures corect as from todays until end of March the system will not perform as good during summer hot bright days.

Oh I know that it is downhill from now on till spring. I think majority who installed PV did it not because of ROI. The ones who think they'll save a fortune, were fooled by sales/marketing. Look what is happening in UK already. We're in wrong country to make any substancial gains from PV power for start.
I'd say probably in 10 years I'll have something added or replaced in the whole setup which will eat up additional cash but I don't mind. I like tinkering with new stuff and if it saves as few $, I'm happy with that.

unkel · 18-09-2019 9:51pm

air wrote: »

It can't, you're completely misinterpreting the financials.
I can see where you are coming from from a cashflow perspective but from an ROI perspective you're completely wrong.

Not at all. You are either trading your 6000 for a cash back of 400 a year. You should put this cash back in the bank where it will compound interest. Or you keep your 6000 in the bank and receive compound interest on top. The first option has a higher return (but also more risks)

Sir Liamalot wrote: »

High frequency inverters lasting 20 years?

It was for simplicity sakes. Panels (including their expensive labour for install) last 30+ years. Battery and inverter maybe a bit over 10 years but can be cheaply replaced without much labour. Overall say a weighted money average of around 20 years

[Deleted User] · 18-09-2019 11:13pm

Depends on the inverter. Ac coupling "hybrids" and DC battery jobbers are north of 1k€. It's a fair chunk of the investment. I'd say 10 years and they won't owe you anything.
For simplicity I count 2 inverters over 20 years (or Studer).

My 4.5kWh setup

Comments