Panel Orientation?

I asked a solar company to install some panels that were supposed to be split between my house and my daughter's to be built house.  The panels were going to go on my roof as her roof was going to be shaded for much of the day.  Now my roof has a 10 degree pitch using a vertical seam metal roof with the seams running  NW to SW.  Initially I asked for the panels to be flat mounted against the roof to reduce the wind on the roof, and that's what my contract says.

But the installers have rack mounted them at an approximate 40 deg angle facing north east.

I think 12 panels (300W) are mine, and 8 panels are the daughter, and we have 14.4 kWh of lead carbon batteries (8 batteries) shared between us, with two Solax hybrid inverters.
My location is about 41.3° S, 174.7° E

Now this my 3rd day with the system active, and it's just after 4 pm, and I expect by 5:30 pm the solar production will drop to neglible levels as the sun will then be behind the panels.  The whole morning was a miss as it was clouded over until midday.

So, my question is,  does one get more sun light with the panels as flat as possible (allowing some inclination for self cleaning), or is the production higher with the panels pointed north thereabouts (in the Southern Hemisphere).  The other consideration is that with a large amount of solar generation the batteries can only be charged at a certain rate and excess is exported to the grid.  So, wouldn't I have been better off with lower peak production spread over the day?

there is a good free calculator here: https://pvwatts.nrel.gov/pvwatts.php

Enter your address and it will find the nearest site with available historical weather data.  Enter different tilt angles (from horizontal) and azimuth angles (0 degrees is north, 180 degrees is south) and it will estimate the yearly production for your system

So, i think your installers are following generally good practice, by having the rack face toward the equator (north for you) and at an angle nearly yout latitude, you will improve the total production of electricity.
1) you say excess electricity is returned to the grid, is it sold back, if so you are maximizing your payback

2) some of us with off-grid systems set our angles to maximize winter electricity knowing that we will still get an increase in electricity in summer (just not as much).  This makes our daily homemade electric supply vary less but reduces overall production.  On grid the winter angle is not an optimal strategy, maximizing total production is better.

3) unless you get a lot of cloudy weather (more than half the days) then the angle of your roof is too shallow for making the most of your panels potential.

4) if you can time your power usage to use 30-40% of your battery by the morning, your panels,at peak can get them recharged right away,  better than draining the batteries in the evening and having them sit at less than 100% charge overnight.  

Thanks for the calculator link.

I tried various degrees of tilt for the azimuth of 45 deg, and found that best was 23 deg giving estimated production of 7,641 kWh/year vs 7,408 kWh/year at a tilt of 41 deg which is my lattitude.

Joseph Michael Anderson wrote:
4) if you can time your power usage to use 30-40% of your battery by the morning, your panels,at peak can get them recharged right away,  better than draining the batteries in the evening and having them sit at less than 100% charge overnight.  

Currently it looks like the batteries for the main house get to less than 10% at about midnight

http://www.solax-portal.com/dz/home/battery_overview/255502

and the rear flat hits 1% at about the same time which is sad considering the flat is empty since no one can live in it while there's no code of compliance, and the only thing running is a half size fridge

http://www.solax-portal.com/dz/home/battery_overview/255503

I'm very new at this so I'd appreciate a strategy of maximizing the use of the panels, and batteries, and minimizing the export back to the grid.  Normally we pay 29c/kWh imported, and get back 8c/kWh exported.  The batteries aren't getting fully charged yet in the morning but we've had some cloudy mornings this week so far.

Graham - NIWA has a great online tool for this very problem:

https://solarview.niwa.co.nz/

You can modify the values for tilt and bearing to see how much influence that changes your theoretical output. The graphical path is based on location and even shows a DEM-derived horizon, which would be pretty useful in Karori.

Thanks Phil.  Is there an explanation anywhere on what all these curves mean?

Those are the paths the sun takes on the given dates.



Along the bottom of the image you'll see these dates and a range from 240 to 0 and back to 120. These numbers are compass bearings in degrees, with north as zero. Along the left side is the vertical angle from the horizon, so you can see that on 21 Dec in your suburb the sun rises about 30 degrees south of due east, then climbs to 72 degrees above the northern horizon at (solar) noon, which looks to be about 1220, or 1320 by daylight savings. The numbers along the paths show the times and the cumulative kWh/m2 that you would get on a cloudless day given the panel orientation specified. For this one, I used the default, which is a 41 degree tilt pointed due north.

So, how would one calculate what the expected output for one's panels from this graph should be?

And today, for the first sunny morning since install, I seem to have a problem.  The 3.6kW string for the front house is reading 212W being produced, and the 2.4kW string is reading 1410W at 9:53 daylight savings, or 8:53 NZST.

I guess I'll have to go onto the roof to see if there's a problem I can see there.  But the sky is virtually cloudless

This might be the issue.  The panels are setup as a row of 8 panels at the front, and then two rows of 6.  I assume that the front 8 panels are for the rear house.  And the front panels are part shading the bottom of the panels in the rows behind them.



This means that the middle and last row don't start producing until 2.5 hours after the front row on account of the minute shading.  There's plenty of space on the roof to install the panels so that they don't shade each other so I'm considering this an installation error.

The racks look like the racking angle can be changed.  Perhaps if the front two rows of panels are lowered to about 20 deg it won't shade the panels behind them.

Hi Graham,   Wouldn't it be better (seeing as you have the roof space) to move the panels further away from each other rather than changing the angle which would affect their performance?

Hi Gerry

Ideally that would be the case, that the panels are at optimal inclination and they not shadow one another.  But I was thinking it would far less work to just drop the angles on two rows rather than have to move the front row forward, and the back row backwards.

But I got onto the roof, and the actual angle the panels are inclined at is 30 degs from the horizontal ( still at North East or a bearing of 45 deg ).  So, I'm not sure how much they can drop the panels to stop the shading issue.
I guess I need to get on the roof at 6:30 am to measure the shadowing then?  I hope not!

I'm a bit disappointed in the batteries.  Or maybe the inverter isn't giving me the correct information.  The inverter does say that the capacity shown is for lithium batteries but these are lead carbon.
It's 8 pm and already the batteries are apparently at 40% or less yet all I'm running is the fridge and a few LED lights ( and the two inverters ).
They get quite low over night and I'm worried about their life if this is what's happening.

Can someone please decipher the markings on the battery above.  I have 8 of these, and the system is running at 48V I think based on what the inverter is saying.

This might be the datasheet on the battery https://solarking.co.nz/SolarKing_PDFS/Narada_12V_200AH_Battery_Specs.pdf which I found on another website.

The interval you really care about (if your panels are aimed at true north) is 10 am to 2 pm sun time. With yours aimed NE at 40 degrees, this shifts your peak to 10-12 and shows a pretty sharp dropoff after that. The NIWA figures for that bearing and tilt say you'll max out at 646 W/m2 around 11 am (10 NZDT) in Feb and your cumulative power per m2 would be 4.79 kWh. Putting the panels flat on the roof will help...this puts your peak back at solar noon and you'll pick up a little more in the afternoon for 5.14 kWh cumulative.

While there might be 12hrs of sunlight 80% of the solar energy is contained in just 4hr.
2hrs before noon and 2hrs after noon. So a bit of shading at 8am or at 4pm, will not amount to a big difference.
Due to the fact that incoming source energy from the sky/sun is not evenly distributed over the '12hrs', the solar panels can't change that.

When it comes to shading/placement.
It is also possible that structural weight bearing consideration were also factored in the placement which lead to the shading.



Energy Production = Solar Panel * Solar Hours = 20*300W * 4hrs = 6kW * 4hrs = 24kWH
Observed Total = Main + Rear = 20.6kWH + 3.1kWH = 23.7kWH
There isn't much of a difference between projected and observed production, based on the SOLAX link provided.


Each Battery is 6v and 300AH = 6v*300AH=1800WH or 1.8kWH
You have 8 of them for a total of 14.4kWH = 8*1.8kWH
Lead acid batteries aren't recommended to go below 50%, so program the system to import energy once it gets to 50%.
The only time it should go below 50% is if the grid is down. Daily discharge to 10% in 6hrs will kill your battery in 2yrs.

The avg solar house uses about 0.6kW per hour.
6am to 6pm = 12hr* 0.6KW = 7.2kWH
6pm to 12midnight = 6hr*0.9= 5.4kWH
12am to 6am = 6hr*0.3kW=1.8kWH
Total = 24Hr*0.6kW=14.4kWH

If you are following avg usage pattern vs 3x, I don't see why you should be using up 90% of your battery (12kWH) in just 6hrs.
It is possible that due to the Lithium battery setting you always have 50% charge in reserve.
Li battery have a constant voltage where as Pb have a variable voltage, so based on voltage a Pb battery at 51%-59% charge would look like a Lithium battery at 1%-9% charge. This seems like a very chunky way for a professional installer to manage the battery life and pull from the grid but it is possible.

Prior to this install how many kwh were you using per month or day?
Have you seen the battery go up to 100% charge, what is the highest that you have seen?
I wonder if it is possible to force the battery to 100%, then disconnect it from the grid and force a discharge at 5kW(4 iron/toaster/etc at 1.25kW each) for 3hrs to use up 15kWH. Proving said amount is actually available for a pull-down.

Once you get your 1st bill, I would check it to make sure that you are producing on avg 24kWH/day aka 360kWH/2week or 720kWH/month.

Thanks guys.  So, I don't need to be too concerned that 12 of my panels don't start producing until 9 am in late summer though I guess that's only going to get later as winter approaches.

Currently the batteries for the rear house are reading 0% with a voltage of 46.99V, and the main house is 3% at voltage of 47.21V.  So, that doesn't tell us what the current state of charge really is.

I don't really know how these batteries are setup.   I guess the two hybrid inverters are both connected to the same battery bank in order to read the 48V rather than the batteries are split into two sets.

Before the solar system was activated, my electricity bills show that I was consuming on average 26.8 kWh per day over the previous 6 weeks.

Total energy usage = 26.8kWH over 24hrs for an average of 1.1kW per hour
Daylight 6am to 6pm = 1.1kW per hour for 12hr = 13.4kWH
Night-time 6pm to 6am = to the same 13.4kWH
So I am not surprised that your batteries that hold 14.4kWH are almost empty. I recommend doubling your battery bank if possible. or limiting them to only a 50% 7.2kWH discharge.

From 6pm to midnight you seem to be using 12kWHr in just 6h aka an average of 2kWH per hour or 2x the daytime average. This 2x is what most homes see for those time period comparison.

S Bengi wrote:
From 6pm to midnight you seem to be using 12kWHr in just 6h aka an average of 2kWH per hour or 2x the daytime average. This 2x is what most homes see for those time period comparison.

There's a repeating pattern of power use through the night so I'm suspecting the hot water cylinder is using up all this power.  I do have solar hot water on the roof pointing north west, so what I'm going to try, based on the above, is to switch the top up power off to the HWC until the batteries are fully charged in the early afternoon, and then switch the HWC off again in the evening.  It'll mean no heating of the HWC again until the next afternoon, and cold morning showers.  But there's only one person who has a morning shower here, and she's moving to the rear house where we have installed an instant gas hot water system for her.  Hopefully that's going to spare the batteries.

The only things we are running are the TV, internet routers and modems, and a few air pumps for the gold fish so they can't be the culprits.  We do have an electric oven but I don't use that since I built my batch rocket stove oven.

Yes I do see the 3kWH spikes every hour like clockwork.
That said your battery bank is only half of your daily usage. When it should be 3x-5x for off-grid and 1x for grid-assisted. So the only real solution is to increase ypur battery bank size or just let run out it's 500cycles in 500days (aka less than 2years), In two years you might have some money saved up to invest in a more flushed out system.

LED TV, phones, LED light bulbs, All-in-One PC and tablets should not use alot of power (less than 2kwh over an entire 24hr period).
So check into microwaves/toaster/iron/washer/etc. Basically how could you cut your daily energy usage.

I notice that from 12am to 5am. You only use about 0.2kWH per hour on avg.

It is currently 11am and your battery is already at 96% snd 90%.
You have gotten around 25% of your daily solar energy aka 6kW* 1hour for a total of 6kWH. And your battery bank was in the single digit and it holds a total of 14.4kWH. 6kWH of solar should not have brought it up to 96% instead it should have been 6kWH/14.4kWH or 42%.
It's currently 6pm EST for me in USA.

S Bengi wrote:It is currently 11am and your battery is already at 96% snd 90%.
You have gotten around 25% of your daily solar energy aka 6kW* 1hour for a total of 6kWH. And your battery bank was in the single digit and it holds a total of 14.4kWH. 6kWH of solar should not have brought it up to 96% instead it should have been 6kWH/14.4kWH or 42%.
It's currently 6pm EST for me in USA.

So, does that mean it charged from the grid?

S Bengi wrote:

I notice that from 12am to 5am. You only use about 0.2kWH per hour on avg.

So, the hot water is not the problem.
And I have a 2kWh load per hour when we're home in the evenings even when we're not using toaster, oven, washers.  It must be the lights and TVs.  I'll have to check tonight what they're drawing by turning them off and see how the house load changes.

Try not opening/using the fridge for 24hrs, that might be where all the power is going. Only use the fridge in the rear house.

It's possible that the battery is only going from 7.2kWH to 14.4kWH and is reporting that 7.2kWH difference as 1% to 100% due to the lithium battery setting on the inverter.
Being able to get it up to 100% charge then disconnect it from the grid. Then draining them with some load, to measure and confirm that 14.4 kWH was actually available for a pull-down.

You generated about 18kWH today. The sun has been down for 24minutes but your battery is already at 56%.

The battery was at 100% this afternoon.  So, I turned on the top up to the solar hot water at about 3pm, and it promptly plummeted to 42% over the next 15 minutes which sounds improbable since the hot water cylinder only draws 3kW/hour, yet the fall suggests a consumption of 8.6 kWh.  I promptly turned off the hot water boost.

I'm going to leave the hot water off and see what happens to the battery bank.  But it sounds like I can't use the battery data for anything.

The batteries were in a much better state this morning.  With the hot water cylinder just on solar since yesterday, the batteries were at "20%" at 6 am vs being at 1% the day before.
The evening load appeared to be about 600W with my running my usual stuff ( heated toilet seat turned off  )
And the water was still hot enough for a shower this morning.
But I checked the temperatures just now, and it's only 40 deg C ( 104F ) at the top of the cylinder, and if it doesn't get to 60C by early afternoon, I'll have to turn the power on so I don't breed Legionella.

The water should be 140F to kill a bacteria that will infect your lungs, when you take a shower it aerosol the water+bad microbe, that then goes into your lungs. If you are only taking a bath or rag-wash, very little of the microbe+water is aerosol and so would be fine. If the water is kept at 70F (ish) there is very little of the 'bad microbe' and so a 'cold' shower is fine even with the water aerosol.

At around 9AM, turn the hot water heater to the max temp (185F if that is possible) then will use electricity and most likely drain the battery.
After we have super heated the hotwater tank we will then turn off the hot water tank and use that stored heat to last until tomorrow.

We have another 8hrs of sunlight, the battery will be back to 100% in no time. and with the hot water tank not pulling electricity we can see how the battery behaves overnight without the hot water load.
 

Sorry, saw this too late but I don't think I can alter the temperature on the tank easily.

I am now wondering if my solar hot water system is actually working.  It's 23 deg C outside apparently, and the sun is directly facing the evacuated tubes, but the collector temperature is only 32.6 C, with the top of the tank at 41.8C, and the tank bottom 29.9C.

S Bengi wrote:Yes I do see the 3kWH spikes every hour like clockwork.

It looks like there was a problem with the Resol Datasol Bx controller.  It should have been set so that it topped the hot water up only between the hours of 6:00 - 22:00 NZST but as you saw it was spiking all night.  Then from the afternoon of the 18th Feb it stopped doing this.  And from then I've only had luke warm showers ( 45C ) :(   Maybe my playing around with the mains power to the cylinder upset something.  Not sure.

I've read the manual and reconfigured the cylinder controller today. So, I've enabled the top up electric to the cylinder only from 16:00 - 18:00.  But if at 17:00 the top of the HWC is not at 60C, it will boost the hot water to 60C for 15 mins which hopefully means I'll start getting hot showers free of Legionella at night again. My reasoning is that at 16:00 the bulk of the solar energy will have been delivered to the evacuated tubes so if the water isn't hot enough, then I'll start to use the power from the panels/battery to heat it up.  And there might be enough sun light after this to recharge the batteries. I'll have to play with the times to see how it works out.

I've shown the pictures of the panel installation to the vendor.  They're in a different city, and they contracted local electricians to do the install.  They're going to ask the installers to move the panels further apart, or adjust the angles so that they don't shade each other.

They're also going to tell me how to read the state of discharge of the batteries since I can't use the Solax information which relates to Lithium batteries.

You use 27kW per day on average based on data since jan to mid feb. And so far you have only produced 14kwh/day to 24kwh/day. So you are pretty much going to have to import 3kwh to 13kwh per day.

What I think would be less stressful and more useful is just a monthly figure of how much:
Produced = 600kWH
Used = 800kWH
Imported = 200kWH (800-600)

I am glad that the hotwater tank is not using electricity during the daytime to heat it up when solar is still available.  

I found some historical data and in the year July 2017 - 2018 I used 17,000 kWh!  So, overall average was then 44 kWh/day.  

Now that I have a way to track my usage hopefully I can find ways to mitigate this.  But since I do have two 5kW inverters I can still add more panels when I have the cash.

My inverters were only active from mid-Feb so I only have data from then.  Pity as Dec/Jan were the hottest on record, and half the panels were installed in early December but the installers couldn't find the time to come back and finish it until late January.

The battery bank is 33% (14kWH/44kWH) of your daily usage so they will go down to 1% everyday.

If the installers are coming back to adjust the panel spacing, ask them to tilt them less while they're at it and you'll pick up another 300-400 Wh per day.

Yeah, so what I need to do is find a way to stop the batteries draining to 1% otherwise this is going to be a very expensive exercise.  Not sure at what depth of discharge I should tell the inverter to stop using the batteries.  And perhaps there's a way to tell the system to import from the grid to charge the batteries.

I haven't heard yet what the installers are going to do.  Shifting the panels looks like a lot of work so they may just choose to adjust the tilt.  I guess it's up for negotiation.

With your controller configured for lithium vs for lead acid battery, your batteries never truly actually go down to 1%, they are probably at 50% it's just that the 50% is not ex-tractable. This is actually a good thing because it makes your battery last longer 5yrs vs just 2yrs.

The manual for the inverter will have all you need too determine if it will import at designated times or on certain setpoints.

Same for the ac battery charger.

Same for priority on grid use.

And for self consumption within the home.

A tankless water heater would allow sterilization much less often or even for the feature to be disabled.
I would avoid electric heating of water at night on batteries, unless they are designed for it.

I am automating an electric water heater now for a client. It is a motorized breaker that disconnects power when battery soc is low or allows it when low but pv power is sufficient. This is only to be used during a long term grid out.

FRANK:
It seems like he has a hot water tank with a temperature differential from top (70C) vs bottom (40C). That is 'heated' by solar thermal with a 'backup' electric heater element. It's possible it is a 50gallon tank. I do agree that with the manual and some time he will be able to save/record all the current settings, And then play around with stuff (worse case he can revert back to a last know working configuration or reset to factory default or call the company to remotely configure stuff.

GRAHAM:
You said that the rear house will soon be occupied and it will have it's own hot water system. just a FYI, the rear house pulls it's energy supply from the same battery bank. So it will not lessen the draw in fact it will increase it a bit, unless the rear house is fossilfuel/biomass (not electric). The good news is that she can have her tiny hot water tank for the morning and you don't have to try to heatup the entire big tank in the front house for a 6am shower instead you can wait until the sun recharge the front house water tank by the afternoon.

Frank, I have the hybrid inverter manual.  I thought it might be outside my technical limits to read but I guess I'll have to have a go.  But this was supposed to be all setup by the installers.

S Bengi, the rear house has an instant gas hot water system installed with LPG gas tanks.  We thought about an electric hot water cylinder and decided against it to reduce our dependence on the one fuel source.

I've had a look at the manual, and the inverter.  It seems you can select either Lithium or Lead Acid for battery type but that setting is hidden to me.  Would Lead acid apply to lead carbon?  I would guess so but I can't see what the current setting is.

Also there appear to be other settings you can control on export/import/charging from grid but none of these options are available without the installer password.

Here is an oddball. Time of use vs array orientation off grid.

I would bring that to their attention. Looks like they need to visit and reprogram or do it from offsite if it has internet comms. Lead carbon is no different than lead acid, it is a lead acid battery, agm, gel or otherwise. That will do wacky things to your system function.

Good on you for reading and understanding the operation manual!

It is common to be locked out of a leased system or one with batteries under warranty by the manufacturer. Id go through the datalogs and make sure the battery has not seen a damaging level of charge or discharge. Lithium has way different parameters, like C/1 charging! Ouch.