Validate my plan (or tear it apart)

[display_Name]

I'm ready to press go on my install (very excited). Would love some thoughts on the chosen setup, and if you have any recommendations for installers in Cape Town Northern suburbs, please DM me.

Here's what I'm looking to set up:
2 x Solis 6kW Inverters (S6-EH1P6K-L-PRO)
32 x Jinko Tiger 560W (JKM560M-72HL4)
Axe Struct Mounting system (Railed Roof Top)
SolarMD SS 4143 14,3kWh
 

Mock up of the roof attached. Mostly East / West, I've added summer / winter sunrise, midday, sunset lines to show morning and afternoon sun. I can't mount on the north facing roof (geometry won't allow).

My thinking is to have 4 strings of 8 panels per string, with the strings of each MPPT split East and West. So MPPT1 String 1 East and MPPT 1 String 2 West.

Was looking at 5kW Sunsynk inverters initially because the software is apparently pretty good, but the panels I'm looking at are 13.6A (over the Sunsynk rating of 13A), and the Solis 6kW at the same price looks like it makes sense with the additional headroom. Unless I'm missing something.

The equipment above comes to R185K excluding shipping. I'd still need fuses, wire, connectors, DB board, breakers, conduit etc etc... and of course the installation.

Does this look about right for a setup?


 

[Scorp007]

13 hours ago, display_Name said:

I'm ready to press go on my install (very excited). Would love some thoughts on the chosen setup, and if you have any recommendations for installers in Cape Town Northern suburbs, please DM me.

Here's what I'm looking to set up:
2 x Solis 6kW Inverters (S6-EH1P6K-L-PRO)
32 x Jinko Tiger 560W (JKM560M-72HL4)
Axe Struct Mounting system (Railed Roof Top)
SolarMD SS 4143 14,3kWh
 

Mock up of the roof attached. Mostly East / West, I've added summer / winter sunrise, midday, sunset lines to show morning and afternoon sun. I can't mount on the north facing roof (geometry won't allow).

My thinking is to have 4 strings of 8 panels per string, with the strings of each MPPT split East and West. So MPPT1 String 1 East and MPPT 1 String 2 West.

Does this look about right for a setup?


 

 

East and West in parallel on 1 MPPT and the same on MPPT2 vs 2 x East strings on MPPT1 and both West strings on MPPT2. 

Let's hear what the installers say

@TaliaB

@Steve87

Edited April 8 by Scorp007

[TaliaB]

1 hour ago, Scorp007 said:

East and West in parallel on 1 MPPT and the same on MPPT2

Method 1.

 

1 hour ago, Scorp007 said:

2 x East strings on MPPT1 and both West strings on MPPT2

Method 2.

Method 1 and Method 2 will produce the same kwh from the east and west arrays.

Method 1 will have an even ouput on both Mppt's during the entire sunlight day.

Method 2 will see Mppt 1 at higher output during the morning compared to Mppt 2 and vica versa during the afternoon.

@display_Name

[P1000]

2 hours ago, Scorp007 said:

East and West in parallel on 1 MPPT and the same on MPPT2 vs 2 x East strings on MPPT1 and both West strings on MPPT2. 

Let's hear what the installers say

@TaliaB

@Steve87

I think he is actually reffering to MPPT1 on different inverters. So each string will go to a separate MPPT, each inverter has 2 MPPTs = 4 MPPTs. Well, if you have 4 strings and 4 MPPTs, I would wire each string to it's own MPPT.

[Scorp007]

34 minutes ago, P1000 said:

I think he is actually reffering to MPPT1 on different inverters. So each string will go to a separate MPPT, each inverter has 2 MPPTs = 4 MPPTs. Well, if you have 4 strings and 4 MPPTs, I would wire each string to it's own MPPT.

My mistake. But if it was a single 8kW Deye. Then back to my question if one could use 4 inputs to the MPPT. 

Which is the better connection. 

Edited April 8 by Scorp007

[Steve87]

9 hours ago, Scorp007 said:

Which is the better connection. 

same amount of power flowing...just the pattern changes.

I have combined East and West on the same Mppt too many times to remember. From a simple 12kW Deye all the way up to a 100kW Atess. The PV Array output is not affected when you combine. You must just make sure that the strings are equal in number of panels. 

The Solis MPPT has some nice expandability more than the Deye and Sunsynk. However, always install any panels with room for expansion if not already maxed out in terms of the roof space or MPPT amperage etc. 

 

Edited April 8 by Steve87

[display_Name]

Loving the replies, thank you.

My thinking is that each inverter has 2 MPPTs... so put the panels for one of the MPPTs facing East (on one string), and the panels for the other MPPT facing west (on one string). I.e. only ever using 1 string per MPPT. Then do the same for the second inverter.

If I instead split 2 strings East / West from one MPPT, would I not be leaving MPPTs on the table? Or are you thinking smaller strings? 

[TaliaB]

26 minutes ago, display_Name said:

Loving the replies, thank you.

My thinking is that each inverter has 2 MPPTs... so put the panels for one of the MPPTs facing East (on one string), and the panels for the other MPPT facing west (on one string). I.e. only ever using 1 string per MPPT. Then do the same for the second inverter.

If I instead split 2 strings East / West from one MPPT, would I not be leaving MPPTs on the table? Or are you thinking smaller strings? 

Nothing wrong with 8/string if you don't need the mppt's for the future go for it. I won't do smaller strings.

[display_Name]

5 hours ago, TaliaB said:

Nothing wrong with 8/string if you don't need the mppt's for the future go for it. I won't do smaller strings.

Brilliant. Yeah I think I'll go with 8 per string then.

From an inverter perspective, I'm fairly comfortable with the Solis inverters, and have been in touch with Solis, understand the return process if needed, who the distributors are etc. But... I've seen a number of posts on Solis here on the forum... any reason I should not be going Solis?

My thinking is that I want to keep this simple, and just put the whole house on the backup load (I'll put them in parallel). I don't think we'll go much higher than 6-7kW draw, and the SolarMD can max out at 10kW, so I shouldn't be pushing that to max. I can always double up on that if needed down the line. Bad idea?
 

[Bobster.]

6 hours ago, display_Name said:

and just put the whole house on the backup load (I'll put them in parallel). I don't think we'll go much higher than 6-7kW draw, and the SolarMD can max out at 10kW,

Got an electric lawnmower? A while back we had load shedding and I told my gardener I'd run an extension lead from a socket inside the house so that he could carry on mowing. After all, it's only 750W. 

Ha ha! The system promptly tripped. 

This weekend just gone, with no shedding scheduled, I used my trusty kill-a-watt to measure what the mower was doing. Once it was running it drew about 614W. But on start up, momentarily, it could draw up to 4.3kW. Hence the trip. 

Your household is full of treacherous devices.

When I got this system installed I'd hoped that everybody would be able to just get on with their lives like nothing was happening. That hasn't happened and we have had to learn to live within the limits the system imposes. 

This isn't much of a hardship. Unless we want to use the lawnmower. 

[Shubs]

On 2024/04/08 at 12:11 AM, display_Name said:

I'm ready to press go on my install (very excited). Would love some thoughts on the chosen setup, and if you have any recommendations for installers in Cape Town Northern suburbs, please DM me.

Here's what I'm looking to set up:
2 x Solis 6kW Inverters (S6-EH1P6K-L-PRO)
32 x Jinko Tiger 560W (JKM560M-72HL4)
Axe Struct Mounting system (Railed Roof Top)
SolarMD SS 4143 14,3kWh
 

Mock up of the roof attached. Mostly East / West, I've added summer / winter sunrise, midday, sunset lines to show morning and afternoon sun. I can't mount on the north facing roof (geometry won't allow).

My thinking is to have 4 strings of 8 panels per string, with the strings of each MPPT split East and West. So MPPT1 String 1 East and MPPT 1 String 2 West.

Was looking at 5kW Sunsynk inverters initially because the software is apparently pretty good, but the panels I'm looking at are 13.6A (over the Sunsynk rating of 13A), and the Solis 6kW at the same price looks like it makes sense with the additional headroom. Unless I'm missing something.

The equipment above comes to R185K excluding shipping. I'd still need fuses, wire, connectors, DB board, breakers, conduit etc etc... and of course the installation.

Does this look about right for a setup?


 

Can you share your requirement?  That can help with comments on the solution design.

What on earth do you need 32 x 560s for, with only a 14kw battery?  

[GreenFields]

On 2024/04/08 at 4:37 PM, Steve87 said:

I have combined East and West on the same Mppt too many times to remember. From a simple 12kW Deye all the way up to a 100kW Atess. The PV Array output is not affected when you combine. You must just make sure that the strings are equal in number of panels. 

Personal 2c. Open to correction. Is there not a risk though of the panels in parallel exceeding the inverter's input current and causing clipping of output? Just making up numbers here, if it's 13A panels, and at mid-day, they are outputting for argument's sake 9A each, maybe more if the roof pitch is flatter, then this Solis will limit current to 16A.

Bear in mind I have zero experience actually installing these, but I'd be leaning towards 8s panels per MPPT, basically just shy of 4.5kW of panels per MPPT, 9kW per inverter, so no real scope for expanding or keeping MPPT's in reserve for future use. One East, one West string per inverter, is just keeping it simple.

I'd go with more headroom on the batteries, depending on your load requirements. Either a second of the same MD, or else 3 x 1C or 1.5C 5kWh batteries, just to be able to make use of the headroom on the Solis inverters, and also cater better for surges coming on top when you're already drawing 6-7 kW.

Edited April 16 by GreenFields

[display_Name]

 

6 hours ago, Bobster. said:

Got an electric lawnmower? A while back we had load shedding and I told my gardener I'd run an extension lead from a socket inside the house so that he could carry on mowing. After all, it's only 750W. 

Ha ha! The system promptly tripped. 

This weekend just gone, with no shedding scheduled, I used my trusty kill-a-watt to measure what the mower was doing. Once it was running it drew about 614W. But on start up, momentarily, it could draw up to 4.3kW. Hence the trip. 

Your household is full of treacherous devices.

When I got this system installed I'd hoped that everybody would be able to just get on with their lives like nothing was happening. That hasn't happened and we have had to learn to live within the limits the system imposes. 

This isn't much of a hardship. Unless we want to use the lawnmower. 

Oh definitely. I was hoping that by having 12kW on the inverters (limited by 10kW peak from the battery in the beginning), I'd be pretty close to supplying what municipal mains can supply for my house (13kW from what I understand).

5 hours ago, Shubs said:

Can you share your requirement?  That can help with comments on the solution design.

What on earth do you need 32 x 560s for, with only a 14kw battery?  

Great question! Check my logic for failure please:) When I do the math, as weird as it sounds, maxxing out the system panels pays it off way quicker when pushing all the excess to the grid as the panels and mountings are the cheap part. Let me explain my thinking:

Essentially, I want to sit as close to the municipal max (I've been told it's 13kW for my house) that I can. Firstly this is so that I can better absorb the peaks Bobster refers to. With 2 Sunsynk 5kW units, I'm at 10kW. In theory this matches better with the SolarMD battery's max. However I can see myself adding another battery at some point, allowing me to fully utilize the 12kW peak of 2 x Solis 6kW inverters and have better backup for winter periods. Then from a panel perspective, I want to get as close to that as possible, so if that's 4.4kWp on each string facing East / West, then midday it's probably going to clip a little and max out at 6kW (hopefully at least some of the time). In winter, I am hopefully never drawing from the grid, and thereby not paying the more expensive rate, improving my differential.

Secondly, I'm now maxxing out what I can send to the grid (as close to municipal mains threshold). This is where the payback figures flip around. Because I've already committed to 2 x 6kW inverters to handle peak load, adding panels becomes the cheaper part that really drives down the overall payback period when feeding back into the grid. Back of the napkin calculations taking an average production per day (averaged over the year) of 3.8kWh per kWp (apparently cape town is around 4.8, but I'm derating that because I'm not going to be at a perfect orientation), I should reach payback in 3.9 years. Even if that extends to 5 years, that's a solid payback period. This is without the eskom price increases coming.

Second to this comes the warm and fuzzy feeling of contributing a little to solving the load shedding issue in a tiny tiny way by creating more supply.

Anyway, that's my thinking... 

Bottom line, even though the overall system cost is more expensive, it's payback period drops by several years if I do it this way.

Edited April 9 by display_Name

[Shubs]

Impressive and solid plan! Up until this post, my (terrible) memory doesn't recall seeing anyone propose that it's worth it to feedback. Not at this scale

What have you calculated as the meter fees with the new discounted model, and the payback return per kW?

Would be cool to see that napkin

[Scorp007]

3 hours ago, Shubs said:

Impressive and solid plan! Up until this post, my (terrible) memory doesn't recall seeing anyone propose that it's worth it to feedback. Not at this scale

What have you calculated as the meter fees with the new discounted model, and the payback return per kW?

Would be cool to see that napkin

Exporting around 15kWh per day at R1.25 can produce an income of over R6 000 per year. Take into account the saving of over R2 500pm in power cost the return is perhaps higher than most other investments. 

Very rough estimate. So it does not take too long to cover the bi-directional power meter. Who knows if one would not get a better price per unit in years to come at the rate of Eskom increases. Time will tell but worth exploring. 

[GreenFields]

40 minutes ago, Scorp007 said:

Exporting around 15kWh per day at R1.25 can produce an income of over R6 000 per year. Take into account the saving of over R2 500pm in power cost the return is perhaps higher than most other investments. 

Very rough estimate. So it does not take too long to cover the bi-directional power meter. Who knows if one would not get a better price per unit in years to come at the rate of Eskom increases. Time will tell but worth exploring. 

How much is the OP's current daily power consumption in kWh? Just thinking that with around 18W of panels - okay but then the panel orientation and inverter rated power will reduce and clip the actual output - but still, that's anywhere around 50-70kWh on average that can be generated daily as a thumbsuck. How much of that is potentially going to the grid? Would really love to see the calcs.

[Scorp007]

1 hour ago, GreenFields said:

How much is the OP's current daily power consumption in kWh? Just thinking that with around 18W of panels - okay but then the panel orientation and inverter rated power will reduce and clip the actual output - but still, that's anywhere around 50-70kWh on average that can be generated daily as a thumbsuck. How much of that is potentially going to the grid? Would really love to see the calcs.

I just took a thumb suck. The OP estimated at 3.8kWh/kW installed per day. I then used only generating 12kW from panels as in line with the 2 Inverters. This then excludes the actual capacity of the panels and takes winter and summer into account and also conservative. 

It would be interesting to see the actual figures used. 

Edited April 9 by Scorp007

[display_Name]

Just able to get back to this thread now. Really appreciate the comments here. It provides an opportunity to correct any thinking I have. I've tried to shy away from extreme optimism, and I also have some unknowns which will create some variance. I'll list them here as you may have some answers for me.

So I took my average consumption over the last year (some heavy months like July / Aug compensate for some lighter months like Jan / Feb), and on average I'm on about 32 kWh per day (works out to about R96.40 cost per day). In terms of panels, I'd have 560 x 32 = 17.920 kWp. I used the factor of 3.8 x kWp to get to Generation of around 68 kWh per day.

What I don't know... is what the impact of the clipping will have on that value of 3.8 (I've tried to max out the 17.9kWp of panels because if you assume a poor power factor of 3.8... I'm not sure the panels are ever really taking the inverters up to max output of 12kW to grid). We have 14 hrs of sunlight a day in midsummer and 10 hrs a day in winter, so 3.8 is fairly low, and I want to keep it that low to be conservative ideally. In winter the panels are probably not pushing anywhere near 12kW taking cloud and Azimuth into account, and in summer we just have sun for so many more hours in the day.

This leaves me with about 36kWh on average per day spare, which would nett me R37.54 if we assume Eskom (78.98c) + CoCT (25c) = R1.04.

So if I take my average daily difference (R96.4 not being spent + R37.54 being earned) I should get to about R133.94 per day or R4152.14 per month change in my current position.

I don't know what impact clipping will have on my calculation above, and while I tried to drop the avg generation of 4.8kWh per kWp down to 3.8 for safety, maybe that's not enough (shrug emoji?). I need to draw a line somewhere, so that's an assumption that may hold out to be true or false. It would be great to get some stats on your measured kWh per kWp on average over the year and get a sense of what a real world value could be.

Ok, now for cost:
From what I can find online:
Panels:
Jinko 560W panels from inverter-warehouse.co.za: R1,695 per panel. 32 panels = R54,240

Inverters:
Solis S6 Pro 6kW from thepowerstore.co.za: R19,585 per inverter. 2 inverters = R39 170

Battery:
Solar MD 14kWh from solarspot.co.za: R62,500 (I've seen them for R60K, and I think with lithium trending cheaper, and the new SolarMD factory happening this year, I expect this to continue the drop over time.

Mountings:
Axe Struct: They helped me get the part list I need for 32 panels in the layout I need (they have a configuration tool that's pretty good), and I got the quote from ARB for the parts: R28,569.11

This gets me to R184,479.11

Now I need cables, fuses, DB boards, and the actual installation. I've priced most of this, and it's not that much, but the big variable... what will installation cost me?

If I had to thumb suck and say shipping, cables, fuses and installation could cost me R40 000 (I may be high or low here... I just don't know what the remainder will cost me... hence some estimates), then I'm sitting at R224,479.11

If I do the divide (R224,479/R4,152.14), I get 4.5 year payback. This is different to the 3.9 I had above because I removed the Solar Tax incentive that I still had in there (that lapsed in Feb this year), and I added more buffer for shipping - also guess work. I really don't know what to expect for the install cost quotes here, and that's the main variable at play here for me. I mean... I really want to do most of the DC side of the installation myself, including mounting the inverters, battery and panels (excluding AC wiring), but I'm not sure how much I'm allowed to do, and where I have to bring in an installer / electrician, or what that cost would be.

Interesting note: What really struck me, is just how far the equipment costs have dropped in the last year. I have been tracking equipment cost over the last year for my install. In July last year, the SolarMD battery was R85,000, Sunsynk 5kW inverters were R28,000, 555W Jinko solar panels were R2,300. That R184,479 figure that I came to earlier (excluding cables, shipping, install) was R234,000 in July 2023 for a smaller system of 28 panels and 10kW of Inverters.

SolarMD cost drop for the 14.3kWh
In July 2023, the battery was R85 600.47 (arb.co.za)
In August 2023, the battery was down to R76,923 (jcsolarpanels.co.za)
In December 2023 it was R73,363.88 (jccolarpanels.co.za)
At the moment I can find one for R62,500 (solarspot.co.za, with others not far off this)


Anyway, apologies if that's too much detail, but that's my thinking anyhow. I'd appreciate any holes being poked. I'd rather them poked here on digital paper where only my fragile ego can be dented, than in my pocket during the install.

Any insight on my unknown costs would be really appreciated.

 

Edited April 9 by display_Name
Typo

[TaliaB]

9 hours ago, display_Name said:

We have 14 hrs of sunlight a day in midsummer and 10 hrs a day in winter

This hours would be location specific but your hours used in your statement above is a bit optimistic. Use the attached list and do the calculation on peak sun hours rather than light hours.

[display_Name]

2 hours ago, TaliaB said:

This hours would be location specific but your hours used in your statement above is a bit optimistic. Use the attached list and do the calculation on peak sun hours rather than light hours.

No we're aligned. The peak value there of 4.7 to 6 (Western Cape), is what I derated to 3.8, so I am assuming that I will only get an effective 3.8 hours of peak sun... just spread out over the day and not in exactly a 3.8 hours period, meaning I'm probably not being limited much by the inverter's 12kW limit too much.


I assume that this (Mid-Summer):


and this (Mid-Winter):


Taking the angle of the sun, the rainy days and all the other negative stuff would equate to the 4.7 - 6 value... and I'm dropping that to 3.8 because of orientation etc.

Does anyone here perhaps have any actuals they could share of what they're getting in terms of kWh per day per kWp, over the year? I'm all in theory land at the moment.

Edited April 10 by display_Name

[GreenFields]

See the attached document. The end result of some online PV calculator's data drawn in, with some further calculations on top.

Basically it is a forecast for Cape Town for the output for your system size. I've assumed a roof pitch of 30 degrees. Maybe not exact, but it's a decent prognosis of what you could expect to see generated per day, per hour, throughout the whole year, from your system. And I think it's a fair basis to work from to add in your own self-consumption , battery charging and discharging, and projected grid export, and calculate your returns.

Note you the last column of AC+DC power generation is only applicable if you are charging batteries and your power use goes over and above the 12kW to which the AC generation is limited for consumption or export, but I don't know how you'd set this inverter up to do that.

 

 

Doc CT.xlsx

[PierreJ]

2 hours ago, display_Name said:

Does anyone here perhaps have any actuals they could share of what they're getting in terms of kWh per day per kWp, over the year? I'm all in theory land at the moment.

This online estimator uses the weather for previous years at your location in order to estimate yields going forward:

https://pvwatts.nrel.gov/pvwatts.php

It's the best one that I'm aware of, and it's free.

[display_Name]

8 minutes ago, GreenFields said:

See the attached document. The end result of some online PV calculator's data drawn in, with some further calculations on top.

Basically it is a forecast for Cape Town for the output for your system size. I've assumed a roof pitch of 30 degrees. Maybe not exact, but it's a decent prognosis of what you could expect to see generated per day, per hour, throughout the whole year, from your system. And I think it's a fair basis to work from to add in your own self-consumption , battery charging and discharging, and projected grid export, and calculate your returns.

Note you the last column of AC+DC power generation is only applicable if you are charging batteries and your power use goes over and above the 12kW to which the AC generation is limited for consumption or export, but I don't know how you'd set this inverter up to do that.

 

 

Doc CT.xlsx 246.94 kB · 1 download

Brilliant, thank you. So If I sum up the AC column and divide by 365, I get to 65.6kWh per day, which is actually not far from my estimate of about 68kWh per day on average. This is fantastic validation.

[Scorp007]

6 hours ago, PierreJ said:

This online estimator uses the weather for previous years at your location in order to estimate yields going forward:

https://pvwatts.nrel.gov/pvwatts.php

It's the best one that I'm aware of, and it's free.

I do find their results for winter in Gauteng very optimistic based on own yields when running at full power all day. 

[display_Name]

Ok, so an update on my side. Contacted SolarMD. The Solis inverters don't communicate with the SolarMD batteries. They would need to measure voltage. Not ideal so I'm going to look at other options. Will revert with more in the next day or 3 after some more hunting around. 

I see online that Deye have a 6kW unit, but I don't see any suppliers here in South Africa. Their 5kW units only support 13A input current, where the Solis 6kW supports 16A.

I have some more homework and will revert once I've done this.

Would love to see when these arrive on the market. 18A input current on the 6kW: https://www.deyeinverter.com/product/hybrid-inverter-1/SUN3-6-5-6-7-6-8KSG05LP1EU-3-68kW-Single-Phase-2-MPPT-Hybrid-Inverter-LV-Battery-Supported-2344.html

In other news... https://mybroadband.co.za/news/energy/532079-solar-silicon-prices-plummet.html