Hi all,

I'm new to the solar component of the game, I'm hoping to get some practical advice from the experienced solar folks on the forum.

First, a breakdown of my current system:

• Synapse 3.0V+ 3000VA/2400W 24V inverter
• 2 x 12v, 200Ah LFP batteries in series (total 5.2kWh)
• HA11 battery balancer
• Raspberry Pi 1B running SolPipLog for monitoring

The above is professionally installed via my DB board, with COC, etc, and provides great backup power for load shedding and outages. It excludes geysers, stove, kitchen appliances, and pool pump, but does include fridges. My consumption is generally between 200-600W at any given moment, with occasional peaks up to 2000W. I'm not in a position to rip and replace with a more robust Sunsynk 48V system, as has been suggested to me a number of times - I'd just like to make the best of what I have.

Now, the inverter (an Axpert type) is quite limited in terms of solar capability. Below is the spec of what it can take:

The panels I'm keen to get are the Jinko Tiger 475W:

I've used @Chris Hobson's brilliant solar calc tool, and it looks like the best I can do on the above spec is 2 x 450W-ish panels in parallel. Panels in series exceed the permissible cold VOC and VMP:

Below is solar production analysis from the PV Calculator app, 30 degree roof tilt, -110 degree azimuth (facing South):

• Based on the above, I would be producing about 3.2kWh per day (1151 kWh per year, ~96kWh per month). Assuming 6 hours of PV production time per day.
• Based on my very loose average system consumption of 400w, I am using about 9.6kWh per day (excluding geysers, pumps, outbuildings, etc which are not on the system).
• It looks like I would therefore cover about 1/3 of my consumption with the PV I want to add.

My questions:

1. Are my assumptions that I can only do 2 x 475W panels in 2 x single panel strings correct?
2. Are my calculations more or less correct?
3. Is it it worth me adding just 2 panels to my setup given the production and consumption estimates above?
4. Can I use a separate MPPT controller instead of the built-in one on the inverter, so that I can add more panels?
5. Can I use a separate MPPT controller in conjunction with the built-in one?
6. If I used the built-in MPPT controller, would the infamous "premature float" bug kill my LFP batteries prematurely?
7. Am I wasting my time with this idea overall? The reputable solar installer who installed my system has quoted me just shy of R20 000 for the additional panel install, with COC, etc. This excludes the cost of the panels themselves which I will buy separately. The initial install cost about the same, and I bought the inverter and batteries myself.

My goal is not to be self-sufficient here. It is merely to supplement my power demands, with the idea that I could charge my batteries off solar instead of Eskom (SBU mode), and perhaps cover my idle load (assuming and average of 9.6kWh per day). I also really like the idea of having some production capability in the event of an extended power outage (a bit of Mad Max thinking here 😄). Any opinions and advise are welcome, and thank you for reading this far.

Edited October 5, 20232 yr by jbroo

Bump.

 

Hi all,

I'm new to the solar component of the game, I'm hoping to get some practical advice from the experienced solar folks on the forum.

First, a breakdown of my current system:

• Synapse 3.0V+ 3000VA/2400W 24V inverter
• 2 x 12v, 200Ah LFP batteries in series (total 5.2kWh)
• HA11 battery balancer
• Raspberry Pi 1B running SolPipLog for monitoring

The above is professionally installed via my DB board, with COC, etc, and provides great backup power for load shedding and outages. It excludes geysers, stove, kitchen appliances, and pool pump, but does include fridges. My consumption is generally between 200-600W at any given moment, with occasional peaks up to 2000W. I'm not in a position to rip and replace with a more robust Sunsynk 48V system, as has been suggested to me a number of times - I'd just like to make the best of what I have.

Now, the inverter (an Axpert type) is quite limited in terms of solar capability. Below is the spec of what it can take:

The panels I'm keen to get are the Jinko Tiger 475W:

I've used @Chris Hobson's brilliant solar calc tool, and it looks like the best I can do on the above spec is 2 x 450W-ish panels in parallel. Panels in series exceed the permissible cold VOC and VMP:

Below is solar production analysis from the PV Calculator app, 30 degree roof tilt, -110 degree azimuth (facing South):

• Based on the above, I would be producing about 3.2kWh per day (1151 kWh per year, ~96kWh per month). Assuming 6 hours of PV production time per day.
• Based on my very loose average system consumption of 400w, I am using about 9.6kWh per day (excluding geysers, pumps, outbuildings, etc which are not on the system).
• It looks like I would therefore cover about 1/3 of my consumption with the PV I want to add.

My questions:

1. Are my assumptions that I can only do 2 x 475W panels in 2 x single panel strings correct?
2. Are my calculations more or less correct?
3. Is it it worth me adding just 2 panels to my setup given the production and consumption estimates above?
4. Can I use a separate MPPT controller instead of the built-in one on the inverter, so that I can add more panels?
5. Can I use a separate MPPT controller in conjunction with the built-in one?
6. If I used the built-in MPPT controller, would the infamous "premature float" bug kill my LFP batteries prematurely?
7. Am I wasting my time with this idea overall? The reputable solar installer who installed my system has quoted me just shy of R20 000 for the additional panel install, with COC, etc. This excludes the cost of the panels themselves which I will buy separately. The initial install cost about the same, and I bought the inverter and batteries myself.

My goal is not to be self-sufficient here. It is merely to supplement my power demands, with the idea that I could charge my batteries off solar instead of Eskom (SBU mode), and perhaps cover my idle load (assuming and average of 9.6kWh per day). I also really like the idea of having some production capability in the event of an extended power outage (a bit of Mad Max thinking here 😄). Any opinions and advise are welcome, and thank you for reading this far.

You can over panel a bit like 180W more than whaf you indicated. The problem is you might be better of using 3x420W panels all in parallel. 

For a average yield throughout the year 5hrs is a closer average than the 6hrs estimated. This is for Gauteng. 

You can increase your PV by using an external MPPT controller connected to some more panels charging the same battery. Here we look at a 30-40A charging current to battery. 

This will make a bigger dent in the grid use if you can try and shift load to day time. 

 

I have just that inverter installed in my in-laws house and  I have 4 330w panels connected 2 series 2 parrel so 2 strings so input into the inverter is around 80v and around 14amps .

This inverter is rated 1000w mppt controller but have never seen it give more than 500w from the controller .

So then an external mppt controller will be best but they costly so save yourself the trouble and get an 3kw inverter that has an 4000w high volts controller this giving your self lots of room to upgrade with time 

Edited October 5, 20232 yr by GMAC

7 hours ago, GMAC said:

 

I have just that inverter installed in my in-laws house and  I have 4 330w panels connected 2 series 2 parrel so 2 strings so input into the inverter is around 80v and around 14amps .

This inverter is rated 1000w mppt controller but have never seen it give more than 500w from the controller .

So then an external mppt controller will be best but they costly so save yourself the trouble and get an 3kw inverter that has an 4000w high volts controller this giving your self lots of room to upgrade with time 

Costly is relative. A 30A external MPPT cost from R2500. 

21 hours ago, jbroo said:

Below is solar production analysis from the PV Calculator app, 30 degree roof tilt, -110 degree azimuth (facing South):

Why are you calculating the yield with panels facing South? Do you have an option to face the panels towards North? If you do, then around 1kW of solar panels installed should generate approx 4kWh daily on average throughout the year, on the understanding that Durban's location is relatively poorer for solar generation. I'm going on solar resource maps for SA.

I think the idea is good to install 2 panels to try to carry your daytime baseload, keep the batteries topped up during daytime loadshedding, and give you a slightly higher degree of autonomy. But not for R20,000 excl panels. At that type of installation cost I'd start looking at another installer, or at a bigger system entirely since you'd be getting more value for money on the fixed costs of installation.

Thanks @GreenFields

30 minutes ago, GreenFields said:

Why are you calculating the yield with panels facing South? Do you have an option to face the panels towards North?

My mistake, I checked the app again, I had actually orientated them NE. My roof does not allow for a direct N orientation.

1 hour ago, GreenFields said:

then around 1kW of solar panels installed should generate approx 4kWh daily on average throughout the year

Thanks, this is good to know.

1 hour ago, GreenFields said:

But not for R20,000 excl panels

Yeah, I agree. It feels excessive, and this is mainly what's holding me back. If it were not for this quote, I would probably have bought the panels and jumped right in already.

I was thinking of orientating one panel (or string) NE, and another NW, to catch morning and afternoon sun, respectively.

Do you think this is a good idea?
Can I therefore also exceed the inverter's rated wattage somewhat, and go for 2 x 600W panels for example?

On 2023/10/06 at 11:28 AM, jbroo said:

Thanks @GreenFields

My mistake, I checked the app again, I had actually orientated them NE. My roof does not allow for a direct N orientation.

Thanks, this is good to know.

Yeah, I agree. It feels excessive, and this is mainly what's holding me back. If it were not for this quote, I would probably have bought the panels and jumped right in already.

I was thinking of orientating one panel (or string) NE, and another NW, to catch morning and afternoon sun, respectively.

Do you think this is a good idea?
Can I therefore also exceed the inverter's rated wattage somewhat, and go for 2 x 600W panels for example?

Yes  you can especially if you have them facing as in your pic and connected  in parallel as only one panel will get  the direct  sun at a time . 

I'm testing 3 340w panels now connected in parallel all facing in one direction to see how that inverter performs with 40v setup . Been cloudy and rainy the last 2 days  hopefully tomorrow will be a better day for solar 

 

On 2023/10/06 at 11:28 AM, jbroo said:

Thanks @GreenFields

My mistake, I checked the app again, I had actually orientated them NE. My roof does not allow for a direct N orientation.

Thanks, this is good to know.

Yeah, I agree. It feels excessive, and this is mainly what's holding me back. If it were not for this quote, I would probably have bought the panels and jumped right in already.

I was thinking of orientating one panel (or string) NE, and another NW, to catch morning and afternoon sun, respectively.

Do you think this is a good idea?
Can I therefore also exceed the inverter's rated wattage somewhat, and go for 2 x 600W panels for example?

Given the low input and operating voltages range so the inverter, I'd go for two 600-ish W panels in series. They usually have a Vmpp around 35V and a Voc of 42V which fall well within the operating bounds of the inverter. I would not mix the directions of the panels as you suggest in the above image. Can you fit two panels in landscape orientation on the NNW facing roof?

4 hours ago, AlexTZA said:

I'd go for two 600-ish W panels in series

Thanks, this is what I gravitated towards also. Below is with the 600W Canadian Solar panels offered by powerforumstore:

The total wattage is exceeded here, but I doubt it would ever reach 1kW even.

4 hours ago, AlexTZA said:

I would not mix the directions of the panels as you suggest in the above image

Why not? My reasoning here was to get more direct sunlight time over the course of the day, at the expense of more wattage at a particular time. I'd like to know your reasoning since I'm fairly new to this.

My main hurdle now is an affordable install. Can someone recommend a decent installer in the Durban area to just install and add panels to my system?

53 minutes ago, jbroo said:

Thanks, this is what I gravitated towards also. Below is with the 600W Canadian Solar panels offered by powerforumstore:

The total wattage is exceeded here, but I doubt it would ever reach 1kW even.

Why not? My reasoning here was to get more direct sunlight time over the course of the day, at the expense of more wattage at a particular time. I'd like to know your reasoning since I'm fairly new to this.

My main hurdle now is an affordable install. Can someone recommend a decent installer in the Durban area to just install and add panels to my system?

Going 2 directions is fine as long as you go parallel which in this case you can as the MPPT range is from 30V which is below the voltage of 1 panel. Series not what I would do.

@jbroo

To clarify, you can take either approach - series in one direction or parallel in two directions will both work. If the split directions is easier for you, use that. The 600W panels have a normal operating voltage near the bottom end of MPPT operating range if used in parallel - personally I'd prefer to not be that close to the minimum due to the slight drop in efficiency but that's just me.

To illustrate, I used the SMA Sunny Design simulator to illustrate the differences in total yield between the two approaches. I chose a random location in central Durban, ignored shading, and approximated the panel azimuths in the NNW and ENE directions you want to use and set the panel tilt to 15 deg. I used 550W panels and I had to add multiple panels to each array so that the configuration would work voltage wise with one of the SMA inverters but the ratio of the overall yield will be comparable to your setup. Note the annual energy yield/performance figures. The difference in overall annual yield between the two configurations is about 4%.

Mixed NNW and ENE in parallel on single string (polystring):

 

Single NNW array on single string:

 

 

Thank you @AlexTZA it seems like a single string is the way to go, facing NNW.