1 hour ago, Stu_D said:

So here is the latest… I now have 12 panels facing North on one string. (2 series strings in parallel) and 6 panels facing South in a series string. 
A new problem appears to be that the North string is reaching the maximum current of 20Amps and then pegging out. 
I can solve this by splitting the 2 strings to have 9 panels each with both strings having 6 North panels and 3 South panels. 
I am wondering if this will affect the output as the south panels may drag the north panels down. 

Yes you will do all the changes just to see you have a throttle due to the south panels. Rather throttle the strings at 20A than throttling the whole system. 

As you have indicated end of May that your 10A panels only produce 2.8A on south string that is what you will get for both strings due to your north series with south.

Edited June 20, 20233 yr by Scorp007

14 hours ago, Scorp007 said:

Yes you will do all the changes just to see you have a throttle due to the south panels. Rather throttle the strings at 20A than throttling the whole system. 

As you have indicated end of May that your 10A panels only produce 2.8A on south string that is what you will get for both strings due to your north series with south.

When I first installed I had 6 North panels paralleled with 6 south panels and I didnt notice any significant change when I split the strings into a North and South string. I have read that my panels (Canadian solar) have built in bypass diodes so that shaded panels do not affect performance of the entire string. With this in mind it may be better to do what I proposed earlier to reduce the Amperage.

16 hours ago, Stu_D said:

the North string is reaching the maximum current of 20Amps and then pegging out

Let's be very careful here first... we need to be sure of the mppt specs on your inverter because I think they allow only 11A per mppt, which means the panels on each mppt need to be in a series string, but we need to be sure to stay under the max Voc of the mppt too to see how many panels in series to put on each mppt. I would then suggest to put the max number of panels you can in series on the N string and the rest on the SW string.

Can you please confirm the mppt specs of your specific inverter?

 

Edit: from what I can see online on the Deye 5kW is 11A per mppt, max 500Voc, which would mean the best would be to put 10 panels on the N roof in series and 8 on the SW in series.

Edited June 21, 20233 yr by jumper

It is a Deye 8kW and has a maximum mppt current of 22 Amps per mppt. The inverter pegs out the current at 20 Amps.

I also checked my output in the history from one day to the next when I originally split the string into two and there was absolutely no difference in kW output afterwards. This leaves me to believe the bypass diodes are very effective and I should still get maximum output if I do an even split of 9 panels North and South. 10 panels will push the limit for the Voc as it will be nearly 500V.

10 minutes ago, Stu_D said:

It is a Deye 8kW and has a maximum mppt current of 22 Amps per mppt. The inverter pegs out the current at 20 Amps.

I also checked my output in the history from one day to the next when I originally split the string into two and there was absolutely no difference in kW output afterwards. This leaves me to believe the bypass diodes are very effective and I should still get maximum output if I do an even split of 9 panels North and South. 10 panels will push the limit for the Voc as it will be nearly 500V.

I that case I would leave it as is. You have previously only managed 10.1A when you had 6 in series on the N roof, so with 2 parallel strings of 9 you will only get approx 20.2A, so if it is throttling at 20A then you won't be losing much. I think if you add N & S panels in the same string you will have a lot more losses overall.

3 hours ago, Stu_D said:

When I first installed I had 6 North panels paralleled with 6 south panels and I didnt notice any significant change when I split the strings into a North and South string. I have read that my panels (Canadian solar) have built in bypass diodes so that shaded panels do not affect performance of the entire string. With this in mind it may be better to do what I proposed earlier to reduce the Amperage.

What one finds after change is always better than what one thinks should happen. 

Yes it's true that about all panels do have bypass diodes in. The problem is each panel with shade will produce no power for the portion that the bypass diode is in operation as the voltage is only about 0.5V. The diodes only allow power from the not shaded panels to flow through. 

Should you have 7 panels and 4 do have shading and your inverter needs 150V to operate and the remaining 3 produce less than 150V the the whole string is of no use. 

Shade on 1-3 columns of cells is OK as only the 3 will be in bypass. The problem is when all 6 columns get even just a bit of shade. Giving shade to say the bottom 12 cells on a panel one can easy see what loss in generation you will get. 

I stand to be corrected on what is mentioned above.