Newbie with mistakes

[Plaashaas]

Hi guys.

I feel like I should start this post like an AA meeting..... Hi my name is Plaashaas...... Some of my friends were doing it.....I tried a little one day, then some more, now I cant get enough. Think I will steal my mother in law's car to buy more Solar!

If only I found this site sooner I would have made better choices on my startup system. With time, and this very cool forum, it will get better. The amount of knowledge and expertise here is overwhelming and I now know that really don't know anything.

For now what I have:
9x330W Canadian Solar

Mecer Axpert SOL-I-AX-5M ..........Yes I know ;-(

4x Royal 1150K 105Ah (Should have gone Pylontech)

Watchpower monitoring (Sort of, I need to permanently install old laptop at inverter)

 

My biggest concern is the way the panels was connected by the electrician. 6 in series, 3 in series, these 2 strings in parallel. The Max PV Array OCV is 145VDC. Does this make sense? I am in the process of shopping for 3 more panels bring me to 12 total. But I see on the online spec sheet that the MAX PV ARRAY is 3000W, Does this mean I have maxed already with the 9x330W?

 

The other problem I have is the the inverter will switch to utility when there is say a big cloud moving over. After the cloud passed it will not switch back when the PV can handle the load? I have done it manually before by switching off the utility input and the inverter was happy with the load. Does this switch over just take some time or is this a setting issue?

Thank you for a great forum. 

[wolfandy]

Hi Plaashaas

I am also a newbie here and have recently added PV to my old Axpert setup. Let me try to answer your questions based on my knowledge/experience (more experienced users here please correct where I am wrong):

Quote

My biggest concern is the way the panels was connected by the electrician. 6 in series, 3 in series, these 2 strings in parallel. The Max PV Array OCV is 145VDC. Does this make sense?

Do you mean that one of your strings consists of 6 panels in series (6S) and the other string only 3 in series (3S)? Then this would be wrong as your 6S string takes you above your 145VDC. I also have 9 panels (315W, so similar) and mine are connected in 3 strings of 3S

Quote

I am in the process of shopping for 3 more panels bring me to 12 total. But I see on the online spec sheet that the MAX PV ARRAY is 3000W, Does this mean I have maxed already with the 9x330W?

I recently had a similar thought - maybe read through this thread: https://powerforum.co.za/topic/4644-what-can-i-realistically-expect-from-my-panels-would-it-make-sense-to-add-more/

Quote

The other problem I have is the the inverter will switch to utility when there is say a big cloud moving over.

This sounds like you have your inverter set to 'SUB' (solar -> utility -> battery). I have mine set to 'SBU' (solar -> battery -> utility) and the inverter then draws the difference from the batteries if PV is not sufficient to supply the loads

[Coulomb]

I would hope that actually they are wired 3S3P. A string of 6 paralleled with a string of three would possibly damage the string of 3, possibly cause a fire.

I suspect he has connected two strings of 3S on the roof with Y connectors, and brought that down with the other string of 3, to parallel the whole lot after breakers or fuses. In other words, two strings of three panels are paralleled on the roof; that set of 6 panels and the other string of 3 panels comes down to a junction box for final paralleling.

Unfortunately, your panels are almost certainly 72-cell. So 3S is too much if you get any sort of cool weather (say less than 10°C). You won't have a problem till winter, and even then you might get away with it, possibly losing some production due to the solar charge controller cutting back power above 130 V.

[Plaashaas]

On 2020/01/09 at 11:50 AM, wolfandy said:

This sounds like you have your inverter set to 'SUB' (solar -> utility -> battery). I have mine set to 'SBU' (solar -> battery -> utility) and the inverter then draws the difference from the batteries if PV is not sufficient to supply the loads

I checked again and is definitely in SBU setting. But still it would switch over to Utilty and and not back again (unless the load drops ridiculously low, +-300W)
I tested it yesterday around 14:00 by switching off Eskom and then running some of my pumps at 2.8kW. The system could do it. When I switch on Eskom again it would drop the Solar. 
At this stage I am considering leaving Eskom off to get some sort of return on investment. Can I do this? If the load is too much for the inverter, will it just trip? And will it restart on its own?
 

On 2020/01/09 at 1:04 PM, Coulomb said:

I suspect he has connected two strings of 3S on the roof with Y connectors, and brought that down with the other string of 3, to parallel the whole lot after breakers or fuses. In other words, two strings of three panels are paralleled on the roof; that set of 6 panels and the other string of 3 panels comes down to a junction box for final paralleling.

You are correct, Coulomb. I am thinking of changing it to 2S6P when adding another 3 panels. (I have that space on my frame)

 

Yesterday I added an old Laptop with Watchpower that will now monitor the system to see what is happening during weekdays when I am not there. 

 

Thank you for all the replies and advise. 

[Coulomb]

4 hours ago, Plaashaas said:

At this stage I am considering leaving Eskom off to get some sort of return on investment. Can I do this? If the load is too much for the inverter, will it just trip?

Yes, when the battery voltage falls below the value in setting 29 (low DC cutoff voltage).

Quote

And will it restart on its own?

I believe it should, though there has been a post recently where it seemed not to. Of course, it will have to see a higher battery voltage (e.g. charged by solar), or you reconnect AC-in.
 

[Plaashaas]

Good morning guys. 

Thank you for the replies. 

Sparkie is coming on Saturday to change panel configuration (And adding 3 more) to 2S6P. 

I also tested al 4 batteries and one is at 8.4V. The other three is 13+. These batteries are a month old. Why would one be defective? From factory?

[Plaashaas]

Hi Guys

I have changed my Inverter from the Mecer  (SOL-I-AX-5M) to the Kodak King. I just could not take the "laziness" of the Mecer anymore. 

Yesterday something weird happend. Any Ideas?

 

Also I am looking for 3x330W Canadian Solar panels (CS6U-330P) but cant seem to find stock.If anyone has please let me know. (Gauteng area)

I have also picked up some weird production from 9x330W panels:

In theory the panels should at best produce 2970W (100%)

[Coulomb]

1 hour ago, Plaashaas said:

Yesterday something weird happend. Any Ideas?

It looks like the charger decided that the battery should not be charged or discharged at that point. I assume that this is with the King, a PylonTech, and there is a suitable cable and the King is set up with battery type = PYL. You should check for warnings like 60 or 69 when this is happening. I note that 54.0 V is 3.6 VPC, right at the upper limit, so that would be why the BMS told the King not to charge or discharge the battery. That's why the battery voltage stayed at exactly 54.0 V. The utility was supplying all the load, so there was nowhere for any solar power to go. So it stayed at 0.

Eventually the BMS saw the battery voltage drop to below that 54.0 V mark, so it stopped telling the inverter not to use the battery, and suddenly the battery voltage dropped all the way to 52.1 V. This triggered another bulk charge. It overshot (not unusual) to 56.6 V then 56.4 V, before settling at 54.0 V again. Probably the battery had disconnected at that point.

So you need to figure out why your battery voltage is so high. Are there any battery related settings available when the battery type is set to PYL? (Assuming my guesses are correct, of course).

[Plaashaas]

Hi Coulomb

At this stage the inverter is set to SUB. I am only getting my Pylontech today so this was with lead acids. I will change to SBU with the new Pylontech. 

14 hours ago, Coulomb said:

The utility was supplying all the load, so there was nowhere for any solar power to go. So it stayed at 0.

With the King, when set to SUB, it should first use all the available power for the load and only "fill in" what is short from utility, if any. So that is why I cant understand the PV input dropping to 0? (It only returned to PV when I temporary switched off Utility)

Edited January 28, 2020 by Plaashaas

[Coulomb]

3 hours ago, Plaashaas said:

So that is why I cant understand the PV input dropping to 0?

I think because the battery is over-full. But that begs the question, why is the battery over-full? I suspect that it's a misunderstanding between the PylonTech BMS and the inverter firmware.

This stuff seems like it should be so easy, but I guess the devil is in the detail.

[Plaashaas]

4 hours ago, Plaashaas said:

At this stage the inverter is set to SUB. I am only getting my Pylontech today so this was with lead acids. I will change to SBU with the new Pylontech. 

With the King, when set to SUB, it should first use all the available power for the load and only "fill in" what is short from utility, if any. So that is why I cant understand the PV input dropping to 0? (It only returned to PV when I temporary switched off Utility)

 

36 minutes ago, Coulomb said:

I think because the battery is over-full. But that begs the question, why is the battery over-full? I suspect that it's a misunderstanding between the PylonTech BMS and the inverter firmware.

This stuff seems like it should be so easy, but I guess the devil is in the detail.

You missed the part where I mentioned it's Lead Acid battery. 

[___]

21 hours ago, Plaashaas said:

Yesterday something weird happend. Any Ideas?

Look at the battery voltage column. Pegged at 54V. Which I assume is your float voltage. So my guess... inverter went to float, and there was nothing for the MPPTs to do until the voltage drifted down low enough for them to pick up again.

But I don't see it at absorption voltage prior to that, it was just barely over 54V at 13.28. Premature float bug, @Coulomb?

In any case, then at 14:28 the voltage pulls down low enough and another Bulk cycle starts.

[Plaashaas]

My question is why would the PV input drop to 0 when there is a load of 3kW. The setting at this stage is SUB so the battery should not have an influence on this? 

[Coulomb]

18 hours ago, Plaashaas said:

 

You missed the part where I mentioned it's Lead Acid battery. 

Arrgh. Sorry.

[Coulomb]

18 hours ago, plonkster said:

Look at the battery voltage column. Pegged at 54V. Which I assume is your float voltage. So my guess... inverter went to float, and there was nothing for the MPPTs to do until the voltage drifted down low enough for them to pick up again.

Except that this is a King, with the double conversion hardware. It should have been supplying as much as possible from solar, and only the remainder (if any) from utility. SUB mode should put S (Solar) before U (Utility).

18 hours ago, plonkster said:

But I don't see it at absorption voltage prior to that, it was just barely over 54V at 13.28. Premature float bug, @Coulomb?

Yes, that's the main premature float bug. They compare with the float voltage (less 0.5 V from memory) instead of the bulk/absorb voltage (less the same 0.5 V), for the purposes of deciding whether to transition from absorb to float stages.

So it doesn't even have to have charged the battery to float voltage, even less than that will do if the charge current is low enough.

[Coulomb]

16 hours ago, Plaashaas said:

My question is why would the PV input drop to 0 when there is a load of 3kW. The setting at this stage is SUB so the battery should not have an influence on this? 

Indeed. So I'm back to considering that this is a dip, as per this topic. Though Jaco seemed to experience a lot more dips, but for a short period of time each.

My theory is that the dip/freeze bug was introduced into the firmwares for many models in the early part of 2019. Various firmware updates have claimed to fix it, but I have seen no confirmation of a fix as yet.

[Plaashaas]

On 2020/01/29 at 6:13 AM, Coulomb said:

Indeed. So I'm back to considering that this is a dip, as per this topic. Though Jaco seemed to experience a lot more dips, but for a short period of time each.

My theory is that the dip/freeze bug was introduced into the firmwares for many models in the early part of 2019. Various firmware updates have claimed to fix it, but I have seen no confirmation of a fix as yet.

Thank for the reply. I suspect it is related to the then lad acid batteries. It has not happened since I connected the Pylontech on Tuesday. 

[Plaashaas]

On 2020/01/27 at 2:35 PM, Plaashaas said:

Hi Guys

I was playing around the Inverter on the weekend and the same thing happened. But this time I realized what the actual problem is. The Inverter can handle 60-115VDC and that is why it would cut off the panels. So I have 330W panels 3S4P with 37.7V each. Can I add some sort of Voltage regulator to not go over the the 115V? 

It seems to only happen when there is a cool breeze on the panels. 

Also with the 3x US3000's it has become less of a problem because during that time I still don't use Eskom anyway. 

Then on a separate matter, I have a 2.2kW (Single phase) irrigation pump. When I start it from the inverter it trips the inverter. Why? And can I add another Capacitor or something to assist in the start? 
(Yes it does run of Single phase Eskom)

Any advise would be appreciated. 

Thx

 

 

[Coulomb]

2 hours ago, Plaashaas said:

I was playing around the Inverter on the weekend and the same thing happened. But this time I realized what the actual problem is. The Inverter can handle 60-115VDC and that is why it would cut off the panels.

The 115 V figure, as far as I can tell, is marketing. The real limit for full power (for models with a 145 V max MPPT) is 130 V. Power output at 135 V is 67%, at 140 V is 33%, and at 145 V it is zero. Straight line linear ramp.

However, in your sample, it does look like the MPPT simply disconnects above 115 V. Can you pull some logs from other days to see if it's consistent? If so, it seems that I'm missing something. On reading through the MPPT firmware for the constant associated with 115.00 V (integer 11 500 or 0x2CEC ), I find no occurrences, but there are plenty of occurrences of 14 500 and 13 000 (representing 145.00 V and 130.00 V respectively).

What is your secondary (U2) firmware version number? I'm looking at 04.10, but the others (04.00 and 01.24) are similar.

[Plaashaas]

Hi Coulomb

Thx. Just remember that my inverter is the Kodak King. So how do I cap the Voltage at 115V then? 

• Solar Charger Type: MPPT.
• Maximum PV Array Power: 4000 W.
• MPP Range @ Operating Voltage: 60 VDC ~ 115 VDC.
• Maximum PV Array Open Circuit Voltage: 145 VDC.
• Maximum Solar Charge Current: 60 A.
• Maximum AC Charge Current: 60 A..

[Coulomb]

3 hours ago, Plaashaas said:

Then on a separate matter, I have a 2.2kW (Single phase) irrigation pump. When I start it from the inverter it trips the inverter. Why?

A moderately sized motor like this pump motor typically draws about 7x rated current at start-up. So that's 2.2 x 7 = over 15 kVA (with low power factor). That's over 3 times the apparent power rating of a 5 kVA inverter; they claim to do 2x rated for a few seconds (10 kVA). 

Quote

And can I add another Capacitor or something to assist in the start? 

There are things called motor soft starters, I don't know much about them. A variac could be used to manually dial up the speed of the pump motor, but they're bulky and you'd soon tire of turning the large knob manually.

[Coulomb]

25 minutes ago, Plaashaas said:

Thx. Just remember that my inverter is the Kodak King.

Yes. My understanding is that they use the same MPPT as Axpert MKS 5 kVA models. [ Edit: ah, but they might impose the 115 V limit in the DSP or MCU firmware. Though I can't find it with a quick search. ]

Quote

So how do I cap the Voltage at 115V then? 

I'm not convinced that this is your problem. But if it is, you could possibly build effectively a 500 W 114 V zener diode, using a large heatsinked MOSFET or IGBT and a simple control circuit. It would need to be well ventilated and have a very large heatsink, possibly fan cooled. You could perhaps test the idea by wiring a 100 W (or so) incandescent lamp across the panel output; note that you need a proper DC rated switch (e.g. a NOARK breaker) to safely turn it on and especially off. When the inverter output goes to zero, turn on the lamp, which should drop the panel voltage, and see if that makes the inverter's solar charge controller wake up. If the voltage drops below 115 VDC but the inverter does not restart, then the theory is blown.

[ Edit: If the 115 V strict limit theory holds up, ] it might be easier to replace every third panel with a 60-cell panel with a similar Imp specification (probably 60/72 = 5/6 power rating of the 72-cell panels). But that might just postpone the problem till winter (when the panel voltages will be higher, by some 8%). So the real solution is to wire all the panels 2S instead of 3S.

Edited February 25, 2020 by Coulomb