52 minutes ago, Calvin said:

My data so far indicates that I never get PV drop-outs on the 2S5P inverter, but quite often on the 3S4P ones (always both together)

Thanks so much for doing that.

53 minutes ago, Calvin said:

the 3S make the SCC run very hot (30C hotter than the 2S,

I know that converters are more efficient when the "turns ratio" (ratio of output voltage to input voltage) is closer to unity. But I didn't think it would be that dramatic! I guess that explains why some installations with voltage near but not actually exceeding the limit are still causing dips; it's limiting due to temperature.

57 minutes ago, Calvin said:

I have however seen no disadvantages to 2S,

From the inverter's point of view, no. But there is no escaping that 2S is more expensive to wire compared to 3S or even 4S in a few cases, and that's presumably what drives people to keep trying it.

Thanks for the vindication 😎

@Calvinthanks so much that is really valuable for me, appreciate you sharing! 

53 minutes ago, Cliff said:

@Calvinthanks so much that is really valuable for me, appreciate you sharing! 

You are welcome.  I have learned so much from these forums, especially people like @Coulomb and @Youda, that it is good to be able to give something back.

I haven't read through this entire forum topic, but note other users reported Axpert king solar PV production issues were fixed with firmware upgrade:

 

@pierre.  Thanks for the link back, I can confirm that after firmware upgrades my system started behaving much better. 

 

Main inverter software version - 71.92  https://powerforum.co.za/files/file/52-axpert-king-main-firmware-version/

Controller / Display software version - 02.49 https://powerforum.co.za/files/file/53-axpert-king-remote-display-firmware/

 

My system should technically be able to give me 6480 watts, if I work on the % given by Coulomb earlier in the post than I am spot on with my solar production.

Reaching almost 5400 watts the other day at around 11 already. 

 

I am not sure if summer will bring through much change, might reach the 5400 watts much earlier than expected.

 

Factors to take into consideration:

Roof angle very flat compared to most roof structures

Being on a hill and winter the sun tracks low across the horizon each day

54 minutes ago, Blowdart18 said:

@pierre.  Thanks for the link back, I can confirm that after firmware upgrades my system started behaving much better. 

 

Main inverter software version - 71.92  https://powerforum.co.za/files/file/52-axpert-king-main-firmware-version/

Controller / Display software version - 02.49 https://powerforum.co.za/files/file/53-axpert-king-remote-display-firmware/

@Blowdart18

I have upgraded to the same firmware for 2 Kings in Parallel but are now experiencing issues with the "Max charging current" jumping around. Are you seeing this?

Discussed in detail HERE

 

21 hours ago, Calvin said:

My data so far indicates that I never get PV drop-outs on the 2S5P inverter, but quite often on the 3S4P ones (always both together),

After looking at my log files in more detail I need to make a correction.  It is NOT always both together.  Sometimes only one of the 3S SCCs trip.

I have had a good look at a few other incident and what they all have in common:

1. Inverter (always wired 3S) has spare PV, so the MPPT voltage has been raised to reduce power output.  Typically around 135V
2. Sun suddenly pops out behind a cloud.
3. Voltage goes up to over 145V. SCC trips. 
4. Takes between 3 and 15 minutes to restart.

Below is an extract from by database - in this case both the 3S inverters tripped. The column PVm show independently measured solar power, calibrated to correspond to what my system can produce.  It clearly shows the sun popping out from behind a cloud.

	I1PvWatts	I1PvVolts	I1AcWatts	I2PvWatts	I2PvVolts	I2AcWatts	I3PvWatts	I3PvVolts	I3AcWatts	PVm	BatteryWatts
2020-08-11 11:06:12	1733	133	708	1275	88.2	762	1259	136.4	745	5770	1786
2020-08-11 11:06:12	1874	133	708	1362	88.2	762	1259	136.4	745	6110	1873
2020-08-11 11:06:13	1874	133.2	824	1362	88.6	806	1259	136.7	867	6720	1873
2020-08-11 11:06:14	2025	133.2	824	1512	88.6	806	1505	136.7	867	7420	1837
2020-08-11 11:06:15	2025	137.3	890	1512	88.6	806	1505	138.7	917	7990	1837
2020-08-11 11:06:16	2035	137.3	890	1512	88.6	806	1374	138.7	917	8830	448
2020-08-11 11:06:17	2035	137.3	890	1512	88.6	806	1374	138.7	917	9060	448
2020-08-11 11:06:18	2035	137.3	890	1512	88.6	806	0	138.7	917	9440	-1558
2020-08-11 11:06:19	2035	145.8	983	1512	92.2	938	0	146.4	1024	9750	-1558
2020-08-11 11:06:20	2035	145.8	983	1512	92.2	938	0	146.4	1024	9750	-1558
2020-08-11 11:06:20	0	145.8	983	1619	92.2	938	0	146.4	1024	9310	-1845
2020-08-11 11:06:21	0	146.9	1084	1619	92.4	1057	0	146.4	1091	8930	-1845
2020-08-11 11:06:22	0	146.9	1084	1666	92.4	1057	0	146.4	1091	7600	-2163

It is interesting that I have only started seeing this since changing my charging paradigm from "get to 100% SOC asap" to "get to 100% SOC as slow as you can".  Under the former the SCC was always at full power until the batteries were full, which means MPPT voltage is around 114V, which means no tripping even when the sun pops out behind a cloud.  Once the batteries were full it made very little difference if some SCCs tripped - one is normally enough to supply my loads.  Under the new paradigm the SCC rarely operates at full power (due to the restricted charging currents) so the MPPT voltage generally hovers in the mid 130s.

Lesson from this: Either live with these drop-our, or do not wire 72 cell panels 3 in series.

 

22 hours ago, Coulomb said:

Thanks for the vindication 😎

Final confirmation, if it was needed.

I took 1 string away from my one 3S inverters, and added 2 of the panels to the 2S inverter.  Now my configuration is: 

• Inverter 1: 3S3P
• Inverter 2: 2S6P
• Inverter 3: 3S4P

Good testing weather today, sun is continuously in and out from behind the clouds.

What I saw:

• Inverters 1 and 3 still drop PV when sun pops out.  Inverter 2 does not, despite now also having 12 panels.
• Inverter 1 & 3 still run much higher MPPT temperatures than inverter 2, this despite inverter 1 having less PV on it than inverter 2.

Off I go to buy some more cables...🤣

3 hours ago, Calvin said:

Lesson from this: Either live with these drop-our, or do not wire 72 cell panels 3 in series.

@Calvin, you've come to the right conclusion. But for other readers, I'd like to point out that:

• 145 V is not the start of PV power throttling, it's the end, and the point where the hardware could blow up.
• The start of PV power throttling is 130 V (undocumented).
• The upper end of "MPPT range" in the specifications is 115 V.
• These Solar Charge Controllers, at least the ones in the MKS models that I'm familiar with, have no way of disconnecting from the PV input. When the PV "drops out", this isn't some "fail safe" condition. The two relays on the SCCs disconnect the output of the SCC from the battery, not the PV terminals from the buck converter input.
• All this is worse at colder temperatures, since panel voltage increases with decreasing temperature. So while the SCC might shrug off a few 146 V situations in the middle of the day, one cold morning with unusual weather conditions it might encounter 152 V and blow up, possibly taking other parts of the inverter with it. In fact, just recently Weber had to visit a remote client whose inverter failed (LC Display comes on only briefly, then off for a time, and repeats). It has a 3S 72-cell configuration, and there is evidence that it disconnected PV the day before. It was unusually cold and windy weather. He's now preparing a replacement inverter, and will recommend that the client gets the array rewired to 2S.

So to me the lesson to be learned is: do not wire 72 cell panels 3 in series, or risk blowing up your inverter. This of course only applies to models with the 145 V (absolute!) max PV voltage.

7 hours ago, Blowdart18 said:

I am not sure if summer will bring through much change, might reach the 5400 watts much earlier than expected.

@Blowdart18 This table from geyserwise is pretty good as a reference for how much difference there is between the months.  Don't mind what the actual value is in this table, just calculate the percentage relative to the best summer month, then do the math to predict what you will get in that month based on what you currently get.

https://www.geyserwise.com/wp-content/uploads/2016/03/0207_PV-system-installation-manual_web.2016March09.pdf

1 hour ago, pierre. said:

This table from geyserwise is pretty good as a reference for how much difference there is between the months

One really important thing that they don't point out: This is per square metre of horizontal surface.

If you tilt your panels the figures change dramatically.  At about 40% from horizontal, facing North, the difference becomes very much smaller.

Below the output for Stellenbosch for panels at 0%.  Note how the daily output numbers are close to the Cape Town ones from Geyserwise.

And here is the output at 40 degrees tilt.

As you can see, at 40 degrees tilt the difference between the months is much less.

Edited August 12, 20205 yr by Calvin

Which power optimizers are compatible with the axpert range inverters?

Hi guys.

I fitted 8 x 480w panels (2s4p) to my Axpert King 5000w 5KV unit with 8 x 135ah agm (4s2p). Been going through recommended settings etc and trying different combinations and have only seen a pv output of around 1700w on one occassion when using a microwave.

Pv voltage is fluctuating between 83v to around 66v.

I don't seem to get the batteries charged and in most cases when checking I see in watchpower battery discharge showing 1 to 2 amps but occassional 1amp  on the charging side.

Generally shows pv wattage at around 500-700w then when using a load it switches to battery and is around 51v.

Its like the solar is not charging or keeping the battery topped up.

have to change to SUB in the evening to charge the batteries from the utility.

Have inverter settings as follows:

01. SBU

02. 80amp

05. Use

10. Eco and tried Auto 

11. 40amp

12. 49v

13. 54v

16. SLB-UCB

23. BYD

26. 57V

27. 54.6

29. 48V

32. AUTO

From this thread it seems to be similar to my issue.

Over the weekend i have updated the firmware to the recommended 71.97a to solve the premature float bug. Downloaded the firmware from the aeva forum for the PIP-5048mk inverter.

Haven't noticed any difference after updating the firmware.

Weather is a bit overcast today and will check again tomorrow provided weather clears up.

 

12 hours ago, razzor13bt said:

Over the weekend i have updated the firmware to the recommended 71.97a to solve the premature float bug. ...

Haven't noticed any difference after updating the firmware.

I haven't identified the cause of this issue, so there are no patches addressing this issue. 71.97a only fixes the premature float issues, nothing else.

Presumably you can see U1 71 97A on the front panel using the up or down buttons.

As a point of interest, what was the version of firmware that came with your inverter?

My unit had v71.90 originally and i updated to v71.92 last year.

I had a charging issue previously and the display went blank so i sent the unit in under warrantee.

It came back a month later and worked fine but was told nothing was wrong with it.

 

been keeping an eye on it when i have a chance at home.

on a bright sunny day with no clouds and inverter set to solar for load only.

If on SBU it will switch to battery if i switch on the kettle and microwave to test.

On SUB i get a pv wattage between 1400w to max 1700w.

surely the panels wont be operating at less than 50% efficiency.

 

Yip my king is doing the same, only way too get PV to work right is to Set to SBU then manually change to SUB to ensure batteries don't go too low.  Very frustrating.

U1 71.97 and u2 02.66

 

When Batteries are full. Pv is poor

 

When on SBU its fine.