Hi Coulomb,
thank you very much. I´ve flashed the test Inverter well. I will update the other 3 later today.

btw, the Axpert MKS4 5,6kw Models FW 74.02 with 120v MPPT Startup have basically the same sticky 120v issues as the Max models.

55 minutes ago, mjmx8k said:

btw, the Axpert MKS4 5,6kw Models FW 74.02 with 120v MPPT Startup have basically the same sticky 120v issues as the Max models.

Yes, I was a little surprised to find that this problem goes way back to the earliest high voltage SCC models. It's surprising that people haven't been complaining about this for years.

Hi Coulomb,

i have updated all 4 Max1 8k inverters now to the modified version you made for me.
Strange that on 2 of them the Led bar is working correct and on 2 of them the Led bar is glimming light red only - as it was before with the U2 12.20 All for of them show correct version numbers for U1 and U2
The 2 where the led bars is working lost all parameters and i had to enter them again. the 2 where the led bar is glimming only remembered the parameters. 2 of them was spinning up the vents during update, 2 of them not. 

Can bus connection is established and they are online in 3Phase mode. To see PV it was too late here today.

Should i try to reflash the 2 with fancy red glimming led bars again or reset them to factory defaults and reconfigure them?  Any way, i dont need the led bars, but somehow this tells me there is maybe more difference inside the box.

thank you

 

Hi, 

 

My led bar was dead with dark red flimmer after 2 weeks of operation.

 

Restart & completely power off did not bring it back to life.

It seems to be an hardware fault .

 

This happened without any firmware flash and was just one morning dark.

Hallo Georg,

hmm, even if the bar is dead, in theory it should stop glimming when set parameter 51 to off / Lds
and interestingly 3P3 was glimming with U12.20 and started working today immediately with MCU Update U12.21. 3p1 and 3P2 did not stop glimming even on 51 Lds
maybe i have different Versions of Display chips...

Hi,

is it correct that it starts up with 80v when in new firmware is set 175v?

1 inverter is still sticky at 96v

Coincidence? the pink line sticky inverter is the one where the led bar is working correct 🙂
 

Switching grid power off and on is reinizializing the MPPT and then MPPT start working at panel voltage

If you read the complete thread you will find information about the internal supply which "stole" energy from both mppt inputs .

If we disable this , you loose the batteryless operation.( Maybe we could steal energy from ac out from the inverter ).

If the solar input is higher than 0.2A then it jumps over the limited 80-110v from internal supply.

 

This is nothing whats happens in the firmware of the inverter.

93 watt? Inverter 3 use only 1 MPPT.

1 hour ago, Georg594 said:

If the solar input is higher than 0.2A then it jumps over the limited 80-110v from internal supply.

it was locked in at 80v 0,2A and after grod off/on it immedately went to 286v 0,5A

2 hours ago, mjmx8k said:

it was locked in at 80v 0,2A and after grid off/on it immediately went to 286v 0,5A

Interesting. The Voltronics algorithm isn't proper Maximum Power Point Tracking, it seems to be a sort of "hill climbing": it tests a little in one direction, if the power increases, it keeps going in that direction. As soon as the power goes down, it stops going in that direction. It never seems to do a complete "scan". So it can get stuck at a local maximum, which may be far from the global maximum.

I'm confused by the current shooting up from 0.2 to 0.5 A. Is it possible that the sun increased just as you were turning the grid input off and on? If you load a panel too much, the voltage goes down, but the current should stay very nearly the same, for a given light intensity (insolation).

Do you happen to have two strings on that MPPT? Or are there perhaps significant shading issues? Either of these might cause more than one peak in the power curve, which could cause the algorithm to get stuck at low voltage.

I assume that this firmware is running with Georg's full patch set, with minimum MPPT voltage set to 175 V?

It's interesting that the PV power seems to go to zero for short periods of time. This is reminiscent of the "dips" (but thankfully not the "freezes") that plagued I think King firmware a while back. Maybe the dips were never properly resolved? At least the amount of time at zero power seems to be short, but it seems to me that it should not be happening.

Hi Coulomb,

the pink line 3p3 inverter has 1 MPPT only with 1s7p (7) 300w mono on exactly the same roof and direction as all other panels. there was no sun peak today. Inveter 3P1 and 3P2 have 2s6p (12) 300 mono on MPPT1 and on MPPT2 on the same roof same direction and they did not show any watt change at this point of time. so definitively no sun peak.

so the pink 3P3 should go almost linear to the other MPPTs just with less watts as its 2100wp compared to 3600wp on the other trackers.

it stayed there for more than 1,5 hours until i dropped the grid breaker

i run the 175v version you made for me 2 days ago:
Firmware version
VERFW:00192.04

the red circle is breaker drop time. you can see other mppts was all nicely and very narrow each other increasing watts while pink line first stay horizontal and after breaker action it goes similar in relation to other trackers. 

at this time 08:00 until 12:00 are no significant shading issues. between 10:00 and 11:15 there is absolutely no shading obstruction, later after 13.00 there is a 1000m mountain between the panels and the sun now in winter.

-----------------------

Later in the afternoon i had again Error 80 CAN BUS DATA LOSS on Inverter 3P1.
I removed this inverter and mounted the spare inverter as 3P1 now.

After some downtime during replacement and full restart now ALL LED bars are working again correctly 🙂
so i guess maybe yesterday after flushing the firmware i maybe did not keep them powered off long enough until fluctuating voltage was gone away....

 

 

"
 with Georg's changes to prevent getting "stuck" near 90V PV voltage
  Uses the 175V MPPT minimum
"

"
Can't implement the next patch from Georg; was a compare 200 use 13 -> cmp 35 use 4
"

Today 3P3 MPPT1 (pink line) 1S7P started without any manual interaction, but still looks like it was locked in some time, while 1P3 and 2P3 MPPT1 and MPPT2 started very well.

1 hour ago, mjmx8k said:

Today 3P3 MPPT1 (pink line) 1S7P started without any manual interaction, but still looks like it was locked in some time,

Looks like as soon as it hit 0.5A it started to work properly.

With Patched firmware 90.06  it looks like this (0.2A is the magic point)

 

 

 

I wonder if I goofed with the 192.04 patches somehow. I had a quick look, and the compare 50s are definitely now cmp 05s. I'll have another look when I get some time.

Looks like Georges Inverter starts 0,2+A and my Inverters over 0,5+A

I had another strange thing today as finally there was a little bit more sun.

On a bit higher PV Power over 2 amps Inverter 2 started to make a mess on both MPPTs while Inverter 1 was completely nice. Panels of P1 and P2 are identical and side by side on the roof, so its not a shading issue. I checked connectors already but could not find any bad connection. and as it happens on 2x 2S6P identical strings (and identical to Inverter P1) i cant believe its a wiring or faulty panel issue in both MPPTs of this inverter.

some days ago before the firmware update there was a similar sunny day with similar load but these drops did not happen.

 

earlier day:
 

@Coulomb checked your changes and everything looks perfect.

 

@mjmx8k

Do you have a dc clamp meter for measuring the real pv input current? 

(Beware of using normal series current measurements with multimeter because of the non isolated dc bus )

My axpert inverter has an offset of 0.12A until 0.16A on both mppts.

If my inverter shows 0.2A the real measurements are 0.34A .

 

Maybe this could be the answer 🙂😂

Yes, i have a 2 finger DC clamp meter and and a Seaward PV100 test kit with a DC Clamp
i will check this.

I hope that you are both zeroing your clamp meters before the reading; Hall effect sensors are notorious for offset and drift. Zero with the meter near the cable you are about to measure, though with at least 50 mm clearance. That's to approximately cancel the effect of the earth's magnetic field. But you probably know all about that.

yes, i will do that. thats my smallest issue at the moment....

7 output power losses on different single phases the last some hours and now again F80 on 3P1 with complete power loss of couse. either its me stupid or these inverters are...

i spend now 20kw of diesel to drive there to restart the system to gain 8kw of green power LOL