Hi All,

Trying to understand why my inverter just drops PV at times. Constant load in blue, with green being PV. 

 

The drop in PV production was cloud rolling in, which from guessing and reading on the forum, could be a voltage spike and MPPT tripping. However, looking at the voltage it doesn't seem to have gone over the max:

 

Inverter main firmware: 71.97

Any thoughts on what this could be? My gut is telling me it's a wiring issue.

 

Thanks
Grant

Layman's 2c. Consider with caution. Looks like it's just the MPPT responding to some fluctuating spikes in the load, hunting to find the new optimum power point by increasing the voltage, but topping out against the maximum panel voltage. And then swapping all load to mains or battery until it re-calculates/finds the correct point to operate at. Might be interesting to know how the panel current changes during this time, if it re-sets to a higher level after the blip.

So the voltage stays high at around 80v, and then wattage slowly ramps up. But something that concerns me is that during these dips, I can very clearly hear a relay flip/click off and then a few moments later flip again, and only after the second flip/click does the wattage ramp up.

And as the wattage ramps up, so does the voltage come down to "normal working levels" 

 

Just went back and saw more drops - this time around, no spike in voltage. So starting to wonder if the MPPT isn't on it's way out. Will enquire from Voltex about this behavior when they are open again

13 hours ago, grant.cpt said:

Trying to understand why my inverter just drops PV at times.

I think it's just because your battery was full, or at least the inverter thought it was full because of the premature float bug. So the PV drew zero for a while, hence the increase in PV voltage. After a minute or two, the battery voltage dropped to the float voltage. This is normal. The battery SOC would have dropped a fraction of 1% SOC; that's the small price you pay for longer life at the float voltage.

If you are bitten by the premature float bug, it's fixed in patched firmware version 71.97a, usually paired with patched display firmware version 02.66a.

5 hours ago, Coulomb said:

I think it's just because your battery was full, or at least the inverter thought it was full because of the premature float bug. So the PV drew zero for a while, hence the increase in PV voltage. After a minute or two, the battery voltage dropped to the float voltage. This is normal. The battery SOC would have dropped a fraction of 1% SOC; that's the small price you pay for longer life at the float voltage.

If you are bitten by the premature float bug, it's fixed in patched firmware version 71.97a, usually paired with patched display firmware version 02.66a.

Hi @Coulomb

Today I had another drop - this time around the battery % was around 93~94. Would this still be that premature float bug? Seemed to have happened as load increased, which would mean a drop in battery voltage.

 

 

4 hours ago, grant.cpt said:

Would this still be that premature float bug? Seemed to have happened as load increased, which would mean a drop in battery voltage.

Without seeing the battery voltage, it's hard to say.

But that one might be with the charge stage already at float (perhaps prematurely; it's sometimes a LOT premature), and when there is a bit of a load, the battery voltage drops too much, and the system goes back to bulk charging. But Voltronic for reasons best known to themselves, only go to bulk via the zero charge state. That lasts for the best part of a minute, causing the dip that you see. That switch back to bulk charge stage could be caused by the lack of battery voltage compensation (for load or charge current, in this case, load current), and will be alleviated with KettleKomp™. The problem would be exacerbated by a battery that was a bit low on capacity, considering the loads. How big is your battery?

KettleKomp™ will be introduced to the patched King firmware when Weber and I get to it; it's waiting for work from me.

Edited January 2, 20223 yr by Coulomb

2 hours ago, Coulomb said:

Without seeing the battery voltage, it's hard to say.

 

From the look at the latest PV drop. Battery voltage was 53.10v before the voltage dropped to 48.5v at it's lowest.

 

2 hours ago, Coulomb said:

How big is your battery?

1 Single PylonTech US3000C - which I think is a bit too small to handle really high spike loads. But the latest spike the battery was full, and solar production was good to cover more than 2/4 of the load.

14 hours ago, grant.cpt said:

1 Single PylonTech US3000C - which I think is a bit too small to handle really high spike loads.

Yes. Even one more battery module will really help.

14 hours ago, grant.cpt said:

But the latest spike the battery was full, and solar production was good to cover more than 2/4 of the load.

LFP batteries are actually moderately high internal resistance when they are full: they drop their voltage the most for any given load in the top ~5% and the bottom ~20% of the SOC range.

14 hours ago, grant.cpt said:

Battery voltage was 53.10v before the voltage dropped

That's just under the 53.2 V that Pylontech recommends to charge the battery to, so I'd guess that the BMS was ready to tell the inverter-charger not to charge any more, if the charger was listening.

Are you using the BMS cable (either direct to the removable display, or via a monitoring Pi)?

If not, what is your absorb/bulk charge voltage setting?

45 minutes ago, Coulomb said:

Are you using the BMS cable (either direct to the removable display, or via a monitoring Pi)?

 

Direct BMS cable to removable display

 

46 minutes ago, Coulomb said:

If not, what is your absorb/bulk charge voltage setting?

Bulk I see it set to: 53.2

Absorb - not sure which setting this falls under (Battery Float?)

3 hours ago, grant.cpt said:

Direct BMS cable to removable display

Ok, so that dip was likely caused by the battery getting nearly full. I suspect that the Pylontech BMS won't allow charging again until the battery voltage has dipped a bit. But of course, just because there can't be any net charge to the battery, doesn't mean it can't or shouldn't cover loads. That seems like a firmware issue, which is disappointing since 71.97 is the latest that we have access to.

Maybe it's just the firmware's silly habit of responding to nearly any change in charging conditions with zero charge for nearly a minute.

I might have another read of the firmware tonight and see if I can figure out how it responds to a "stop charging" command from the BMS.

5 minutes ago, Coulomb said:

Pylontech BMS

The battery was manufactured in the start of 2021 - I've seen chatter about the Pylon battery module firmware upgrade. Do you think this is perhaps, something to consider?

18 minutes ago, grant.cpt said:

chatter about the Pylon battery module firmware upgrade. ... something to consider?

Sorry, I have no idea.

4 hours ago, grant.cpt said:

Absorb - not sure which setting this falls under (Battery Float?)

Absorb and bulk are really the same setting. Bulk is a bad term for this (but Voltronic use it), because the battery voltage only reaches this setting at the end of the bulk charge stage, but it tries to stay at that same voltage for the whole of the absorb stage. So it's much better called the absorb voltage setting. I call it the absorb/bulk battery voltage setting.

Sorry for the confusion.

33 minutes ago, Coulomb said:

Sorry for the confusion.

No stress. Makes more sense now too.

1 hour ago, Coulomb said:

Maybe it's just the firmware's silly habit of responding to nearly any change in charging conditions with zero charge for nearly a minute.

I might have another read of the firmware tonight and see if I can figure out how it responds to a "stop charging" command from the BMS.

Indeed. I see code that sends a reset to the Solar Charge Controller (this will guarantee zero solar charging for nearly a minute) every time the stop charging flag changes. So once when the battery becomes full and net charging should stop, and again when net charging is allowed again.

So even if you have a large load on, when the battery becomes full, a near-one-minute dip in PV production is baked in, so the loads have to be supported by the battery for that near-minute.

To fix this would require some careful thought, which is probably why it's not fixed.

24 minutes ago, Coulomb said:

so the loads have to be supported by the battery for that near-minute

Understood. I'll just have to get 1 or 2 more batteries, as I see the one module complains when a high load is pulled from it - Red Alarm light flashes for until solar ramps up, and less amps are pulled.

26 minutes ago, Coulomb said:

To fix this would require some careful thought, which is probably why it's not fixed.

Makes sense, perhaps 1 Pylon US3000C module isn't what Axpert Inverters should be paired with, but perhaps 2 as a minimum. As I'm sure 2 will be able to level out the voltage better.

 

But do wanna say HUGE thanks for digging into this. 

Hi all iv installed a brand new axpert king and it is showing me an error code 53 and when I looked it up it said  (soft start failed) is it an internal problem 

Fault code 53 doesn't sound like anything that could be caused by an installation error, unless perhaps there is a short circuit on the AC output. So that sounds like a warranty repair or replacement to me.