Parameter 12 Voltronic

[EmilBenvenuto]

I think I have a problem with my Axpert Max 7.2 KW inverter, I have paired it with three Pylontech US3000C 48V batteries, I also have two strings each of six 375W panels. I have set inverter setting 1 to SBU, (ie: Solar power provides power to the loads as first priority. If the solar power is not sufficient to supply all the connected loads, the battery power will supply power to the loads at the same time.

The AC grid will only supply power to the loads when the battery voltage drops below the low level alarm voltage or below the reference value in setting 12). Instead the 12 setting I have left it default at 46V, however I have noticed that the inverter, especially in winter, trips to supply power from the AC even if the batteries still have three notches, so not using all the stored energy, how do you think I would solve this?

[BritishRacingGreen]

1 hour ago, EmilBenvenuto said:

I think I have a problem with my Axpert Max 7.2 KW inverter, I have paired it with three Pylontech US3000C 48V batteries, I also have two strings each of six 375W panels. I have set inverter setting 1 to SBU, (ie: Solar power provides power to the loads as first priority. If the solar power is not sufficient to supply all the connected loads, the battery power will supply power to the loads at the same time.

The AC grid will only supply power to the loads when the battery voltage drops below the low level alarm voltage or below the reference value in setting 12). Instead the 12 setting I have left it default at 46V, however I have noticed that the inverter, especially in winter, trips to supply power from the AC even if the batteries still have three notches, so not using all the stored energy, how do you think I would solve this?

It is not unusual for a battery's voltage to fall during  times when there is a substantial sudden discharge eg. a hairdryer / kettle etc.  That fall in voltage might be low enough to trigger setting 12 (back to grid). Do you have logs/graphs  that you can check this. 

If you cannot do anything about it, you can set the back-to-battery voltage setting  just above setting 12.this will reduce the grid import time, and before you know you back to SBU. 

Edited July 1 by BritishRacingGreen

[BritishRacingGreen]

39 minutes ago, BritishRacingGreen said:

It is not unusual for a battery's voltage to fall during  times when there is a substantial sudden discharge eg. a hairdryer / kettle etc.  That fall in voltage might be low enough to trigger setting 12 (back to grid). Do you have logs/graphs  that you can check this. 

If you cannot do anything about it, you can set the back-to-battery voltage setting  just above setting 12.this will reduce the grid import time, and before you know you back to SBU. 

Note to @Coulomb : I have noticed on the MAX 7.2 that the machine assumes 'native' offgrid mode when in SBU. in other words the safety/grid relay is negated until either the mode is changed or setting 12 is triggered. Why so, is it a pedantic measure to save grid energy by illiminating the need to import the odd 100W buffer power from the grid when the machine is ongrid.? 

Edited July 1 by BritishRacingGreen

[Coulomb]

2 hours ago, EmilBenvenuto said:

The AC grid will only supply power to the loads when the battery voltage drops below the low level alarm voltage or below the reference value in setting 12). Instead the 12 setting I have left it default at 46V, however I have noticed that the inverter, especially in winter, trips to supply power from the AC even if the batteries still have three notches, so not using all the stored energy, how do you think I would solve this?

A few factors.

1) There is an undocumented (except by me, of course ) interaction between setting 12 (back to utility battery voltage) and setting 29 (low battery cut-off voltage). See FAQ #2).

2) The factory firmware does not use load/charge compensation (like fully patched firmware does, where it is called KettleKomp™). Also covered in the above FAQ.

3) Factory firmware still has "lead acid thinking". The LFP flavour of fully patched firmware has lowered thresholds that are far more suitable for LFP batteries, which most of is are using these days.

1 hour ago, BritishRacingGreen said:

how do you think I would solve this?

If you have recent enough software and your battery type is set to PYL or similar, then setting 12 can work with SoC% instead of bare, uncompensated battery voltage. It's hit and miss at this stage as to which firmware pairs (you need capability in the DSP and in the display firmware for this to work). I'd say that this would mitigate the problem somewhat.

Unfortunately, unless you are using a lot PV voltage 48 V model, there will never be fully patched firmware. It's just too much work, and my colleague and I don't have a model to test these comprehensive changes on.

[Coulomb]

1 hour ago, BritishRacingGreen said:

I have noticed on the MAX 7.2 that the machine assumes 'native' offgrid mode when in SBU. in other words the safety/grid relay is negated until either the mode is changed or setting 12 is triggered. Why so, is it a pedantic measure to save grid energy by eliminating the need to import the odd 100W buffer power from the grid when the machine is ongrid.? 

Only when the output source priority is SUB and AC-in is present and within specification does the Voltronic firmware connect AC-in to AC-out and the AC side of the DC<->AC converter together, which is what allows blending of AC-in with battery and/or solar power. All the high PV voltage Axpert firmwares seem to be like this.

Edited July 1 by Coulomb

[EmilBenvenuto]

You are fantastic! It's great to find experienced people who help people in trouble! Thanks to your excellent answers, I think I have understood the problem. I will describe my current setup of parameters that might be useful to you:

In the last message I misstated the model of my battery, I actually have Pylontech Turbo Energy US2000C

5: PYL
12: 48V (obviously wrong)
13: 53V (perhaps too much?)
29: 45V

Now, as you can see I think my inverter is set up wrong as I am not an expert and you certainly know more than me.

Could you tell me on the basis of my batteries (whose data sheet I enclose anyway US2000C-specs.pdf) the correct settings for these parameters? Thanks very much.

[Coulomb]

1 hour ago, EmilBenvenuto said:

5: PYL

I assume that you're not getting warnings about battery comms, so you must have a cable between the inverter's BMS port and the battery's BMS port. Is that correct?

Or do you have ICC or similar that connects to the battery's BMS on the inverter's behalf?

1 hour ago, EmilBenvenuto said:

12: 48V (obviously wrong)

Actually, I think that's good.  At least, that's what is in my recommendation image from years ago. Bear in mind that I don't use a Pylontech or any battery with a factory BMS.

1 hour ago, EmilBenvenuto said:

13: 53V (perhaps too much?)

I have 51, or 50 if you want to reduce utility use,  especially if you have more than 3 battery modules (of US2000 size).

1 hour ago, EmilBenvenuto said:

29: 45V

I have 46.5 V, which effectively makes setting 12 48.5 V, which is kinder to the battery. It could go down to 46.0 V to make the battery run longer during load shedding, but suffer a longer life. This is the one to adjust to get the behaviour that you want.

Other battery settings will be over-ridden by the battery's BMS via the cable.

Actual Pylontech owners may care to add their recommendations.

[EmilBenvenuto]

For parameter 5 everything is OK as it is a mode provided by my inverter for communication to the Pylontech battery. I have 3 US2000C in total.  The default 12 is 46V, could the battery data sheet which says Discharge Voltage (V) 44.5 ~ 53.5 be useful?
The parameter 29 actually in practice what should it be used for? As safety? Because then 12 and 29 should do the same thing in theory.

[Coulomb]

10 hours ago, EmilBenvenuto said:

could the battery data sheet which says Discharge Voltage (V) 44.5 ~ 53.5 be useful?

No. That 44.5V is the absolute lowest voltage that you can get any useful energy from the battery. There is no implication that this is safe to do regularly. 

10 hours ago, EmilBenvenuto said:

The parameter 29 actually in practice what should it be used for? As safety? Because then 12 and 29 should do the same thing in theory.

It won't be used for its original purpose in your case, as the battery BMS should cut off well before this setting. It's designed to stop discharging the battery in case the battery gets too low, and further discharge would damage the battery and possibly the inverter. 

Setting 12 should come into effect earlier, i.e. the value of this setting should be higher than that of setting 29. Setting 12 is where you should routinely start charging if possible. But sometimes charging isn't possible. In that case, you can run the battery down a bit lower, for that presumed rare case. Note that regular load shedding breaks that presumption. 

So yes setting 29 is meant to be an emergency thing, used rarely. Despite this, its undocumented effect on setting 12 is permanent. 

[EmilBenvenuto]

Thanks for the information. Do you think this configuration will work?

12: 46V
13: 50V
29: 44V

I have no idea if 12 is too little, but isn't 48V too much? I don't know how a battery works from a technical point of view.

Also, do you think it is essential that I update the inverter's firmware?
Connecting with watch power I read this data:
cpu version: 69.01
remote panel cpu version: 112.19
ble cpu version: 0

[Coulomb]

16 hours ago, EmilBenvenuto said:

I have no idea if 12 is too little, but isn't 48V too much?

48 V for a 15S battery is 48/15 = 3.20 VPC. Assuming light loads, that's already well down the sloping part of the voltage versus SoC curve, so we can say that this is about 14-15% SoC. If you have higher loads, then the SoC is actually a little higher at that battery voltage, assuming no compensation for loads.

Personally, I try to keep my LFP cells above 20% SoC, even though the latest trend seems to be to go down to 10% SoC. That's why I suggest a higher setting 29, making the effective SoC for setting 12 more like 48.5 V, though that might be getting closer to 25% SoC. So perhaps 46.2 V for setting 29.

You really need to experiment with these values.

16 hours ago, EmilBenvenuto said:

cpu version: 69.01
remote panel cpu version: 112.19

Those are rather old now. Consider upgrading to main (U1, DSP) version 69.70, and removable display version 12.28.

This is factory firmware version 69.70 for the Axpert Max E 7.2 kW non-twin, with the '28066 DSP. Do not use with any other model, not even other Axpert Max 7.2 kW models whose existing firmware doesn't start with "69". Use at your own risk. Hex file is dated 6th March, 2024.

Firmware upload instructions for models with a removable display.

 

MAX_HV7.2K_MAIN_6970.7z

[EmilBenvenuto]

I have never flashed firmware on an inverter, and I wouldn't even know how to do it, but I would tend to prefer not to, except to have the Kettle Comp modification. What would be the firmware containing this modification for my inverter? Also, I have another question, although it has nothing to do with this topic, the Voltronic inverter parameter 2, i.e. Maximum charge current by default is 60A, should I change it?

[Coulomb]

On 2024/07/04 at 9:50 PM, EmilBenvenuto said:

I have never flashed firmware on an inverter, and I wouldn't even know how to do it,

It's all in the upload instructions, which I link to from every firmware post that I make.

On 2024/07/04 at 9:50 PM, EmilBenvenuto said:

I would tend to prefer not to, except to have the Kettle Comp modification.

Unfortunately, KettleKomp™ is a complex patch, so it's only available for the 48 V battery, 145 V max PV models, e.g. Axpert King 1. You really need a way to adjust the strength of the compensation, so that involves display firmware as well as main firmware for most models.

On 2024/07/04 at 9:50 PM, EmilBenvenuto said:

Voltronic inverter parameter 2, i.e. Maximum charge current by default is 60A, should I change it?

That depends on how many inverters you have, and what battery. Every battery has a maximum charge current rating.