Inquiry About Low Battery Voltage Issue with MultiPlus 24/3000/70

DIY
, ,
1

Dear Sir/Madam,

I hope this message finds you well. I am currently conducting experiments using the MultiPlus 24/3000/70. Below is the system configuration and the issue we are facing:

System Configuration:

  • Generator: Maximum output of 1000W
  • Battery: 24V, 10kWh capacity
  • Operating Conditions: Continuous operation for 24 hours

Issue:

  • When a load exceeding 1000W (e.g., 1100W) is applied, the battery voltage gradually decreases.
  • Once the battery voltage drops below 20V, the MultiPlus detects a “Low Battery” alarm and ceases to function as an inverter.

Questions:

  1. Is it possible to adjust the settings on the MultiPlus (e.g., battery profile or alarm thresholds) to prevent this issue?

  2. Alternatively, is this phenomenon more likely due to the characteristics of the battery being used (e.g., degradation or temperature effects)?

  • For instance, if the voltage drop and overload are primarily caused by the discharge characteristics of the battery, I would appreciate advice on possible countermeasures.

I apologize for this basic question, but I would greatly appreciate your assistance in resolving this matter.

Thank you in advance for your response.

Best regards,

2

When you say that the battery voltage gradually decreases, over what period of time?
The battery voltage is expected to decrease, as you are pulling 1kw from the generator, and 100w from the battery, so the battery is in discharge and will take around 10,000wh/100w=100 hours to fully discharge.
If your battery has a fault, bad cell, bad connection etc, it might drop in voltage a lot faster than the expected.

Let us know what chemistry you have (lead, LiFePO4, etc) and whether your battery has a BMS which communicates with the multiplus via a GX device, but i suspect that from your question that the battery has no communication. In either case, you can change the 20v setting by using a laptop and MK3 cable and the free Victron VEConfigure software to configure this.
https://www.victronenergy.com/support-and-downloads/software

3

Thank you for your response.

1. Time Variation of Battery Voltage Drop

The “Low battery” alarm appeared in less than one day (24 hours). The battery capacity was over 80%, indicating that it was adequately charged. If the “Low battery” alarm were to appear when the battery is near 0%, battery degradation could be considered. However, despite being around 80% capacity, the inverter suddenly stopped with the “Low battery” alarm.

2. Battery Type

The battery in use is presumed to be a Lithium Iron Phosphate (LiFePO4) battery, based on the following specifications:

  • Cell Voltage: 3.2V
  • Capacity: 320Ah
  • Combination Configuration: 3P4S
  • Rated Capacity: 960Ah
  • Nominal Voltage: 12.8V
  • Max Charging Voltage: 14.6V
  • Discharge Cut-off Voltage: 10V
  • Internal Resistance: 0.13 ± 0.05mΩ
  • Weight: 80kg
  • Dimensions: 500 × 380 × 238mm
  • Operating Temperature Range:
    • Charging: 0°C–45°C
    • Discharging: -20°C–60°C

3. MK3 Cable

Is it correct to consider the MK3 cable as an item specifically designed for use with Victron Energy products? In other words, my understanding is that the MK3 cable can be purchased via the following URL and that it can be used with the designated software to adjust voltage settings:

MK3-USB Interface - Victron Energy

4. Voltage Settings

When the “Low battery” alarm is triggered and the inverter shuts down, it always happens when the battery voltage drops below 20V.

If the voltage setting for triggering the “Low battery” alarm were adjusted to 10V, would there be any safety risks involved?

Specifically, is it possible that altering the default lower and upper voltage settings might lead to malfunctions or other issues? Could making such changes inadvertently damage the system?

4

What you need to determine is if the battery was truly at 20v. If it was, then the shutdown was appropriate.
A 24v LiFePO4 at 20v has no power left

LiFePO4-Battery-Voltage-Charts-24v
LiFePO4-Battery-Voltage-Charts-24v250×650 39.6 KB

As your specs for the battery show,
Discharge Cut-off Voltage: 10V
So with 2 batteries in series, a discharge Cut-off Voltage for the pack of 20V is indeed appropriate.

Reducing that cutoff voltage for the pack to 10v would lead to permanent damage to the cells.

Now … other than a very bad cable issue, it is most likely that the battery was truly at 20v, in which case you have a battery that was either not fully charged when you started, or the load is actually higher than 100w, or the generator is not actually putting out 1000w.
To troubleshoot this, can you simplify the system? I would fully charge the battery, then put a 100w load on the battery, no generator, no other load, and graph the voltage. You can easily do this by setting a timer and recording the voltage of the battery with a multimeter every hour or half hour.
I think you will find that you have a bad battery, but if the 100w load does indeed take around 100h to flatten the battery, then the battery might be ok. To be sure, you need to repeat the test with a high-current draw, something like a 1kw load which should last for 10 hours. If you have a bad connection inside the battery, it might run OK with a 100w load (which is around 4A) but fail on a 1kw load (which is around 40A)

The MK3 cable is indeed specific to Victron. Based on your most recent post, i don’t think you need the cable, as it looks more like a battery issue to me.

5

Thank you for your response. I will try repeating the experiments and testing further.

One point to clarify: currently, I am keeping the load (AC loads) suppressed to ensure that power is not drawn from the battery, keeping the generator output below 1000W.

I had understood that the “Idle 95%” display represents the “current battery charge status.” However, is it correct to interpret this percentage as showing the Voltage Capacity instead?

VRM
VRM1080×2400 106 KB

6

Have you got a cerbo GX? Just go to the device list and Remote VEConfigure to download the file. Modify the settings and then upload it again. You can put in a low battery alarm recovery voltage.

The % you see is from the multiplus bu you need to add the battery capacity for that to be correct in the Remote VEConfigure file.

7

“Idle” just means that “not very much” is happening - the charge or discharge is very low. If a lot of current is leaving the battery, the label will say “Discharging”.
“95%” is the SOC of the battery, calculated by AmpHours in the battery divided by AmpHours Capacity of the battery, and as Owen says, the MultiPlus is doing the “counting” of the amps in and out, so its fairly accurate but not as accurate as when the calculation is done by the battery BMS or a shunt.

I didn’t realise you have a GX device. Go to the Advanced page, set the time period to something like 2 days, and then screenshot the interesting graphs - probably the one that shows battery voltage and inverter power

Here is one of mine

image
image498×451 43.6 KB

You can see the blue line goes between 26v and 27v - my battery doesn’t drop very much overnight because the battery is large (50Ah) compared to the load (a few Ubiquiti radios at 8w each, an NVR at about 60w, and some cameras at 7w each, and a RaspberryPi at ~6w, total = 41w)

image
image498×451 28.3 KB

My power usage is very flat (the graph is sharp only because the Y axis doesn’t start at zero) - min 38w, max 41w, which is probably similar to yours.

So … post your screen shots of graphs like these and we can probably work out whats happening

8

Thank you, everyone, for your responses.
I have prepared graphs showing the voltage and load conditions at the time the “Low battery” alarm was triggered. Please take a look.

画像
画像1486×696 40.1 KB

When checking the Battery Voltage and Current, the voltage remained stable until 15:00. However, after 15:00, the voltage dropped sharply, and the inverter stopped when it fell below 20V.

At this time, the battery’s state of charge (SOC) was maintained at over 90%.

画像 (1)
画像 (1)1467×682 15.4 KB

In addition, AC Output Power and Battery Power showed significant changes during this period, but the variation was limited to around 150W.

画像 (2)
画像 (2)1467×705 34.6 KB

I also checked the values set in VE Configure. The Battery Capacity was set to 480Ah, which is half the capacity of the 960Ah battery we are using. Could this discrepancy be a factor affecting accurate measurement?

画像 (4)
画像 (4)951×747 111 KB

Furthermore, the shutdown voltage was set to 18.6V. Our battery’s nominal voltage is 12.8V, and the maximum charge voltage is 14.6V. Should we consider the possibility that there is a mismatch between the specifications of the battery and the inverter? Normally, for the Multiplus 24/3000/70, it would be appropriate to use a 24V battery to ensure proper recognition.

画像 (5)
画像 (5)937×736 103 KB

What are your thoughts on this?

9

Lets start here

Untitled
Untitled1380×646 60.1 KB

During A, the current is positive, the battery voltage is rising, the battery is being charged, or towards the end more like “idle” as it alternates back and forth between charge and discharge.
During B, the voltage is dropping, but the current is apparently still positive, so this implies that there is some problem here with sensing current, or that there is a DC load somewhere that is not being measured.
During C, there is a brief period of charge
During D, there is a discharge of (if you can believe the Y axis) about 125w (5A x 25v) that results in voltage collapse. This is not a healthy battery - its SOC is very low.

During A, the charge appears to be around 56 watts, for around 7h, putting in about 18Ah - nowhere near enough to charge the battery, and i would suggest that the battery never even got to 10% SOC.
Your voltage during A should be around 29v, so you are 4v short of being fully charged.
The collapse at D is totally expected in this scenario.

You should fully charge the battery, and then repeat your test. The SOC reported in the 2nd graph is misleading you - we can work that out later, for now, you need to get that battery up to 3.5v per cell, 14v per battery, or 28v overall, and hold it there for around an hour on a small charge current (about 2A should be ok) to allow the cells to balance if the BMS wants to.

It is possible that with the capacity set to 480, the MultiPlus was counting the Ah going in, and reached 480 - only half the required energy, so fixing that might help a lot. Change that in the “General” page and upload the file to the MultiPlus.

The shutdown voltage is a red herring - it makes no difference if this is 20 or 18 or 16 - at that point in the voltage curve, there is nothing left in the battery. If it were my battery, i would set the shutdown voltage on the “Inverter” page to around 10% SOC which correlates to about 24.5v

1 Like
10

Thank you very much.

In other words, when the “low battery” alarm appeared, the actual battery charge level was much lower than the displayed state of charge (over 90%).

The reason why the displayed value is not accurate could possibly be due to the MultiPlus setting being configured for a smaller capacity of 480Ah.

I understand the situation now, and I will first try the suggestions you provided.

11

Yes, you understand perfectly.

12

Thank you very much. I will give this a try first.

There is one correction to be made regarding the battery specifications used by our company. The correct specifications are as follows:

▼Battery Group Specifications

■Cell

Model: 3.2V-202Ah
Nominal Capacity: 202Ah
Rated Voltage: 3.2V
Internal Resistance: ≤ 0.2mΩ

■Combination Standard:
A. Capacity Difference: ≤1%
B. Internal Resistance: ±0.02mΩ
C. Current-maintaining Ability: ≥90%
D. Voltage: ±0.01V

■Combination Method:
Single Cell 2P8S (28P8S)

■Battery Index

Rated Capacity: 404Ah
Minimal Capacity (0.3C5A): 400Ah
Nominal Voltage: 25.6V
Max. Charge Voltage: 29.2V
Discharge Cut-off Voltage: 17.6V
Charge Current: 100A
Working Current: 1–100A
Output and Input Terminals: P+ (red) / P– (black)
Weight: 75kg
Dimensions (L × W × H): 500 × 380 × 238mm

■Charge Method:
Standard: 100A × 12hrs
Quick: -

■Operating Temperature:
Charge: 0°C to 45°C (32°F to 113°F)
Discharge: –20°C to 60°C (–4°F to 149°F)

In this case, the battery capacity is 404Ah. The value set in VE Config is 480Ah. While there is a slight discrepancy between the values, this is significantly closer to the previous incorrect battery specification of 12V 960Ah.

Despite this, the GX device does not display the correct battery status (it shows over 80% battery capacity even when the battery is actually out of energy). Could there be a possible reason for this?

Regardless, I plan to set the correct capacity in VE Config. I apologize for the complexity of this question and thank you for your understanding.

13

One thing to check in the VE.Configure app is that the correct battery type is selected. You have a LiFePO4, but if the MP thinks that the battery is an AGM, its not going to be using the correct curves.

14

image
image1051×747 98.8 KB

I have checked, and the current battery type is set to “No corresponding default.”
I understand that it needs to be changed to “LiFePO4.”

15

Yes