Behavior while AC running in heater mode

[micmak]

Hello everyone,
We've had a local company install a solar system for us and it has been working great. I have a concern though about running the AC in heater mode. We're running two banks of 4 batteries, each battery is a tubular 12V/240AH. The inverter is a 5.2KW 48V 230V inverter (we use 220V where I live, 50Hz)

I have a victron smartshunt and I have my inverter connected to home assistant to log the data. I ran the AC in heater mode (~1500Watts) and it ran for 2 hours before the inverter cut off due to low voltage, from 3AM (prior to that the grid was available) to approximately 5AM:


Is this drop in voltage normal? Total home power usage was around 1700Watts, less than 50% the inverter capabilities. it happened twice before I woke up and turned off the heater at 4:50AM. Then without the load the voltage kept recovering. Victron SmartShunt was showing an SoC of 88% around that time.

I am aware under load the voltage drops, but then again state of charge is at 88%, way higher than the 50% I try to stay above as those are flooded lead acid batteries. What am I missing here?

This is my first post, so I apologize in advance if I broke any rules or posted in the wrong place.

- Michel

 

[sunshine_eggo]

You're pulling 1500W/12V/0.85 = 147A

That's a substantial current and a large voltage drop is expected. Worth checking that all your connections are tight, and that you have battery cables suitable for both current and the voltage drop. Should likely have 4/00 cables sized for your 3000W load.

 

[micmak]

Thank you for your feedback.

Shouldn't it be 1500W/48V/0.85? I have 2 battery banks in parallel, each 4x12V in series. As for the voltage drop, is the state of charge irrelevant here? or do I reduce the cut off voltage on the inverter? or I am expecting too much from my batteries?

let's assume that the wires are fine, I double checked the connection and everything is tight. I also calculate a while ago the needed cables and it matches with what they installed.

 

[Wibla]

I eyeball around 41 Volts from that graph, that indicates around 1.7V per cell, which is nowhere near 88% SOC under a relatively small (<0.1C) load.

 

[micmak]

I eyeball around 41 Volts from that graph, that indicates around 1.7V per cell, which is nowhere near 88% SOC under a relatively small (<0.1C) load.

It dropped to 41V when the heater was on and the load was around 1500W, then jumped back to 45V when I turned it off and kept going up to around 46V before the sun came up.

 

[RCinFLA]

I assume 'heater mode' means A/C reverse cycle heat pump in heating mode. It is not clear in your description if you have total 1700 watts from inverter or 1700 watts house loads plus 1500 watts for heat pump. I would expect that for about a 15-20 kbtu A/C.

It appears you are running 2p4s Optima-like lead acid batteries from your description of 12v 'tubular' batteries. These are likely AGM type batteries. I have not seen 240 AH spiral lead acid batteries before. Most I have seen is less than 100 AH batteries. Are you sure they are 240 AH batteries, not 240 CCA?

1700 watts would be about (1700 /0.9 effi.)/ 48v = 39 amps from battery array. That is not excessive current for 480 AH 48v lead acid battery array.

Dropping quickly to 48vdc would be fairly normal for lead acid batteries in the first five minutes of moderately heavy load. This would not be normal if they are LFP batteries which enforces my guess they are lead acid batteries.

What is not normal is only getting an hour before batteries start to fizzle for what is suppose to be 480 AH's of batteries.

Check each battery's terminal voltage when the voltage starts to fizzle.

Looks like your batteries are in poor condition, not 240 AH batteries, or you are drawing a lot more current from them than you think. You should get a battery monitor with current shunt so you can monitor their load current and AH of consumption from batteries.

 

[Mike Jordan]

A/C in heater mode? As in a heat pump?

Thats is a very energy sucking method to produce heat

 

[Wibla]

Thats is a very energy sucking method to produce heat

We're veering into offtopic here, but generally heat pumps made for colder climates are quite efficient heaters, generally giving you 2-5 times (depending on dT) the input power in usable heat.

 

[Mike Jordan]

We're veering into offtopic here, but generally heat pumps made for colder climates are quite efficient heaters, generally giving you 2-5 times (depending on dT) the input power in usable heat.

You are definitely veering off topic, as the opposite of that is true. You only find heat pumps in typically warm climates, where heat demand is very low. A heat pump running in a very cold climate will freeze over, as the exterior coils cannot exhaust air colder the the ambient temperature

 

[Wibla]

You only find heat pumps in typically warm climates, where heat demand is very low.

This is nonsense, I live in Norway, there's a shitton of heatpumps here, and they have good efficiency down to -15C to -20C, and some still provide net heat (over pure resistive heating) at -25C.

A heat pump running in a very cold climate will freeze over

When temps go below freezing, there's not a lot of humidity in the air to actually cause the coils to freeze over, so while this DOES happen, it only requires a defrosting cycle (lasting a few minutes on newer heat pumps) every few hours.

as the exterior coils cannot exhaust air colder the the ambient temperature

Please explain to me how they can't physically do so, as long as the refrigerant isn't outside of its rated temperature range.

E: Don't take my word for it, here's what the Canadian government has to say about heat pumps

 

[micmak]

I assume 'heater mode' means A/C reverse cycle heat pump in heating mode. It is not clear in your description if you have total 1700 watts from inverter or 1700 watts house loads plus 1500 watts for heat pump. I would expect that for about a 15-20 kbtu A/C.

It appears you are running 2p4s Optima-like lead acid batteries from your description of 12v 'tubular' batteries. These are likely AGM type batteries. I have not seen 240 AH spiral lead acid batteries before. Most I have seen is less than 100 AH batteries. Are you sure they are 240 AH batteries, not 240 CCA?

1700 watts would be about (1700 /0.9 effi.)/ 48v = 39 amps from battery array. That is not excessive current for 480 AH 48v lead acid battery array.

Dropping quickly to 48vdc would be fairly normal for lead acid batteries in the first five minutes of moderately heavy load. This would not be normal if they are LFP batteries which enforces my guess they are lead acid batteries.

What is not normal is only getting an hour before batteries start to fizzle for what is suppose to be 480 AH's of batteries.

Check each battery's terminal voltage when the voltage starts to fizzle.

Looks like your batteries are in poor condition, not 240 AH batteries, or you are drawing a lot more current from them than you think. You should get a battery monitor with current shunt so you can monitor their load current and AH of consumption from batteries.

Thank you for your extensive feedback, these are the batteries: Robuste Tubular Battery

"Tubular" batteries have flooded the market here since electricity became a luxury and the currency went down the drain. Anyway, that's off-topic


The AC is a very old 9k BTU cheap one, it seems to have ran all the time without taking any breaks.

I have a Victron Smartshunt already attached and the inverter measures current, both seem to agree (although the smartshunt seems to think the SoC is at ~80% still, not sure why). I didn't have a voltmeter at hand, but I checked that time the cables and none of them were hot. Your amps calculation seem correct:

So I am not drawing more current than I think. This leaves us with bad batteries? Any other possible reasons?

 

[Ampster]

Thats is a very energy sucking method to produce heat

As others have posted earlier it actually it is three to five times more efficient to transfer heat with a heat pump than to produce it by using electricity to heat a wire or heating element.

 

[Wibla]

So I am not drawing more current than I think. This leaves us with bad batteries? Any other possible reasons?

Can you post graphs of voltage and current to/from the battery over the last 72 or so hours?

 

[Quattrohead]

should answer all the doubters about newer heat pumps

 

[micmak]

Can you post graphs of voltage and current to/from the battery over the last 72 or so hours?

I purged my home assistant database earlier that day so I have the below only:



I can get some data from the wifi datalogger that comes with the inverter (updates every 3 minutes or so when internet is working, so I wouldn't rely much on it, but it covers 25th all day and 26th til 6AM, issue happened between 3 and 5am on 26th):

 

[Wibla]

That does look like batteries with diminished capacity, for some reason.

 

[micmak]

I called the company that installed the system, they passed by to check on the batteries and said something about the acid, and they increased the charging voltage and asked us to keep the generator/grid charging the batteries for a day. Looking at the graphs after they left, it's still charging at 54.9V.

I think they were talking about acid stratification (took a while to get to this given the weird translations in the local language). But shouldn't it be 57.6V for this? Solar panels will provide that kind of voltage, but not the grid (not sure why, maybe that's how the inverter works).

 

[HRTKD]

Generically speaking, every charge source should provide the correct voltage (charge profile). It is doesn't, then you either have a configuration problem or the wrong equipment.

According to the PDF linked below, a fully charged 48 volt battery bank should be 64.8 volts. I don't work with 48 volts, but that seems high. Either way, you're not even close.

https://robusteenergy.com/wp-content/uploads/2021/11/ROBUSTE-CATALOGUE.pdf

 

[sunshine_eggo]

Generically speaking, every charge source should provide the correct voltage (charge profile). It is doesn't, then you either have a configuration problem or the wrong equipment.

According to the PDF linked below, a fully charged 48 volt battery bank should be 64.8 volts. I don't work with 48 volts, but that seems high. Either way, you're not even close.

https://robusteenergy.com/wp-content/uploads/2021/11/ROBUSTE-CATALOGUE.pdf

Wow. That is absurdly high. Most 48V lead-acid are between 56 and 60V absorption. Those voltages are equalization level.

I have no experience with tubular type. Maybe that's a difference. Personally, I would be concerned as some inverters cut off below the fully charged levels.

 

[micmak]

Actually those voltages are configurable in the inverter, Bulk Charging Voltage (was 57.5, increased yesterday to 58.4) and Floating charging voltage (55V). Solar can charge up to 80A and grid up to 40A. I see that solar does hit the Bulk CV but the grid is always limited to 55V. There is an option in the inverter for "battery stop charging voltage when grid is available, which is set to 55V", maybe that needs to match solar...

I can't find much information on tubular as well. The country plunged into darkness and hyperinflation mode so solar became extremely popular to whoever could afford it. And out of nowhere, the market got flooded with these batteries, mostly made in India. I wanted to go initially for lithium but couldn't find them in the local market. I guess by the time the current batteries need replacement, the market would have saturated.