Off-grid installation diagram, batteries and other equipment.

hostii wrote:

powgreg you are wrong, the battery will not discharge to the end (10.5V) because the power supply has a higher voltage and the devices draw power first from the generator (device) which has a higher voltage in this case it will be the power supply, you can't get more out of the battery, unless that we will draw more power from the power supply than its nominal power. Then the battery will discharge to zero and the power supply will turn off. You have a similar situation with on-grid, the inverter increases the voltage and first the energy in the house is consumed from PV and when there is no voltage, it drops and is selected from the network where the network is our power supply in this case.

We have a parallel connection so as not to complicate the two sources (PV is omitted), the converter will draw energy as you rightly noticed from the source with a higher voltage, i.e. first only from the battery itself and as it is discharging, when the voltage is equal to the power supply, from both sources until the battery is completely discharged, i.e. SEM its will be equal to the voltage of the power supply.
To sum up, the current from the battery will decrease but it will not stop flowing when the voltage on it drops to 11.5v only when the emf, i.e. the voltage on the open terminals reaches 11.5v, it will stop flowing because all the current will be taken over by the power supply.
Of course, the voltage at the battery terminals will not drop below 11.5V, but this does not prevent it from being completely discharged because it is assumed that the 12V EMF of an unloaded battery means 100% discharge.

Indeed, too long operation of the battery with a parallel power supply can be deadly for it. When I was doing my experiments, the power supply had about 12V because it was too weak to "carry" the load without battery support. I didn't think about such details because I didn't notice them, the battery was always charged. In my opinion, 12.2V is a bit too much to pay off in parallel with the power supply, so I decided to look for the lowest possible voltage and found this:

https://techtron.pl/blog/jak-prawidlowo-ekploatowac-akumulatory-zelowe-i-agm/

A lot of detailed information about the operation of batteries, of which the most important in this case are the voltages of 11.8V and 10.8V. If the battery is discharged with a high current and then lies uncharged for a long time, the limit is 11.8V. But if we discharge it calmly, for 20 hours, it can go down to 10.8V. 20h is enough time for PV to start giving voltage even in winter charging the battery so that it will not be discharged. So if I use LEDs and other trinkets as a load, I can safely charge the battery up to 11V and immediately charge it. In the case of larger ones, it is better not to go down to those listed in the table because they will be constant for too long, the limit is 11.8V. In my opinion, the parallel operation of the battery and the power supply with a heavy load can be treated as 20 hours of discharging and even 10.8V will not kill the battery. I guess I'm good at it?

gaz4 wrote:

A lot of detailed information about the operation of batteries, of which the most important in this case are the voltages of 11.8V and 10.8V. If the battery is discharged with a high current and then lies uncharged for a long time, the limit is 11.8V. But if we discharge it calmly, for 20 hours, it can go down to 10.8V. 20h is enough time for PV to start giving voltage even in winter charging the battery so that it will not be discharged. So if I use LEDs and other trinkets as a load, I can safely charge the battery up to 11V and immediately charge it. In the case of larger ones, it is better not to go down to those listed in the table because they will be constant for too long, the limit is 11.8V. In my opinion, the parallel operation of the battery and the power supply with a heavy load can be treated as 20 hours of discharging and even 10.8V will not kill the battery. I guess I'm good at it?

It is exactly the opposite with high current you can go low with the voltage because there is a voltage drop on the internal resistance of the battery so when you remove the load the voltage will increase and with a very small current going down to zero the limit of 2v on the cells is considered to be about 100% discharge, the value of this voltage changes by a few tenths of a volt, depending on the manufacturer, AGM gel technology, etc.

Please note that the degree of charge is determined by the SEM voltage and the disconnected battery for 1-2 hours.

I know from my own experience that basing an off-grid installation on a simple algorithm of measuring the battery voltage alone leads to its rapid wear...

I understand, but that's not what I meant, the battery will never reach SEM during the day because it will still be slightly loaded. I will give an example based on my own case. For 24 hours a day, I will have the power supply of the converter and the pump controller turned on. At the moment I do not remember what current they consume, for simplicity I will assume 0.1C, i.e. the discharge of the battery without external support will take place after 10 hours. But I connect the power supply in parallel, so the load on the battery after several hours without recharging will drop to, say, 0.05C. So on the terminals we will never have emf but voltage at minimum load. In my opinion, if after 20 hours of exemplary discharging, charging from PV takes place, even 10.8V should not kill the battery, because such work will feel like a slow discharge of 0.05C.

I'm not going to go down to 10.8V if only because more or less at this level my converter starts beeping warning against discharge. I consider the level of 11.5V at the above-mentioned continuous load, i.e. the controller + own converter consumption, to be optimal. Unfortunately, my experience with the battery is too small to clearly state that such work will not kill it. Earlier experiments with parallel power supply were conducted on an old car battery and a higher voltage, for several months of such work I did not notice a decrease in capacity. But it was in the summer, he had time to recharge in the evening (the main load was the collector pump) so it's hard to say what will happen in the winter when I hold the battery for a few hours under a light load at 11.5V and start charging when the PVs start working.

This mode is extremely interesting for me because I have a tariff where electricity during the day costs over 70 groszy, but at night (22-7 in the morning) below 30 groszy. Even the losses on the power supply and the converter with such a price difference are "neutral". Anyway, in winter it's hard to talk about losses because I have electric heating and the energy "wasted" by the above-mentioned devices will stay at home anyway - too many pluses not to take this solution into account. Five times the door, a pump working on PV with a parallel power supply will save about PLN 50 per year, the power supply itself comes from the "drawer" but even as a new one it was not more expensive than 5 tenner. So after a year it depreciates, after about 6 years the purchase of the inverter pays off, and after another 4-5 years I earn 100W PV ... No risk of boiling water in the event of a power outage is just a bonus

gaz4 wrote:

I understand, but that's not what I meant, the battery will never reach SEM during the day because it will still be slightly loaded. I will give an example based on my own case. For 24 hours a day, I will have the power supply of the converter and the pump controller turned on. At the moment I do not remember what current they consume, for simplicity I will assume 0.1C, i.e. the discharge of the battery without external support will take place after 10 hours. But I connect the power supply in parallel, so the load on the battery after several hours without recharging will drop to, say, 0.05C. So on the terminals we will never have emf but voltage at minimum load. In my opinion, if after 20 hours of exemplary discharging, charging from PV takes place, even 10.8V should not kill the battery, because such work will feel like a slow discharge of 0.05C.

The crux of the problem is that in this parallel battery + 11.5V power supply system, the energy is first taken from the battery and especially for small currents drawn because they do not cause a large voltage drop on the battery so that part of the energy is taken from the power supply.
So in most cases, such a system will discharge the battery to zero or almost to zero, and whether and after what time it will happen depends on the parameters of the given system as well as the "sun" just a few worse days, etc.

gaz4 wrote:

Earlier experiments with parallel power supply were conducted on an old car battery and a higher voltage, for several months of such work I did not notice a decrease in capacity. But it was in the summer, he had time to recharge in the evening (the main load was the collector pump) so it's hard to say what will happen in the winter when I hold the battery for a few hours under a light load at 11.5V and start charging when the PVs start working.

In this system you had three plus factors: higher voltage, summer and a receiver that did not work without the sun, so it had the right to work well.

gaz4 wrote:

... so it's hard to say what will happen in winter when I hold the battery for a few hours under a light load at 11.5V and start charging when the PVs start working.

If it's not too deeply discharged, nothing will happen...
I do not know what your system parameters are, battery capacity, max current, etc., but I would try to increase the voltage, the easiest way is to disconnect the power supply, let's say the battery is 50-70% charged, turn on all loads (those that most often work simultaneously) measure the voltage and if it does not come out too low, then this slightly lower on the power supply.
It is known that higher voltage means more frequent "switching on the power supply", and much above 12v means the possibility of unnecessary charging of the battery from the mains (you can add a separating diode).

I don't have a power supply at home because there were quite large standstill losses and double processing, only the system below 11.9V counts the set time and "dumps" the receivers to the network, if the voltage "jumps" during the countdown time, the counter resets and does not switch to the network.

The battery capacity is not a constant value because I can also use screwdrivers and trimmers. Those for lower voltages (14.4V) will constantly work with acid, and those for higher ones (18V) will be charged separately (also with PV) and inserted, if necessary, into a stand equipped with a high-power DC-DC converter that lowers the voltage to 13V. So, apart from the power supply, acid battery and NiCd / NiMH battery connected permanently, I have a few Ah of spare capacity which I can support the system at any time. I decided on the battery from screwdrivers for a simple reason: during the year I will use them maybe 30x, so they will fall apart faster from old age than they will perform 1000 cycles. 200 cycles each winter will shorten their life to, say, 5 years, but in normal use, after a maximum of 20 years, they would still be thrown away. The ones I bought with the screwdriver died after about 10 years, i.e. no more than 300 cycles.

To sum up: the lower the voltage, the greater the risk of battery damage in the absence of the sun. The safest would be suggested by @powgreg 12.2V, which protects the battery from discharging below 50%, but this would mean too much consumption from the network during the day and the whole fun will become financially unprofitable. Since at the moment I am not able to determine how everything will work in short days, I will try to minimize the risk of battery sulphation. I will make a system in which the power supply loaded with the converter and the controller (with the pump turned off) will have 11.8V. This is the safety limit for gel batteries, there should be several hours a day when the battery is recharged, so they should never discharge to zero. If I notice any disturbing symptoms, e.g. increasing power consumption from the power supply, I will increase the voltage to exactly 12V, which for me is the upper limit of profitability. If this does not help, the purchase of a large (>100 Ah) battery or the abandonment of the idea of powering the pump with PV. In this case, I would bet on the 2nd option, it is not worth investing too much for the pump itself, because with PLN 50 of annual profit, I can't wait for the return

gaz4 wrote:

If this does not help, the purchase of a large (>100 Ah) battery or the abandonment of the idea of powering the pump with PV. In this case, I would bet on the 2nd option, it is not worth investing too much for the pump itself, because with PLN 50 of annual profit, I can't wait for the return

I would not touch the solar pump because it is your best receiver because it works when there is sun and then there should be no shortage of energy and the battery should only work as a buffer. And the power supply should be switched on from the mains side when the battery voltage drops too much, because on standby 1W / year is about PLN 6.5.

You also need to be careful with the parallel discharge of old used acid batteries because they are easy to damage at discharge voltages even slightly lower than, for example, 12V, and the higher the installation voltage of 24v 48v, etc., the harder it is to see that the single weakest 2V cell has been excessively discharged.
I have often observed such situations
5cell after 2.05vi and one cell 2.03V on the whole battery 12.28V after an hour

5cells at 2.03V and one 2.00V cell around 12.15V and later it was even more interesting

5cells at 2.08V and one 1.6V cell across the 12V

If it was one battery, the voltage would immediately drop on it and the protection system would disconnect the converter and so the current from the other parallel batteries would support the power supply and everything seemed to work ok.
Therefore, the voltage across the entire battery connected to the system says little about its condition...

I don't have a collector anymore, it's a pump for a coil in the fireplace. This one will work at the worst time when the sun is like medicine and worse, very short days. I have a cheap tariff for half a day, which is used for heating anyway, so during this time the losses on standby de facto do not exist. Because I smoke in the fireplace in the afternoon when the 1st tariff >70 gr/kWh applies, the time when the power from the power supply in idle state will be wasted is reduced to a maximum of 5h/day in the morning. In this mode of operation, an additional attachment does not make sense, as in the 24-hour G11 tariff - apart from savings, it can remove most of the disadvantages of a solution with a parallel power supply.

I encountered the problem of dead cells when the old acid gave up the ghost. And you are 100% right, several parallel batteries will make it very difficult to control. Fortunately, there are Chinese with their electronic trinkets sold for half free, so I found a simple and cheap solution In my charger based on DC-DC converters, I used a voltmeter with an integrated ammeter and thermometer, which is also useful. I connected the whole thing in such a way that the batteries from the screwdrivers are charged bypassing the ammeter, which only indicates the charging current of the acid battery. This is well justified because the way of connecting NiCd prevents them from discharging

gaz4 wrote:

I don't have a collector anymore, it's a pump for a coil in the fireplace. This one will work at the worst time when the sun is like medicine and worse, very short days.

In such a system, it is worth carefully recalculating the economy of such a solution, because if the greater part of the current supplying the pump is selected from the mains, then after "multiplying" the efficiency of the double processing of the power supply-converter, it may turn out that we gain little and use the batteries heavily.

gaz4 wrote:

I don't have a collector anymore, it's a pump for a coil in the fireplace. This one will work at the worst time when the sun is like medicine and worse, very short days.
I encountered the problem of dead cells when the old acid gave up the ghost. And you are 100% right, several parallel batteries will make it very difficult to control. Fortunately, there are Chinese with their electronic trinkets sold for half free, so I found a simple and cheap solution In my charger based on DC-DC converters, I used a voltmeter with an integrated ammeter and thermometer, which is also useful. I connected the whole thing in such a way that the batteries from the screwdrivers are charged bypassing the ammeter, which only indicates the charging current of the acid battery. This is well justified because the way of connecting NiCd prevents them from discharging

powgreg wrote:

gaz4 wrote:

I don't have a collector anymore, it's a pump for a coil in the fireplace. This one will work at the worst time when the sun is like medicine and worse, very short days.

In such a system, it is worth carefully recalculating the economy of such a solution, because if the greater part of the current supplying the pump is selected from the mains, then after "multiplying" the efficiency of the double processing of the power supply-converter, it may turn out that we gain little and use the batteries heavily.

I am aware of this risk, so I am trying to think everything through before the season starts. There was no problem with the collectors because the power supply was needed only in the event of a sudden increase in cloudiness. At that time I had an old battery + low PV power and before the collector had time to cool down to turn off the pump, the inverter often beeped signaling a voltage drop on the battery. The author of the topic is also considering off-grid powering the collector pump, but at current PV prices, I consider it unprofitable. It would be better to go for DHW on PV, with 1 kW per person you can have a nice piece of power that will work all year round. With 2 kW on the roof in winter, I have a large surplus of power (DHW is from the fireplace), which I can fully direct to power the pump. In the worst case, the PV will be 100W, of which the pump + inverter will consume half, and the rest will go to the battery. It's a pity that working hours are a bit different from the Sun. The pump must work after sunset, i.e. it will load the battery in 100%, which is why I am considering a parallel power supply. Until 10 pm, most of the energy should be provided by the battery, and after this hour the electric heating turns on, so the concept of "loss" does not exist. In fact, everything will run on PV despite the power being drawn from the grid for half a day Worse than the battery can't handle it, its faster consumption + the cost of additional electricity used during the day and the savings may result in a loss. But I'm good-hearted, I'll manage and I won't lose

@hostii Never mind that it's archived, but it's still useful - regards S. Ludwik