See for more themes. | Q25: Some final remarks Finally some find the off grid 240V system is too costly. Getting some $20,000 is hard. Cheaper than commerical quotes, do it yourself. You do not need 500amp/hr Ni-Fe batteries. You need only power for night use. 200 Amp/hr is fine for one fridge, etc. So purchase 200amp/hr batteries, at $4,000. A second set allows you to switch over to them, if you have run the other dead flat on a cloudy or rainy day, or at the wrong time of the day. Do both auto charging and manual charging. One MPPT 100|30 controller for when you go away on holidays, auto is nice :) Add a fuse breaker to limit power in and out to a safe load, say 100Amps. This protects battery from overload charges. Purchase three Victron 1200VA inverters for $1,800. Purchase as many second hand 200W solar panels as you can, say 100 of them, at $50 each, will cost you $5,000, and provide at least 3 amps each, you switch them on until the amps reaches 80Amps. You need to source proper voltage panels, this is hard but not impossible. And all solar panels must be of similar manufacture, otherwise one amp fights another amp in parallel. Since you are manually adding distilled water weekly it's no labour issue to a grassroot person, maintaining the system. This way you get a 240V fridge, as large as you like, a 240 Water pump (700W) and another invertor to run a bore perhaps. Plus you have over in theory 300 amps available on a rainy day, actually they come in at 50amps, but who cares, no more generators, and the second hand solar panels will continue for another 20 years just fine, and the Ni-Fe batteries for 100 years. The total cost is under $9,000. Now it seems OK even to a poor person. Beware of voltage depression or Battery Memory . I suggest once a week you flatten the N-Fe bank to 20 Volts and every day use all the storage you have, so if it's 300 amp/hr make sure you consume every drop every day, and things will be fine. Or with a second battery set, run the first dead flat every day, and charge them ONLY when the cel voltage reaches 1.05V Or if you can't afford a second bank, make sure you run the bank fully flat before sunshine time, and you switch the solar panels on. DUring the day, the batteries are not used, the solar panels are used to run big loads, you only need the bank for the nght loads, from 4pm until 8 am. Make sure you can run the bank dead flat before 8 am, every day. Another saving is you do not need fuse breaker or safety switches, you cannot protect the Invertors anyhow, so do not bother wasting your money. You also don't need a BMV monitor either. And if you get stuck and want advice, just email me. Just remember I am no expert, but have wasted thousands too on learning the experience of the basics. I have just realized that my Solar Trina Solar panels are rated at a working voltage of 34.2 Volts. I am thinking of removing the wiring configuration, and building my entire system into a manual charging system, and keep Victron MPPT solar controllers as an auto back up. I can regulate the charging better myself, for summer less panels, for winter more panels, for cloudy days mlore panels and for cloudless days less panels, a solar controller cannot do all this, as a human can. And all the overcharging will do is remove too much water, but I am adding distilled water every week still anyway, so who cares? Switch on as many strings of current as required to get the batteries for bulk, from 60 Amps to 90 Amps. Come back in a few hours time and get my voltage per cell, if is 1.5 V, than switch some strings off. Come back in more hours time and switch off the strings of current and leave only a trickle running, like the auto Victron controllers. When I am using a big load, such as bore pump, water pump and distillation of water, all this is 60amps, I switch on 60 amps of current, plus another 60 amps and see if the fuse breaker holds. It should, 60 amps going out and 60 amps left passing through the 100 Amp fuse breaker into the batteries. Things to try. What does all this do? It reduces your costing by $3,500. No need for lots of solar controllers. You might keep one MPPT for when you go away on holidays. Hopefully readers will understand how Ni-Fe behaves and how different it is to Lead Acid or Lithium batteries. You also potentially save money. Nickel Iron likes to be worked hard and flogged hard. And every day you must the bank dead flat, to 1.05 Volt per cell, or 20 Volts on a 24 V system.
Or you purchase the solar panels secondhand, but all strings need to add to 30 to 40 amps at 34 Volts. You switch on manually as many strings as you need, summer, winter, cloudy or rainy. You cannot switch them all on, as it goes past the limit of the 100 Amp fuse between the battery bank and your system. But you can switch current greater than 100 amps if you pulling amps away at the same time, ie using a Inverter to run something. The 100 Amp fuse protects your battery bank, so no worries. Manually overcharging your batteries every day, no worries either. Cost for this system? Less than $12,000 If you can source 200 W 3 Amp to what used to be 5 amps when new solar panels at say $20 each A hundred of them gives 300 amps in theory and cost $2000, so you are better off. Just make each is 34V and gives the same amps (test them individually). If you have any questions, email me. This is a grass roots installation, for grass root people. If you seek bigger systems, get a commerical person to expertly do it for you. I expect my current system to run fine for the next ten years. I will design it as the above diagram. Maybe new electrolyte will restore things :) Shalom
Recent update on the water loss from my battery cells. On advice from David Bartlett of Iron Core Batteries, Victoria, I was able to enter each device and going into battery/ than/ expert mode, you see the tail current. David advised me to set tail current to 0.0V, which Victron see this as disabled. Now hopefully the Victron controllers will not go to bulk from 4pm and allow the battery cells at 29 Volts to remain that way. No more bulk and over charging the cells from 4pm. We will see the results soon. Update 2 September 2023.
update April 5 2025. The NiFe battery bank cannot run a fridge and freezer for more than 2 hours into the night, so I suspect it's no longer able to deliver 150 Amp hr any more. We have had our batteries for 5 years, and I feel it's time to lay the technology down for now.
So I purchased on the advice of family 400 amp hr LiFePO4 battery technology. Sadly I don't like this technology either, but my friends were adamant. I said Litium Iron Phosphate also have memory issues. They say ridiculous. The company who sold them say, they come with their own battery management system. So need no water top up ( a big bonus) cannot be overcharged or undercharged. Great I think, but what about memory issues? They say they will last just six years. Not much good with that. I suspect any battery chemistry does not like little discharing and little charging, like solar does. One has to discharge the battery once charged to its optimum discharge, and than charge to it's full charge limit. Once charged the battery bank MUST be removed from solar panel charging until it is optimumly discharged. Once discharged, you switch on the charging. However timing these cycles to match solar sunshine is near impossible. Even with two banks it will be hard to achieve. So my friends are my experiences with memory issues with chemical batteries. Lead acid does not have memory issues as much because Pb Acid does not like to be deeply discharged in the first place so a small loss of due to memory is barely noticed until the Lead sulphates. Shalom Another update: April 25 2025
I am about to install my new Lithium Iron Phosphate cells, and have been told they also have memory issues. So you have to run the battery bank (in my case one cell in parallel to another cell) [ 24 V at 400 Amp hr] down to 21V every day, consuming all 400 Amps all the time. This is the same idea as Nickel Iron, which I didn't know about before.
Five inverters on right, mostly 900 W, or 37 Amp max Two LiFePO4 batteries with Smart Battery Monitors in parallel. Four solar controllers wired to supply maximum 30 Amps each, 120 Amp for the battery bank. The system for Victron is still on bulk, so I suspect the BMS is not using the 120 A available, but ignoring what it needs and I am unable to measure this? Bother. You could save 1000 for there is no need for a Victron Solar Controller, except to convert all current and voltage to the right levels for the BMS system on top of your NiFePO4 battery cells. Now all we need to do is avoid memory issues and use all the current before sunrise.
Notice the preset Victron settings for smart LiFePO4. Installed on Anzac Day, God bless our freedom thanks to the soliders who died to make us free. April 25 2025. Hopefully my shunt battery monitor might confirm I have 400 AH, but I doubt this. The first night before sunrise, the voltage was only 26.4V telling me with a fridge, freezer, and two fans (2 LED light for a while and a 24V shower pump for a while) should have used more than 120 Amp H. I used to consume 200 Amp Hr on this use? Wow So I disconnected a parallel cell, reducing the SOC to 200 Amp Hr. Currently the voltage is 26.2, still too much in this single bank bank. This is still 50%, and thus 70AmpH to go? The water pump is using 27Amps to run, it usually takes 23 Amps, so the amps rises as the voltage falls. I have connected the 240V 700Watt garden water pump to drain the battery down to 30% SOC. Still got some way to go and its 6:30 am Still 26.1 V only 50% bother. Need more load? So I plugged into another water pump, now we have just under 50Amps coming out, the voltage drops to 25.9V nearly there but sadly light hitting the panels at 7.00am, so I switch off all the loads, the voltage rises to 26.2. This is still 50% to go, but on load nearly there, at 25.9V. So it's hard to know when your LiFePO4 chemistry is flat ready for the next day of sunshine. The SOC curve is almost too flat, hardly any difference between 70% SOC and 30% SOC. This is a bother? It is 9:00am and I suspect the 200AmpH bank is already fully charged as the shunt meter says 200 Amps in. The vitron solar contollers are all BULK on 22A, 22A, 26A and 4 A, because the last one faces due west. So I suspect the BMS on the LiFePO4 is already switched off charging and the Victron controllers are still in bulk. I have no way to testing the wires inside the bank, under the BMS system. See what happens on another day. It's May 3 2025, since installing the battery on 25 April 2025, just 8 days later, the bank is at 21 Volts and completely dead, the BMS has shut the entire bank off - no lights, no power - nothing ! Listening to a video, one must make sure the batteries never get too hot or too cold. One must make sure the batteries are fully charged and fully discharged once a day on Solar systems. And once cannot left a bank in storage, as with self discharge, it becomes permanently dead. I discovered the problem was I left the garden pump (25 amps per hour) connected to the Inverter and so it ran the bank flat. So I am purchasing light senor relays to switch off bigger loads when the solar light fails. My family relatives all say LiFePO4 does not have memory issues, so I was told to hook up the 400 Amp hr battery and use it a little, and often. Forget about memory issues. Arrh well, I am happy to do a 7,000 dollar experiment and prove them wrong. Now to see how long the LiFePO4 lasts on a 400 amp hr system with once daily charging per dayand at night we use around 8 hours per hour, or about 170Amp hr for the entire night. With my energy shunt meter, I can graph the loss of State of Charge over the next few years, to prove them wrong, that LiFePO4 does have memory issues. So much for listening to family advice. Abstract. Lithium iron phosphate batteries are widely used in various fields and have long been considered to have no memory effect. It is not until recent years that the memory effect of lithium iron phosphate batteries has been discovered and studied. For lithium iron phosphate batteries, an incomplete charge and discharge will cause the battery to produce a memory effect, causing a jump of voltage during the memory-releasing cycle. And this memory effect can be erased after the battery is fully charged and discharged. The memory effect of lithium iron phosphate cathode material is affected by active particles, relaxation time. The paper refers to charging electric cars rapidly often to only 70% and not 100% as Solar charging does. Perhaps memory is not an issue with once a day 100% charging using solar controllers. They have also previously been widely believed to exhibit no memory effect. That's how experts call a deviation in the working voltage of the battery, caused by incomplete charging or discharging, that can lead to only part of the stored energy being available and an inability to determine the charge level of the battery reliably. Another discussion : link to memory issues When it comes to modern lithium ion batteries, the memory effect is not as common as it is in older battery technologies. However, this doesn’t mean that lithium ion batteries are completely free from it. Under specific conditions, lithium ion batteries, especially those using lithium iron phosphate (LiFePO4) chemistry, can still experience something similar to the battery memory effect. The article says to restore any mild memory issues, every now and than, do a proper discharge and charge of the system. Hmm? Might some big electric loads, like a airconditioner? So I will test the system and keep readers informed. I will let you know how it all turns out. Shalom
|
Created by Rob Thompson. Hosted since 10/01/2012. Visitors HOSTED by Prologic, my Son. A thin website.
