 Notice. New forum software under development. It's going to miss a few functions and look a bit ugly for a while, but I'm working on it full time now as the old forum was too unstable. Couple days, all good. If you notice any issues, please contact me. |
Forum Index : Solar : LA battery charging
| Author | Message | ||||
| Alastair Senior Member  Joined: 03/04/2017 Location: AustraliaPosts: 161 |
I am just starting a project to build an irrigation system to water our garden whilst we are away traveling. It will be controlled by a micromite and powered by a 12v 7AH security system style battery. I plan to charge it from a 20W solar panel. The numbers work out ok. I will charge the battery to 14.0v once per day and then float it at 13.8 - sun permitting. I will use the MM to control the charge by PWM. My question (finally) is what is the best regime for the pulsing? I will adjust the pwm duty cycle based on the voltage but what is best low frequency longer on period or higher frequency shorter. I am familiar with linear charging, Mppt chargers but have never look seriously at what pwn chargers typically use in terms of frequency. Cheers, Alastair |
||||
| Alastair Senior Member Joined: 03/04/2017 Location: AustraliaPosts: 161 |
I have been doing more searching and thinking. There are commercial & amateur units that have mentioned frequencies from 25Hz to 5KHz. I have no idea how LA battery chemistry responds to frequency. Any insights? I am thinking that actually what I should do for my simple needs is more on/off charging based on voltage rather than pwm. What is the advantage of pwm in this situation? I can see perhaps when in float mode a gentle pwm may be better than say charging to 13.8 and then waiting to drop to 13.75 to charge up again. The purists suggest that all voltage should be measured off-load to get a better SOC indication. My feeling is that I will not bother as when under load I want max charge unless the voltage rises above 13.8 Cheers, Alastair |
||||
| Warpspeed Guru  Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
The simplest way would be to not bother with PWM, but just switch the battery straight across the solar panel until the voltage reaches 14.0v then disconnect the battery. Allow the battery to sit, until the voltage falls back to about 12.6v. A fully charged battery in good condition should be able to attain and hold 12.6v almost continuously. If the battery is flat, it will struggle slowly up to 14.0v, then when the charger disconnects the voltage will fall VERY rapidly back to 12.6v. This gives a long charging time and a very short off time. Its ultra slow motion PWM. As it gains more charge, the on times become shorter and the off times longer. This goes against much of the established theory, but it doers in fact work very well. You just need to make very sure that the solar panel and the maximum rate of charge is not too much for a fully sealed lead acid battery. Gassing must be avoided at all costs, or the battery will quickly dry out and be ruined. The battery should have some specified maximum charging rate, and the solar panel sized to that. If your charging current is still a bit high, one thing you might like to try is connecting an 18 watt car turn indicator bulb in series with the battery and solar panel. With 12v across the brilliantly lit bulb the filament resistance should be about 8 ohms. But the cold resistance of the filament will be about a tenth of that, maybe 0.8 ohms or less, a multimeter will tell you. How this works, is that if solar conditions are very poor, an extra 0.8 ohms makes little difference to the charging current. But with a clear blue sky, the filament might run just red hot, and reduce the charging current slightly. Its a poor man's current regulator. Its also a handy charging indicator. By trying various wattage bulbs, (or combinations of bulbs) by trial and error its possible to make a system that is not only very cheap and simple, its also very reliable and can work extremely well over long periods, as there is really not much that can go wrong. I have a system like this keeping the battery in a very rarely driven vehicle fully charged. The charge and discharge cycles are much more healthy for the battery than a constant float voltage, especially for a fully sealed lead acid battery. My system runs off a 15v dc plug pack, but it should work just as well from a small solar panel. This system has been in operation for just over two years, and my vehicle starts instantly with an obviously still very potent battery. Cheers, Tony. |
||||
| Warpspeed Guru Joined: 09/08/2007 Location: AustraliaPosts: 4406 |
Yes indeed. You posted that while I was still writing my reply. Cheers, Tony. |
||||
Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
PWM works fine, lots of commercial PWM controllers, we are also doing with an Arduino. Just set it to maintain absorb voltage for 2 hours, I have coded to stop counting time when the voltage drops due to clouds etc. Then drop back to float voltage, then when voltage drops back to near 12 volts reset ready for the next absorb cycle. You may find this thread helpful. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| hotwater Senior Member Joined: 29/08/2017 Location: United StatesPosts: 120 |
For such a small system I would consider using one of those little 2596 type buck converter boards and operate a pseudo MPPT with fixed panel voltage. The on/off pin can be lifted from the board ground and pulsed on and off from a micro if something elaborate is desired. I have a simple garage system with a SLA for lights. It was an old battery I had recovered from recycling and it has been running for years. Panel voltage is set at about 17V and output of regulator at 13.5 volts. A single transistor sinks the on/off pin with a pullup resistor. A 12V zener does most of the voltage drop to a voltage divider pot on the transistor base. Just have a 1,000uf across the panel for charge storage. |
||||
| Alastair Senior Member Joined: 03/04/2017 Location: AustraliaPosts: 161 |
Thank you everybody for your suggestions. Many ways to tackle this. Yesterday I tried numerous ideas on the test bench in between domestic tasks. I think I have settled on 'slow' pwm as my approach. Since I have a microcontroller in the box to do the watering switching I could not resist doing something else with it. The panel will be full on up to 14.0v switched by a mosfet controlled by the micro. PWM will kick in at something like 20Hz and hold the voltage at 13.8 during the day when sometimes the solenoid valves will be switched on and load the system. The system will then obviously be off overnight. I have decided to use solenoids that need a hold open current rather than latching ones. My reason is that with the latching units there was actually no way to know if the valves had opened without putting in flow sensing which is over the top for this. I have some of those cheap Hall effect current sensors. I will use one and assume that if the right current flows, then the valve has operated. If the current is wrong then we have a fault and a warning is shown. My logic is based on the system being needed because we will be away quite a bit over the next few years and I want it to be robust. If there is a fault then it will be obvious and a neighbour will intervene hopefully. The idea of modifying a converter board had not occurred to me. I have a few sitting in the cupboard too. Hmmm perhaps have a little play today as I don't watch football and others will be occupied with their ball shape of choice!! Will post the final result when it happens. Have to get the valves and other bits this week so I have some time to play. cheers Cheers, Alastair |
||||
| Madness Guru Joined: 08/10/2011 Location: AustraliaPosts: 2498 |
Have you considered measuring soil moisture and having it connected to the internet? You could have it send 1 or 2 emails a day with the data. There are only 10 types of people in the world: those who understand binary, and those who don't. |
||||
| Alastair Senior Member Joined: 03/04/2017 Location: AustraliaPosts: 161 |
Pls do not keep suggesting additions - I have enough trouble constraining projects and keeping them within bounds  Actually yes I have thought about measuring the soil moisture and overriding the timer if it has rained. At the moment I have to focus on getting the basic system going as we will be away all december and I need to have the main part working well in advance so I can check that it is reliable. Winter has been dry and the forecast for summer is hot & dry. As to having the data on the internet - I am not keen. Ignorance may be better. Cheers, Alastair |
||||
 Print this page |
