Automatic Battery Switching.

Discussion in 'The Projects Forum' started by ron roberts, May 7, 2008.

  1. ron roberts

    Thread Starter Active Member

    May 7, 2008
    49
    0
    Hi everyone you may have seen this request on many forums from me, over the last 2 years I have been trying to monitor and switchover 2 12volt batteries.

    I have a motor that runs on 12 volt lead-acid battery and needs about 2 amps but I want to rate the circuit about 4 amps so has not to over load things. Now if the voltage drops below 10.5 volts it just comes to a full stop but the motor at this point is still drawing some current that completely flattens the battery if left unattended. The motor will free run at this voltage but not under load so I need to restore the 12-volt supply. The motor runs a small pump that just circulates water for my pond. Now I don’t want the pump to stop at all but if it does I want to be able to start it ASAP. So that means 2 batteries one in use and one on charge, by solar panel, But the sun doesn’t like to play ball some days this has happened many times.
    Yes I could use both batteries in parallel and charge them at the same time. But if the sun don’t play ball I have both batteries under 12 volts and no motor running. Now that seems to happen a lot when the batteries are in parallel, I know that should run just fine but it doesn’t.

    My system at the moment as 2 switches, switch 1 (primary switch) if I flick this one every day or so it seems to keep going but only with my input and for most of the time it’s fine on the switch. Switch 2 (secondary switch) puts the batteries in parallel regardless of the position of switch 1.

    Now lets say I go away for some time which happens, I cannot throw the switch so motor will stop. Yes I can put the batteries in parallel with the flick of a second switch just for such occasions and this will all most double the run time. Now in the ordinary world that would be fine. But my job means I get a call and whatever I’m doing stops and off I go regardless. When I get a call I don’t have the time to go and set things up. So I get the wife to go and flick the parallel switch, this will last about 4 days.
    Now if the wife uses the primary switch daily it seems to run forever.
    The location is in a remote-ish wooded area and I don’t want the wife do go to it unless absolutely necessary, all though keeping the pump running is a priority, but not as to put the wife at risk.

    Just to note I have run both systems independently without any switchers to see which would be the most suitable for my needs. And the results of these tests were to switch the batteries and not to parallel them.

    Tests carried out over 12 weeks, 6 weeks on changing the batteries over daily and the next 6 weeks with batteries in parallel. Now I know that the solar panels output will have a big deciding factor on this but I have been running the same tests over a 3-year period and it’s always the same result.

    Test 1.
    Week 1: running on primary switch, all runs ok.
    Week 2: running on primary switch, all runs ok.
    And so on for 4 weeks all ok when using the switchover. Switch 1.
    But maybe stopping at night, I ant stopping there all night to watch a pump, no thanks it’s just too spooky.

    Test 2.
    Week 1: running on secondary switch, mostly ok but pump starting to run slow by 4th day.
    Week 2 run on secondary switch pump running intermittently.
    And the same for the next 4 weeks when using the secondary parallel switch. Switch 2.

    So the next test was to use a timer circuit so every 8 hours or so it flips over the batteries this seemed to be OK, looked like it was sorted!
    NO, it would change regardless of the battery condition resulting in intermittent running of the pump.
    I played around with this one a lot, it does work to a degree.

    Next test light activated switch when it gets dark it flips over batteries, so the one that as been on charge all day now runs the pump all night, sun comes up and the other one that charged overnight now runs the pump all day.
    Doo! Solar panels don’t work to good in the dark. Well it seemed like a good Idea at the time.
    So as you may now start to understand my frustration with this maybe someone will help.

    So my worst case scenario, I left the system on one battery I get the call and off I go, maybe for weeks. Now maybe the solar panels will run things ok. More batteries would be in order or a different motor but I don’t want to do that the system works well but only with my involvement. Now for relays I have used all different manor of then but had nothing but problems, some of the problems were sticking on or not switching on, getting hot even one that had melted. I don’t want to go in to depth on this, I want to keep well away from them I just cannot express this enough so please no relays.
    Now I paid my local electronics company to build me something to do the job it cost me £150. I installed the unit it worked fantastic for weeks and completely solved my problem. So why am I looking for help well someone broke in to the shed and it got damaged the electronics company that built it sealed it with resin and I cannot get it repaired and they want another £150.
    I know properly need a voltage comparator and a current comparator, and reset flip-flop circuit.
    The outputs of the flip-flop drive will drive a Mosfets that connect the solar panel and disconnect the pump.

    But I don’t have a clue how to set the above out. Build it yes.
    Please help with full diagram even if it has relays. I’m so desperate to get this done.
    thanks Ron
     
  2. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Hello Ron,
    This certainly sounds do-able.
    But first, what country do you reside in? I ask this because what kinds of components that are available varies widely by country.
     
  3. ron roberts

    Thread Starter Active Member

    May 7, 2008
    49
    0
    Thanks for your reply i reside in the UK.
     
  4. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Ok, so you will probably be using someplace like Maplin for your parts source. I suggest that you at least put your country in your profile.

    I see they carry IRF540 power MOSFETS, which would work nicely (given appropriate drivers, etc.) They also carry IRFZ24's, which are considerably less expensive, and will also do the job.

    Storage batteries will last much longer if they are not discharged more than 30%. Running them completely flat will shorten their life drastically.

    I completely agree with your assertion that relays are not a good thing for your application; not just because the contacts will burn out either. While energized, the coil will be a constant drain - something you absolutely don't want.

    It seems to me that your pump system needs to be able to adapt to what the power remaining in the batteries are. This could be as simple as having the circuit adjust the ratio of "pump on time" to "pump off time". I suspect that your pond is supporting aquatic life, such as fish.

    What is the minimum duty cycle over a span of time that is required to sustain that life? For instance, for a 15-minute period of time, how many minutes must the pump be running in order to sufficiently oxygenate and filter the water?
     
    Last edited: May 7, 2008
  5. Digi Dave

    Member

    Apr 25, 2008
    26
    0
  6. Cornelius

    Member

    Mar 17, 2008
    19
    0
    You didn't mention the wattage of your existing panel? To break even on an average sunny day, you should have at least 100W of panels to run 2A 24hrs...

    I have no answer for you for the switch, but to me it seems that another good solution would be an addition of a small windturbine... :) It would depend on the windconditions in your area of course...

    But both these solutions costs more than a new £150,- switch... :(
     
  7. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Cornelius,
    That's sort of where I'm heading - trying to figure out what kind of power he actually needs vs what he actually has available, while attempting to maximize the life of the batteries - and perhaps a few other "whistles and bells" that didn't exist previously.

    Running his batteries down as far as they've been in the past really is not a good option. This is increasing his operating costs by accelerating the wear on the equipment; namely the batteries, which are expensive to replace. Allowing the life forms (if any) which are supported by the pond to die due to lack of oxygen or murky water is also not a good option.

    Wind turbines are quite expensive to put up. I'm thinking this is not a realistic option at this point, as the O.P. is balking at spending £150 for a replacement of a module that worked, and even a small wind turbine would cost considerably more than that.
     
  8. Cornelius

    Member

    Mar 17, 2008
    19
    0
    Seems i was a bit tired when i wrote this... ;)
    I also meant to suggest to double up his panels as an option...:rolleyes: But that will also cost a bit... (i estimate a 200W windgen at £550,-, and a 100W solarpanel at £500,-; 50W at £300,-)

    For as you say Sgt., the way he runs his batteries now, he's killing them; too many deep discharges, and the cells will sulfate beyond revitializing. To have batteries live a long life, you should consider them flat at a open voltage of 11.9V...

    If you're a bit handy in mechanics Ron, you could build yourself a small windgen; either from scratch, with magnets and coils, or get an old dc pm-motor to convert...
     
  9. ron roberts

    Thread Starter Active Member

    May 7, 2008
    49
    0
    Thanks Digi Dave I do have an account with CPC and RS and most of the large component distributors.
    Sorry to all that have answered I should have stated in my original posting about solar panel output.
    The solar panel is 120 watts

    SgtWookie
    I have now put my country in my profile thank you I never even considered the need for doing so until you pointed it out.
    Sourcing the components is no problem I have an account with CPC and RS and most of the large component distributors.
    Storage batteries and discharged more than 30%:
    To stop running them completely flat is why I want the circuit to prevent this happening. Not only shorten their life if they run to far down or completely flat, the solar panel seems to take a longtime to recharge them before they can be used especially when paralleled. I have even tried doubling the batteries up but they never get a full charge unless I disconnect the pump some times for days depending on the sunlight.
    Like I have mentioned I have 2 batteries one running the pump and one on charge and I just change them over by a switch. It all works fine I have tried all manner of batteries at the moment I have 2 45 amp hour. (Well that’s what I think they are but I just cannot remember if that was the total amps or just one battery but will look when I go down next time.) If I go larger with the batteries it gives the same effects as paralleling. I know it all depends on the sun, but the setup I have seems to be the best setup I can find. Large solar panel, small batteries. This way if the sun don’t play ball the solar panel still gives the batteries a good charge.
    So without being pushy I don’t really want to go in any other direction with this as I know that if I can build the correct circuit it will work perfectly.
    I haven’t stocked the pond yet I won’t do that until I can prove to myself everything is working. All I can say is I would like the pump to run has much has possible preferably all the time. I’m not allowed to use a wind turbine. I had a hell of a job with planning for the solar panel as that was refused. I got round it by incorporating it in to an existing roof that needed replacing. The roof was just the right orientation and pitch. I was lucky as there were no plans set down for the roof I just informed them that there used to be a skylight in the roof and that’s where I wanted to put my solar panel, and they agreed with me about the skylight with a shady grin.
     
  10. ron roberts

    Thread Starter Active Member

    May 7, 2008
    49
    0
    Well I hope I haven’t offended anyone with my last post only all seems to have stopped.

    thanks from Ron
     
  11. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Sorry Ron,
    Have had a whole slew of new folks on board. Got distracted. :rolleyes: ;)

    Seems to me that this project is simply begging for a microcontroller to monitor things; to ensure both batteries get topped up daily while keeping the pump running, at intervals if need be, during periods of inclement weather, or a battery cell failure. A microcontroller could be programmed to "sleep" for a period of time to save power, then "wake up" to perform system checks, switch things over, and possibly even report back statistics via a low-powered RF transmitter to a receiving module in your house; perhaps a USB device connected to your computer. It's just a concept at the moment.
     
  12. plasma

    New Member

    May 13, 2008
    1
    0
    I have only had a very little amount of experience in all of this so far but what about the possibility of a small logic circuit that throws a relay once a day?

    If simply switching it once a day makes it run continuous then would it not be easiest to set up a timer circuit that clocks a flip flop that would activate one of two relays? As I've said my experience is little but your project has peaked my interest I would love to see what is done for a resolution.
     
  13. adrian_brooks

    New Member

    Nov 3, 2009
    1
    0
    I have a suggestion. Perhaps you could have a voltage detection circuit developed for you that would be preset at 11.9VDC. This circuit would be monitoring which ever battery was currently running the pump. If the voltage of that battery dropped below the threshold, this would initiate causing a signal to be fed to a flip flop chip that would then fire one MOSFET switch while turning off the other and thus allow for each battery to run to it's maximum performance without deep cycling it.

    Hope this helped your design ideas. :)
     
  14. BMorse

    Senior Member

    Sep 26, 2009
    2,675
    234
    I am coming in a little late in this thread (like I always am in most threads....) but just to throw my 2 cents in here....

    Ron, I have made a DIY windmill generator that uses a circuit similar to what you are looking for, this circuit is based on 2 comparators.... but it does use relays for the output.... I know you said you wanted to stay away from those, but My circuit actually uses the IRF640 Mosfets to drive the relays, so maybe you could alter the circuit a bit to accommodate your needs, take a look on my site here >> http://www.morse-code.com/id184.htm << and let me know if you are interested in the circuit.....
     
  15. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    Gee, this thread had been asleep for nearly a year and a half.

    I wonder what happened to Ron Roberts? :confused:

    As I asserted previously, the system really does need some kind of microcontroller monitoring charge levels and system performance, as well as adapting the pump run-time to the battery capacity that's available without running them below 30% charge.

    A 45 AH battery with a 2A load will become 30% discharged after 6.75 hours of run time. Since you may only get 8-10 hours of useful charging time during the day during good weather, you have quite a shortfall of available reserve power. Discharging your batteries below 70% charged will cause them to have a decreased service life. If you went to 50% discharged, that would be 11.25 hours of run time, and you would be buying new batteries every year, or possibly even more frequently.
     
  16. Duane P Wetick

    Active Member

    Apr 23, 2009
    408
    19
    It sounds like your solar panel is incapable of fully charging one battery with no load on it. Without this capability, you will never achieve success.The first step in this endeavor is to acquire sufficient battery charging capacity, one that will adequately charge a 50 AH battery over a 16 hr. period. Actually, a 50 AH battery is marginal, assuming a 2 amp draw over 24 hrs, a 65 AH battery is a better choice. The battery manufacturer can provide you with a graph showing how long a fully charged battery will last with various amp. loads. Note, I said a FULLY CHARGED battery. Which brings us back to the solar panel; can it charge the battery to 100% in 16 hrs?
    I am concerned that the solar panel is the only charging means. Around here, 42 deg N. latitude (US), we have an average of 19% sun. over 1 year. Given this truth, it seems like another charging regime is needed.


    Regards, DPW
     
  17. BMorse

    Senior Member

    Sep 26, 2009
    2,675
    234
    2 other alternatives to charging batteries with besides his solar panels could be:

    1. Add a hydro generator on the output of the pump, so when the pump is pumping it will also produce some energy to store in one of the batteries......
    2. A windmill pump / Generator, he could also add in a windmill generator to generate some power to charge the batteries with. And also a pump driven by the windmill......
    Both can be made using a permanent magnet DC motor (micro generator), lower RPM's and higher voltage ratings would be good.......

    My .02
     
  18. SgtWookie

    Expert

    Jul 17, 2007
    22,182
    1,728
    A pump-driven hydro generator wouldn't work; you're attempting perpetual motion there.

    He can't build a windmill. Building codes and all.
     
  19. BMorse

    Senior Member

    Sep 26, 2009
    2,675
    234

    I am not suggesting perpetual motion with the hydro generator..... the output of the pump can be used to drive a water wheel to turn the generator, with the combined power from solar, hydro, and wind all going into the battery banks to charge, the combined output will be pulled down to the battery levels anyway while charging, as long as he puts in a blocking diode for each source before going into his charge controller..... but most likely he would have to change charge controllers also, since all solar charge controllers are not very efficient, when the battery reaches a full charge the controller just shorts out the output of his panel which is wasted and dissipated as heat through the transistors.....

    I have a setup at home with a micro windmill generator and a hydro generator to charge a bank of 3 12volt 21AH Lead Acid Batteries ( I use the power for outdoor lighting).... the hydro generator is just placed over a natural creek that flows near my property (the hydro generator is built on a platform that lowers or raises with the water level...), and the flow of the water turns a water wheel with an output ratio of 1 to 10 to the generators shaft (1 turn of the water wheel = 10 turns on the motor shaft).... and at about 30 rpms on the water wheel, I am getting around 18 + volts output on my PM DC motor (rated at 90 volts DC, 300 RPMs).....

    And as for the windmill, I am not talking about some gigantic 100 Meter pole windmill, mine is mounted on a 10 foot pole, [​IMG]
    (with a 87" swept area and produces 18+ volts in 4 to 7 MPH winds...)
    [​IMG]
    and they do make smaller ones just for pond aeration and such......

    Besides I was just throwing some ideas his way anyway...... and we will probably never know what he thinks of them unless he comes back around....

    My .02
     
  20. CDRIVE

    Senior Member

    Jul 1, 2008
    2,223
    99
    Woa, don't end this yet! Please post a pic of the water wheel gen.
     
Loading...