Announcement

Collapse
No announcement yet.

Which Voltage is accurate for 50% battery condition? Loaded or not loaded?

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • #16
    Originally posted by Sage Oldmann View Post
    What kind of problems are you talking about? I have all the battery cables the same length. I draw the positive of off battery #1 and the negative off the far end of battery #6. How should multi AGM battery banks be hooked up?
    Your 'diagonal' wiring is a problem. The way to wire batteries in parallel is use a bus bar (method 3 on that smartgauge web page).

    I fault the smartgauge site for showing how to put batteries in parallel, without explaining why it is such a poor design choice. Parallel batteries are inherently unstable. That means that if they are in equilibrium, it is an unstable equilibrium. It's sort of like trying to stand a pencil on its point. Mathematically, a pencil on its point is in equilibrium. In the real world there will be tiny perturbations from equilibrium. When the pencil leans one way, the forces on it cause it to lean further in the same direction. That's positive feedback.

    Batteries in parallel are much the same way. Suppose you had 6 impossibly identical batteries and the wiring was impossibly perfect. The batteries (to start out) would be in equilibrium (but an unstable equilibrium). Consider, for an example, the effects of temperature. If just one battery is a tiny bit warmer than the other, then when charging the battery bank the warmer battery will get more of the current. Because it gets more current, it warms up more than the other batteries. Because its warmer it gets more current. Because it gets more current it gets warmer. Because its warmer it gets more current. Because it gets more current it gets warmer. That's positive feedback... thermal runaway.

    Parallel batteries (for the above mentioned reason and several other reasons) should have a fuse in each string. Of course all the extra wiring for fuses adds more resistance and potential points of failure to the system.

    Getting a DC clamp ammeter (as has been suggested by others) will be an eye opener for you. You may find out that when the battery charging stops, some batteries will be discharging into other batteries.

    --mapmaker
    ob 3524, FM60, ePanel, 4 L16, 4 x 235 watt panels

    Comment


    • #17
      Originally posted by mapmaker View Post
      Your 'diagonal' wiring is a problem. The way to wire batteries in parallel is use a bus bar (method 3 on that smartgauge web page).

      I fault the smartgauge site for showing how to put batteries in parallel, without explaining why it is such a poor design choice. Parallel batteries are inherently unstable. That means that if they are in equilibrium, it is an unstable equilibrium. It's sort of like trying to stand a pencil on its point. Mathematically, a pencil on its point is in equilibrium. In the real world there will be tiny perturbations from equilibrium. When the pencil leans one way, the forces on it cause it to lean further in the same direction. That's positive feedback.

      Batteries in parallel are much the same way. Suppose you had 6 impossibly identical batteries and the wiring was impossibly perfect. The batteries (to start out) would be in equilibrium (but an unstable equilibrium). Consider, for an example, the effects of temperature. If just one battery is a tiny bit warmer than the other, then when charging the battery bank the warmer battery will get more of the current. Because it gets more current, it warms up more than the other batteries. Because its warmer it gets more current. Because it gets more current it gets warmer. Because its warmer it gets more current. Because it gets more current it gets warmer. That's positive feedback... thermal runaway.

      Parallel batteries (for the above mentioned reason and several other reasons) should have a fuse in each string. Of course all the extra wiring for fuses adds more resistance and potential points of failure to the system.

      Getting a DC clamp ammeter (as has been suggested by others) will be an eye opener for you. You may find out that when the battery charging stops, some batteries will be discharging into other batteries.

      --mapmaker
      Thanks, I learned a lot from this post.

      Comment


      • #18
        Originally posted by mapmaker View Post
        Your 'diagonal' wiring is a problem. The way to wire batteries in parallel is use a bus bar (method 3 on that smartgauge web page).

        I fault the smartgauge site for showing how to put batteries in parallel, without explaining why it is such a poor design choice. Parallel batteries are inherently unstable. --mapmaker
        So this method I used is not good? batt_new.gif

        It seems logical. Can I use the six batteries with a terminal post with equal length wires to them?

        Is my current configuration just inefficient or the source of the problem I'm having?

        Comment


        • #19
          Originally posted by Sage Oldmann View Post
          So this method I used is not good? [ATTACH=CONFIG]7837[/ATTACH]

          It seems logical. Can I use the six batteries with a terminal post with equal length wires to them?

          Is my current configuration just inefficient or the source of the problem I'm having?
          It is better than putting both wires at one end, but if you look at the smartgauge website you will see that for more than two batteries the current to the middle batteries is still not equal to that at the end batteries.
          A bus bar is the best method for that many. Among other things it lets you fuse each battery more easily and also remove a battery from the bank without having to wire around i to keep the others working.
          SunnyBoy 3000 US, 18 BP Solar 175B panels.

          Comment


          • #20
            Originally posted by Sage Oldmann View Post
            My starting resting voltage was 12.7 volts.
            Uh oh, brand new agm's should be resting at 12.9 to 13.2v. That's a bit low and I suspect either poor quality agm's or has been discussed here, bad balance. You may want to measure your voltage at rest on each battery *individually* to see if you can weed out a low charge culprit.

            I don't know if you did this, but ideally before putting those 6 batteries together in a bank, charge each one individually at least once, and THEN operate in your parallel arrangement.

            Inetdog is right about the wiring - it is a vast improvement over the typical ladder "at the end of one battery" connection, but in this case, you can do even better with the busbar, or at the very least, moving the positive and negative attachment to the bank itself, one or two batteries down the string. What we're aiming for is to try and get in the middle as possible.

            What's really weird to me is as soon as I turned off the TV, within minutes the voltage meters (3 of them) showed the batteries jumping up to 12.4 volts. The next morning they showed 12.6 volts. How is this possible with no charging going on?
            Batteries will "recover" a little bit when the load is taken off them. That is why most SOC measurements that rely solely on voltage (like with AGM's since we have no choice) want you to wait 4 hours or more for better accuracy as this self-recovery climbs higher and higher.

            According to your hours of *resting* voltage of 12.6 after a tv-watching event, that is actually OK. That is just about 25% DOD for an agm - so cool.

            Again we're assuming that your batteries are actually balanced. From the sound of it, they are operational, but you CAN do a bit more to tidy that up to get the most out of them, such as at least an individual charge on each at least once before going back to a bank-style of operations.

            This will be a good learner agm bank, when once it works and you have tidied up a bit, you may want to opt for just ONE or perhaps only two in parallel of higher quality.

            Tip: if your television is a backlit-lcd, try turning the *backlight* down as far as you can stand it. It will save considerable power. Do not confuse with brightness - it is the backlighting that will save a lot of energy.

            Comment


            • #21
              Originally posted by PNjunction View Post
              Batteries will "recover" a little bit when the load is taken off them. That is why most SOC measurements that rely solely on voltage (like with AGM's since we have no choice) want you to wait 4 hours or more for better accuracy as this self-recovery climbs higher and higher. According to your hours of *resting* voltage of 12.6 after a tv-watching event, that is actually OK. That is just about 25% DOD for an agm - so cool.
              If my goal is to not go below 50% for my batteries, which is around 12.2 volts as I understand it, removing the load when the meter shows 12.2 volts while under load it is not actually 50%. As soon as the load is off, my voltage goes up. My true resting voltage after several hours is 12.6 volts. So practically speaking if I wait to remove the load until the meter shows 11.9 volts I'd actually be closer to 50% than when it shows 12.2 volts. As I see it my true resting voltage will probably be around 12.3 volts. or just over 50%. Am I understanding this correctly? I'm not going to do it until I know it is safe to do so. BTW, I use 4 different meters to check voltage to be sure I'm getting an accurate reading.

              What does DOD stand for?

              Comment


              • #22
                Originally posted by Sage Oldmann View Post
                If my goal is to not go below 50% for my batteries, which is around 12.2 volts as I understand it, removing the load when the meter shows 12.2 volts while under load it is not actually 50%. As soon as the load is off, my voltage goes up. My true resting voltage after several hours is 12.6 volts. So practically speaking if I wait to remove the load until the meter shows 11.9 volts I'd actually be closer to 50% than when it shows 12.2 volts. As I see it my true resting voltage will probably be around 12.3 volts. or just over 50%. Am I understanding this correctly? I'm not going to do it until I know it is safe to do so. BTW, I use 4 different meters to check voltage to be sure I'm getting an accurate reading.

                What does DOD stand for?
                Depth of Discharge or Department of Defense, depending upon context

                Comment


                • #23
                  Originally posted by Sage Oldmann View Post
                  If my goal is to not go below 50% for my batteries, which is around 12.2 volts as I understand it, removing the load when the meter shows 12.2 volts while under load it is not actually 50%. As soon as the load is off, my voltage goes up.
                  Correct - voltages under load and while at rest differ in their ability to show the DOD. Under load, the "puekert effect" is in play. That is, the faster you discharge, the larger the voltage drop, and hence the quicker the voltage recovery. Many manufacturers may have convenient charts to show the DOD while under various loads, as well as the "resting" DOD.

                  Ie, a ridiculous example (values are made up here) - you run a band-saw for a minute from an inverter off your battery and it instantly drops to 10v. Stop the bandsaw, and it recovers quickly to 12.7v.

                  Now, take a 10 watt CFL or LED bulb off that same inverter, and run it for 10 hours when the voltage shows 12.2v under load. Turn off the light (and inverter!). Now after 12 hours, the rested voltage won't make much of a recovery - maybe 12.3v !

                  My true resting voltage after several hours is 12.6 volts.
                  That is just a bit under the typical 12.7v rested voltage for 25% DOD, maybe 30%. So in reality, you only took 30% from the battery, and had 20% left to play with.

                  You can see how this is a crude general guideline, and it is far better to mathematically calculate or physically measure it. But, yeah, you could just run stuff for awhile without obviously killing it, let it rest for at least 4 hours (12 hours better), and try again little by little. Man, this is a really rough way to do things.

                  What does DOD stand for?
                  Depth Of Discharge. 100% DOD is dead. 0% DOD is fully charged.

                  Compare this to the opposite you sometimes see, aka "SOC" or State Of Charge. It is just the opposite. 0% SOC is dead. 100% SOC is fully charged.

                  The only time you can't mess this up accidentally is if you reference 50% SOC or DOD.

                  Comment

                  Working...
                  X