Voltage readings- "bounce?"

johnelaruejohnelarue Solar Expert Posts: 33
Hi all,

Another super newbie question, tried searching the forum, but nothing specific to my question.

I have three, 50watt PV's hooked to a 15A mppt CC, and a 115Amp/hour RV/Marine battery, 150watt automotive DC-AC inverter, powerstrip. all 12V.

When running lights at night the battery voltage runs down to near 12.3V, which is 70% of full charge, as far down as I want to go on my battery without damaging it , so I read. I double check battery voltage at the CC and the battery terminals.

My loads are super light, a single strand of LED's rated at 1amp.

So of course I shut everything down to be safe, but after a while battery voltage goes back up to safer levels, and by morning back up to 90-110%.

I have an analog voltmeter connected to the battery input at the CC. This for some reason always reads 0.5V lower than my digital multimeter, not a big problem (the analog meter is old- I have 3 identical, they all read 0.5 lower than the digital multimeter)

So, how can I get accurate voltage readings while running loads? And not "false" readings that make me shut everyting off.

Sorry for the long roundabout explanation,




  • AntronXAntronX Solar Expert Posts: 462 ✭✭
    Re: Voltage readings- "bounce?"

    You can go lower that 12.3V without damaging the battery. When you read advice saying avoid repeatedly discharging the battery below 50% state of charge, that's not so much to avoid the damage, it is more like to extend that battery's useful life. The damage supposedly starts when you go below 10.5V, but even then batteries may recover if caught early and recharged soon and never taken below 10.5V again. So, by discharging down to 0 - 10% SOC (10.5 - 11.5V) you may get about 200 cycles out of your battery before it looses 20% of it's original capacity. On other hand, you may get 1000 cycles if you only discharge to 50%. That information should be in your battery's data sheet.
  • BB.BB. Super Moderators, Administrators Posts: 32,741 admin
    Re: Voltage readings- "bounce?"

    Part of the issue is that battery voltage under load or under charge does not represent the state of charge of the battery bank... Only resting voltage is (roughly) accurate.

    You can get a good quality hydrometer and read the state of charge by electrolyte specific gravity.

    Another way is a cumulative Amp*Hour / Watt*Hour meter... Lead Acid batteries are pretty much 100% efficient in Amp*Hours out vs Amp*Hours back in on charge... Where they do diverge is during the final bit of charging and equalization.

    So, if you have a load that is drawing 1 amp from a full battery--then you would have to run it for 25 hours to get:
    • 1 amp * 25 hours / 115 AH battery = 0.22 or 22% discharged (or 78% state of charge)
    Assuming that is the only load--you are nowhere near stressing out the 115 AH battery.

    A steady load (or charge) on a battery is pretty straight forward to estimate state of charge... It is the variable loads that are difficult to estimate.

    I always suggest a Battery Meter so you can monitor the estimated state of charge of an operating battery bank. Checked once in a while with a hydrometer.

    But battery monitors are not cheap... If you have a small bank and just want to experiment--these Amp*Hour / Watt*Hour meters used by RC (remote control model) hobbyists are pretty nifty. Not as sophisticated as the Battery Monitors--but much more useful than a simple volt or amp meter.

    Near San Francisco California: 3.5kWatt Grid Tied Solar power system+small backup genset
  • AntronXAntronX Solar Expert Posts: 462 ✭✭
    Re: Voltage readings- "bounce?"
    BB. wrote: »
    But battery monitors are not cheap... If you have a small bank and just want to experiment--these Amp*Hour / Watt*Hour meters used by RC (remote control model) hobbyists are pretty nifty. Not as sophisticated as the Battery Monitors--but much more useful than a simple volt or amp meter.

    Oh I got 3 of those, love 'em.
  • mike95490mike95490 Solar Expert Posts: 9,583 ✭✭✭✭✭
    Re: Voltage readings- "bounce?"

    When you have 3 meters "agree" and one that does not, you may want to get one calibrated, or measure a reference something, to see who's right.

    Powerfab top of pole PV mount | Listeroid 6/1 w/st5 gen head | XW6048 inverter/chgr | Iota 48V/15A charger | Morningstar 60A MPPT | 48V, 800A NiFe Battery (in series)| 15, Evergreen 205w "12V" PV array on pole | Midnight ePanel | Grundfos 10 SO5-9 with 3 wire Franklin Electric motor (1/2hp 240V 1ph ) on a timer for 3 hr noontime run - Runs off PV ||
    || Midnight Classic 200 | 10, Evergreen 200w in a 160VOC array ||
    || VEC1093 12V Charger | Maha C401 aa/aaa Charger | SureSine | Sunsaver MPPT 15A

    solar: http://tinyurl.com/LMR-Solar
    gen: http://tinyurl.com/LMR-Lister ,

  • icarusicarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Voltage readings- "bounce?"

    Bill has it right.

    A battery will only reveal "true voltage" after it has not been loaded or charged for at least an hour, making voltage a poor way of measuring battery state of charge. If you don't have a battery monitor, try logging your loads for a while (number of watts X time) to see what your average daily draw is. Then measure the battery voltage/specific gravity at rest. If you consistently use ~20% of battery capacity relative to the stated AH capacity of the battery, AND the at rest voltage agrees with that within a few percent then you know how to manage your battery a bit better.


    By reading and understanding the above links, you will learn a bunch more about care and feeding of batteries.

    In round numbers your ~150 watts of panels should give you ~300 watt/hours of power per day (figuring ~4 hours of sun) Your 115 ah battery should give you ~ 1400 wh if you were to draw it to zero. Drawing it down 25% ( a good compromise amount) might be ~350 wh so in rough numbers you should be able to take out 20 hours of 15 watt cfl on a good day. If you wish a three day reserve, figure ~6.5 hours or so.

    As has been suggested, as you use your system more, a battery monitor is a must. I suggest the Trimetric. FYI, we have ~300 watts of panel and we use ~600 wh day in, day out, into and out of 450 ah of battery.


    PS Welcome and good luck
  • john pjohn p Solar Expert Posts: 814 ✭✭✭
    Re: Voltage readings- "bounce?"

    I would be hopeful of getting more than 300watt hrs per day from 150w of panels...In round numbers your ~150 watts of panels should give you ~300 watt/hours of power per day (figuring ~4 hours of sun
  • icarusicarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Voltage readings- "bounce?"

    The general rule of thumb for battery based systems is a derate of ~45% Net system efficiency of ~55% I use 50% just to keep the math easy. I then use 4 hours as a good average. Unless you live in a very clear place, 4 hours of "good sun" is a fairly good average. MIght get 6-8 in the summer depending on latitude, four or less in the summer. (Not everywhere is as clear and dry and low latitude as much of Aus.)


    On other observation from experience is that people tend to over estimate the amount they can harvest. You might double that on many days, but you might halve it as well. That is the problem with "averages". You gotta figure out how you're going to play them. As I suggested in the earlier post, we live with ~300 watts of PV (more now) and we routinely draw ~600 wh/day,, maybe a few more since some doesn't show up on the Trimeteric due to load shifting. We only have to run the generator after 3 days of no sun, or 4 or 5 of partial sun, which is almost never. I've added another 125 watts of late just because I got the panels cheap, and the loads (as always) keep growing.
  • johnelaruejohnelarue Solar Expert Posts: 33
    Re: Voltage readings- "bounce?"

    Excellent advice as always guys,

    Will be looking for a hobby monitor/meter soon.

    A gold medal to you all,

  • bryanlbryanl Solar Expert Posts: 175 ✭✭✭✭✭
    Re: Voltage readings- "bounce?"

    re: "A battery will only reveal "true voltage" after it has not been loaded or charged for at least an hour, making voltage a poor way of measuring battery state of charge."

    The same could be said for specific gravity.

    The truth of the matter is that voltage is as good as specific gravity for measuring state of charge and that both of these measures need proper procedure.

    The 'net charge' or what I call electron counter type battery meters infer a state of charge based on a number of assumptions and their programming. I do not know of any of these that really account for all of the major factors that can influence available capacity. Temperature and Peukert effects are about it from what I have seen.

    For live battery state of charge assessment, the only reliable method I have seen is SmartGauge Electronics - battery metering. That one uses an analysis of voltage as a function of current for its determination. It makes use of the "voltage bounce" and a knowledge of lead acid battery behavior to assess condition and status.

    No matter what you use to assess battery capacity, assuming that you have an accuracy of better than 20% is probably not a good idea for small battery banks.
  • icarusicarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Voltage readings- "bounce?"

    I agree with all of that. For a system as small as yours, possibly the best way is to keep a log, and see how the battery performs. The reality is your battery is fairly cheap, so if you expect it to go 3 years and you get five, you are probably doing something right on balance. On the other hand if you kill it in two, you might change your habits with the next set, or buy a bigger battery (or batteries).

    No matter what you do, there is a trade off between battery capacity and battery longevity. When I built my current system I went from 4 L-16s to 4 smaller T-105s. My calculation was, if the t-105 last 5 years with the loading/charging I do, they will be cheaper than the L-16s lasting 7 years. I fully expect the T-105s to last ten, but since they were ~1/2 price relative to the L-16s it is all a gamble I am willing to take.

    IMHO, the key is keeping the discharge % manageable ~20%, and making sure the batteries charge with enough current to charge fully every day, but to also charge fast enough to keep the electrolite well mixed. ~5-10% of AH capacity. Lower current may get them charged, but you will pay a price in stratified acid. (that sounds wierd!)

    Good luck,

  • johnelaruejohnelarue Solar Expert Posts: 33
    Re: Voltage readings- "bounce?"

    I ordered another battery to boost AH and provide more of a cushion.
    I have a log in the control box and will be keeping records as well.

    Thanks for the info bryan
  • icarusicarus Solar Expert Posts: 5,436 ✭✭✭✭
    Re: Voltage readings- "bounce?"

    Be advised that doubling the AH capacity of the battery without raising average charge current may lead to other problems. A battery that is too small for it's charging array, has the potential to be chronically under charged and or over drawn. You might wish to target your charging to at least 5% of AH capacity, ideally closer to 10%. So your ~200 ah would like somewhere between 10 and 20 amps. Pretty hard to do consistently with only 150 watts of PV.

  • RCinFLARCinFLA Solar Expert Posts: 1,460 ✭✭✭✭
    Re: Voltage readings- "bounce?"

    The recovery you are seeing is the result of reestablishing kinetic equilibrium back to battery.

    Whenever current flow out of or into battery there are chemical reactions and molecular movement (kinetics) that require some plate polarization voltage that causes the terminal voltage to be lower on discharge and higher on recharge.

    This includes conversion of lead and lead dioxide in plates to lead sulfate and sulphuric acid to water during discharge. Process is reversed on recharge. Also there is required movement to diffuse electrolyte water and sulfuric acid.

    The amount of this plate polarization overhead depends on the age, condition, and type of battery. Higher discharge rate consumes more overhead polarization voltage to accomplish the higher rates of kinetic activity. The Peukerts effect is just some guy that measured and defined these effects on discharge rates. There are two major componenents to this degradation and Peukerts does not fully account for all the factors. When data says the battey capacity is reduced to 65% of 20hr discharge rate at higher discharge current it is interesting info but most applications have a continously varying rate of discharge current.

    The first is the pure kinetics of moving material around and molecular transformation. For example, when discharged the electrolyte becomed locally diluted to water where sulphuric acid is needed to interact with plate to create lead sulfate. It takes some time for the diffusion of electrolyte to reach an equilibrium in the acid depleted surface. When the current stops much of this will recover to the equilibrium were the uniform electrolyte mixture now represents the rested (kinetic equilibrium) state of approximately cell voltage equals specific gravity of electrolyte + 0.845 volts.

    In a normal flooded cell, the electrolyte remix time lag is usually a large piece of the loaded voltage drop and is almost totally recoverable when current flow stops. AGM battery for example, captures electrolyte more locally on a smaller per square area of plate so there is less time lag in the remixing of electrolyte. AGM's don't drop in voltage as much under load and recover equilibrium state much quicker when current flow stops or is reduced.

    The second involved access to the lead surfaces of the plates. Higher discharge rates can prematurely plug up channels in the negative plate sponge lead surface with lead sulfate. This effectively takes some of the plate surface out of use. This will reduce remaining battery capacity to some degree but will not have an impact on the recovered equilbrium 'rested' battery voltage which is primarily based on acid concentration (SG). It requires recharging to clear the 'artery blockages'.

    There are also pure IR loses due to conductor resistance of grid, lead, and terminals. This is lost in heat and consumes actual battery capacity.

    A columb counter meter, like Trimetrics, with a 'fudge factor' setting that represents your particular current rate use profile is the best way to have a battery gas gauge.
Sign In or Register to comment.