Hi Welcome to the Eev blog an Electronics Engineering Video blog of interest to anyone involved in electronics design. I'm your host Dave Jones Hi The previous blog was a tutorial on battery capacity now I Thought I'd Follow that up just quickly with a very, uh, quick little practical demonstration of how to measure battery capacity. Now, as I mentioned previously, there's two different ways to measure or specify battery capacity. The first one is W hours and the second one is Amp hours.

Amp Hours is a more simplistic figure uh, as I explained in the previous blog I Won't go through it again, but the true capacity is measured in wat hours. Now let's have a look at this. if we've got our Uh, the voltage of our battery on the Y AIS on this left hand Y axis here, and the current on the right hand y AIS and time discharge time on the x-axis Then, as we've as we saw last time, the voltage of the cell is not constant. It will drop or the voltage of the voltage pack.

Depending on what you're actually measuring, it doesn't have to be just one cell. It could actually be multiple cells in series parallel, a combination of both uh Etc, but it will have a characteristic curve which may look something like that. Now, if you're measuring a simple Amp hour capacity down here like this, uh, and you're using, you typically use just a constant current uh discharge so the current curve won't change over time. it'll just be completely flat like that at one continuous figure.

But if you're trying to measure W hours like this down here, then you have to monitor both the voltage and the current. And uh, you would do the wat hour one typically using a constant power source like that. and well, well, you don't have to. but that's what you might typically uh do because a constant power might be uh to characterized, say an ID or DC to DC converter or something like that.

So in that case, your uh, you won't have constant current. Your current will change with the drop in of the cell voltage. So as the cell voltage drops like this, the current will increase like that. As you'll see, to measure, Wat Hours is a bit more complicated and requires a bit more gear than a simple Amp measurement.

For just to measure amp power, all you need is a constant current like this constant current load which I used in the previous Vog. Very simple to BU build. quite trivial and all you need is a stopwatch to time it and a multimeter hooked across the battery to determine the cutout voltage which we'll call V cut. But what hours to measure that? Uh, you need to log over time.

You actually need a Data Logger Um, either a PC or a multimeter that can data log and you've got to measure both the voltage and the current until the desired voltage cutout point. So here's what you need to measure W hours or Amp hours for w hours Here, As you can see, it's quite complex. You got your battery or your pack, which you're actually measuring. You need a current shunt resistor like this to be able to measure the current.

You can put it in the low side or the high side, depending on which amp you got, but you know it typically might go in the low side. You've got your constant power load. Once again, that's got to be fairly intelligent. To get a constant power load isn't as simple as a constant current load, so take that into account.

And then you need two differential amplifiers, which is very important as we'll see in the Practical demonstration. And then you need some way to log it. Uh, it can be I've done it as a as a PC into a data acquisition card. Or you can do it with two data login multimeters or something like that.

You need some way to accumulate all that data over time so that you can do your individual what hour calculations over time and then accumulate them all up to get yourself a total wat hour figure. Now that's you know, it's fairly complex to do. You got to have the gear to actually do uh, W hours. But and power measurement.

It's simple. You've got your battery under test, a simple dumbass constant current load which is a Fet and a Opam and not much else really. and a multimeter and a stopwatch and that's it. Um, and you just measure the um, uh, you just set it up and just uh, count the time with the stopwatch until it gets from the Uh.

the voltage on the cell gets from fresh right down to your cut off voltage. Simple So anyone can do amp hour measurements W hours bit harder. As I mentioned in the previous blog, there's two ways to get your W hour uh, capacity figure. Actually, to actually log it, one is to get the Uh the voltage and current curves I've representing V * I up here.

So you power curve effectively and then do an integral of the area under that curve. But you've got to do integrals and well, you just don't have to because there's an easier way one is to Simply take the Um the Uh regular measurement say like it might be at 1 second intervals like that and because the voltage isn't going to change a massive amount in one second usually um for most applications, then it's a very good estimate and you can calculate the wat second figure for that particular chunk and then you do it. I've expanded that because it might take hours to discharge, but if you accumulate and just add up all these what uh second measurements, you can get a total figure. When you come to the end, a total figure in what hours or you know what seconds or whatever you want or Jewels doesn't matter.

So that's how we're going to do it in today's practical experiment. So what actually is the true what? Our capacity of a doublea cell? Well, we can't get it from the data sheet, so let's measure it, shall we? Now it just so happens. I've got this brand spanking new Metrahit Energy multimeter here which actually allows me to measure uh, not only voltage and current at the same time, but it allows me to actually measure capacity in W hours as well. Fantastic! So what we're going to do is discharge this cell at a no and constant current using my constant current um load here which is s in a previous blog and I'm going to discharge a cell a Um standard ablea Energizer alkaline and see what we get.

Now, as I mentioned before, there's two ways to actually measure the wat capacity of a battery. One is to get the characteristic curve and then integrate it uh, over time and do some math and actually integrate it. Or you can use log the voltage and the current directly on the cell over time and that will build up a w hour figure for you. And that's exactly what the Metrahit Energy Meter does.

So let's give that a go. As you can see, it actually displays the cell voltage and it displays the cell current. There we go. 260 mamps.

It's actually 262.50 You can see the accumulation there of the Wat hour or The Milw hour figure and by the time this battery gets flat, it'll actually get down to um A. It should get down to a Wat hour figure which we guesstimated is probably about 2 and 1/2 Wat hours for a standard Energizer Ablea cell. Now the Metr Hit Energy uses what's called a three terminal measurement system. It's got a Volts terminal, Com and an amp just like a normal multimeter, but because it can sample both at the same time.

This is my little Dave C P drawing here and you can see that we've actually inserted the current uh, measurement part of it into the negative terminal of the battery. We've got our constant current load here, which I'll set for 250 milliamps and we've got the cell. now. the disadvantage what with this three terminal measurement is that this value here because this internal voltmeter here okay is actually measuring the differential voltage between the two input terminals.

Like that the ADC is there. Then any current flowing through this wire down into here from the cell over to here that is actually going to cause a voltage drop and upset the measurement. So that's a disadvantage of this three terminal measurement. So what we want is a big chunky wire as short as possible right here.

Now as a practical measurement here, I've set this up and it's drawn about 260 milliamps as you can see, but the Metroid energy is only measuring 1.03 7 Vols Now, if we get the fluke here and actually measure the voltage directly across the battery like this, you'll find it's actually 1.20 So what's going on? There's a discrepancy here. Now, if we actually move this voltage terminal from the end of the battery to the actual input jack, you'll see it's 1.05 six, which is basically the same as what it's showing here. So that little tiny bit of wire there going jumping from there over to there and the contact resistance of the spring terminal and all that sort of stuff is enough to cause that voltage drop at 250 milliamps. Now you notice if we turn the current right down 1.25 9 Vols and we'll measure the battery voltage.

There we go. One point, it's practically spot on because there's no current causing a drop in that little tiny lead there, so we have to work work on optimizing that. Now, normally you would actually do this with what's called a four terminal measurement which I've mentioned in another blog for resistance measurements, but in this case you would actually measure the differential voltage straight across the cell into an amplifier like that and you take it off and then you'd measure the current into another amplifier and you would actually log the voltage and current using a PC or data acquisition card or something like that. But in this case we've only got the three terminal resistance measurement on the Metr hit extra.

Now what I've actually done is I've squeezed the wire in there between the spring terminal and the battery just to avoid that actual spring terminal. Now let's see if that makes a difference. Take it up to 250 where we were before and 1.25 volts. Now let's measure directly across the cell.

We can actually measure here here and here. 1.25 3. There we go. pretty close.

so it was the spring terminal actually causing the problem with our measurement there. Now the actual cell I plan on measuring is actually a Duracell proell. It's a standard alkaline, just a rebadging to stop pilfering. Hey, go figure.

Um, it's exactly the same as a regular alkaline. It's March uh, 2016 expiry, so it's It's not brand spanking new, but it is straight out of the Uh box. It is original condition so shouldn't have dropped too much capacity at all. Let's consider it brand new and we're going to do 250 milliamps, which um corresponds to the characteristic curve here and we should get about 9 hours.

uh, use out of it down to 0.8 Vol So it's um, late night here, so I'll head to bed now I'll set this up and I'll leave it running overnight and we'll accumulate the charge on here and see what we get. And here we go. It's 1.52 Vols at 250 milliamps a smidgen over. But let's not worry about that.

and let's start the Uh Energy Measurement Here we go, it's reset and it's counting down. Well, it's counting up. So I'll come back in uh, 8 or n hours and we'll see it's accumulated M hours already. Look at it.

Go look at the resolution on this thing. Oh, all right, it's morning time and as you can see, 7 hours and 53 minutes later, not quite 8 hours. We've got 2.06 W hours total and as you can see, the current has dropped drastically to 37 milliamp so it looks like it's completely dead if we switch back. Yep, the battery is only 144.5 molts.

at 36 milliamps, it's completely dyed so there it didn't even get close to meeting its uh spec here of let's take a look at it. It was supposed to at uh uh, 250 milliamps There, it was supposed to get at least 9 hours down to 0.8 Vols We didn't even get 8 so it's dropped off completely before that. Unbelievable. So let that be a lesson to you.

You can't always trust batteries to meet their performance spec even when they're well within date, even when their quality Brands like this. So the answer is for a quality alkaline cell like this Jurus cell proell with with 4 years left on its sh on its stamped shelf life Fresh out of the Box has got just over two wat hours capacity at 250 milliamps continuous current discharge. So what did we get from that practical measurement? Well, as it turns out, I didn't get here quick enough and it had already gone past the 0.8 Vols cut off voltage that I wanted. So I'm going to round it down to say to roughly guess that it was at about round about the 7.5 hour mark that it got down to 0.8 Vols.

So that and we know the actual the meter because it's really cool. It can calculate wat hours for us. We know the actual wat hour figure of that battery even below 8 volts. but it drops off sharply is around 2.06 W and we can calculate the milliamp hour figure as well because that's trivial cuz we were using a constant current load of 250 milliamps.

We uh, know it's well. or we guessed it's 7.5 hours and that gives us a capacity of 18875 milliamp hours. So what does that tell us? Well, it actually doesn't tell us very much at all. And this is the Crck with battery capacity measurements.

We know accurately what the figure is for for a 250 Milah constant current load. But is your load for your product going to be 250 milliamps constant current? Probably not. So really, you can't use this data, this milliamp figure, or this wat hour figure down here to calculate the capacity for your product. Um, you really have to measure the uh, the capacity of the battery for exactly the type of load you're going to have on your circuit.

So really, we can't see much at all what will happen? Um, well, we can because we know that anything above 250 milliamps constant current due to the IR this capacity is going to drop. It's not going to go up, it's going to drop. but at lower capacity. Say if at 100 milliamps constant current discharge, we could expect this figure to go up and the water and the corresponding water hour figure to go up as well.

But how much it goes up by, or how much it goes by down by, we don't know. We have to do further measurements. And here are two simple examples where constant current and constant power might be used and which one you might have to use to measure your battery capacity. Now, Constant current? Uh, you would might typically use that if your circuit.

Here if this resistor represents your circuit, and let's say your circuit is drawing roughly a constant average amount of power, it might be pulsing or something like that. But let's not complicate it. Okay, it's drawing a constant amount of power because it has a constant voltage. It's being driven by a something you should know, a 7805 voltage regulator, right? It generates a constant voltage over a constant resistance, which gives you a constant current load.

Okay, and a constant power load. and that's going to give you a fixed current into here. Now because of the nature of linear voltage Regulators like the 7805 or the LM 317 or something like that, this input current here is going to be the same as the output current. It's a little tiny little bit lost down here, but you know, let's not include that input current equals output current.

So it's effectively constant current being drawn from your battery. So that's an example of where you might use constant current. Constant power, on the other hand, requires something like a DC Todc converter. exactly the same load.

This is your product down here. It's once again a fixed voltage. Let's say it's 5 volts or something powering your circuit drawing. Once again, a constant amount of current.

It's drawing exactly the same amount of power as it was up here, but in this case because of the nature of DC to DC converters, the input current will actually vary. It'll vary as the input voltage drops. So if when your battery voltage drops as it follows the characteristic discharge curve, your input current is going to go up. So as you can see, you can't just measure the current because the current from the battery will not be constant.

It will vary. So you have to actually look l or measure both the battery voltage and the battery current to get power. And that's what you want to do. because it's a constant power in the load.

We're assuming an ideal DC to DC converter. I Won't go into details about how the efficiency of converters, you know, drops at both ends of the current scale, but let's not go there. If it's an ideal DC to DC converter, which for the sake of many arguments, you can say it is an ideal converter, then you want to me be measuring constant power. So that's why when you're measuring the capacity of the battery, you want to simulate a constant power load.

So as you can see, battery capacity measurement and specification is not easy. It all depends on a whole bunch of factors and so if anyone tells you, comes along and says oh, this battery has a capacity of X, tell them there, they don't know what they're talking about, Tell them to provide more information or say that's assuming a constant current or that's assuming a constant power or something like that. What does it happen under pulse words, What does that happen under of this? What happens when the battery voltage drops, etc etc. Far too complicated.

Anyway, that's a simple battery capacity measurement. And in case you're wondering, it's Australia day here in Sydney January 26th and it's pretty darn hot in the here in the Eev blog lab. Over 35 C Fahrenheit No idea you Yanks figure out what 35 Celsius is. It's getting quite warm and I'm sweating.

Time to go back into the air conditioning. See you.