Hi yes, it's alkaline battery leakage time again. I Thought we'd revisit this after my previous video linked in down below at the end and up on the card here. If you haven't seen that, where I go through some long term like nine month testing of various brand alkaline batteries to see if they leak. Unfortunately, that video didn't produce any results.

so I'm gonna redo it again. Yes, I'm glutton for punishment with a different testing methodology at this time. Check the previous video for how I did it last time. Anyway, I Thought I'd be fun to redo it and maybe we'll get some results this time.

I just look through like a random older battery box and I found this vard and it's made in Germany Vata battery and wow, that is a shocker. So what you're looking at here is a leakage of potassium hydroxide from the battery through the seal in the negative end of the battery and that's where the seal is. There's no seal on the top. That's why you'll never see them like leak from the positive terminal of the battery.

And so let's say that eventually leaks out and then that potassium hydroxide forms as carbon dioxide in the atmosphere and that forms all these little dendrite e growth and crystalline structures that you see here. and it's rather funky to look at actually. I Do actually enjoy looking at that even though it kills all your gear that you've got these things in. Never leave like discharged batteries long-term in your any product because this could be the end result at least for our Malka lines.

Anyway, this phenomenon is particular to the chemistry used in alkaline batteries, but crusty. so once again I just got a whole bunch of different brands and stuff. These are the only energizes that I could get from my local supermarket. they're currently you know in like a social distance in lockdown.

Anyway, they had so Energizer Max Plus I've got two different types of Jura cells because Jura cells are notorious of course for leaking. So I wanted two different types I've got the regular cop top and the jurors Coppertop Ultra as well. Then I've got to Sheba Jobbies I got Phillips I got max also all our quality brands Panasonic that we used last time two different Panasonic's actually ones the E Voltar type which is supposed to have leakage protection in it so that'll be interesting. two different types of mating Germany Varda ones ones a high energy ones, just a long life job.

and then I've got a couple of just no-name cheapies. I've got wallaby. What do you want to be? whatever you wanna be? I've got that Kohl's brand which is just like my local supermarket chain brand so who knows who makes those and just some generic. eBay one cord juice bank and all of these have reasonable use by dates on them, so that really the use by dates not going to factor into it a huge amount.

although you know if they're expired, you wouldn't leave them in product. So the self discharge is probably going to kick in. But anyway, it's not a concern for this particular test. They're all well within date.

And for those curious here are the batteries that from the previous test and I've kept them in filing cabinet in their in the drawer and you can see that there is no leakage on any of them. so even after all that time Yeah bugger. So the difference to last time is that I just had these battery holders here and I had one half discharged and one fully discharged because I thought like reverse charging one of the batteries might have done the business and you see in the previous videos that some cells did actually reverse voltage. they reversed charge and that is actually a thing but none of them linked.

So what I'm gonna do is before: I had a justice double-a to double a holder and a single 47 ohm resistor on here and I just left these on here indefinitely and as I said, I think that load was too low and it didn't have time for the pressure to build up inside the battery and hence cause the leakage of the rubber sea or plastic rubber, whatever material they're using on the negative terminal of battery and that's how it leaks out. So I thought time I would actually discharge the batteries either 90 percent or a hundred percent in quote marks and then put like a light load on it like a hundred. K for example. So a hundred one volt divided by 100 K is 10 micro amps and that 10 micro amps is a fairly typical like stamp products To end by current figure.

So I Figure: Like you know, because you get leakage and a lot of this equipment that has like soft power buttons and standby and things like that. Of course you can get them with absolutely no standby power at all. But anyway I Thought I would So I'll change these to a hundred K Please leave it in the comments down below if you don't necessarily agree with that limit. but it's something, right, we've got to try something.

But because I've already got a whole bunch of these things already made up. I Figure: I will use these to actually discharge the battery itself. Now let's have a look at a datasheet for a typical Alkaline. here.

we've got the Duracell Copper Top double a nominal 1.5 volts hundred 20 millions at one kilohertz for those playing along at home. Yes, if you want to measure the true battery impedance, you have to measure it AC at one kilohertz. Anyway, that's just a nominal figure. It's just like the ESR for capacitors, for example, is measured nominally at 100 kilohertz.

Why it's just a nominal figure, it's just yeah. you've got to pick a frequency. It's just like why does every oscilloscope have a 20 megahertz bandwidth limit? Even today, many decades after, we've gone way beyond 20 megahertz analog bandwidth scope. Still, they still over 20 megahertz analog bandwidth figure.

And why is that 20 megahertz? The standard for measuring power supply noise. For example, check out these specification: I Know going off on a tangent here. That's what they do. Sorry, can't help myself Check out a power supply datasheet.

The noise. For example, we've measured over 20 megahertz bandwidth. Why 20 megahertz and Wow Because scopes have a 20 megahertz bandwidth limit. Why do they have a 20 metres bandwidth limit? It's just a it's just an industry value the industry picked and a de facto standard.

Same thing here: 1 Kilohertz. There you go anyway. So there's a couple of ways to discharge this battery. Could use a Const current load for example.

I've got several proper electronic loads here of course and but doing it but draining if one of these batteries could take like 24 hours a pop for example. and I want to do like dozens of these batteries so I could string them all in series. Of course I can get two of these per thing and I can like wire on them all up in series and then do that because the load has I think my load takes up one of my loads, take sixty volts maximum compliance folders or something, or a couple hundred volts anyway. I'm so I could string them all in series and do it that way.

But the problem is I don't know the exact capacity of these batteries and they're all different brands and different types so you don't want any to sort of like reverse charge and be completely out. So what? I thought I've already got these made up. They've already got a forty seven Ohm resistor on them. I Thought: I Just work two of them in series like this.

and hence I there's less risk of one of them reverse charging. For example, work two in series. 47 Ohm resistor. That should do it.

So let's go take a look at the curves down here and see how long it lasts. The Jura Cell: Not all data sheets are going to have the same characteristic curves. The Jura cells have constant current curves like this. These ones for low currents down here.

Five millions, 250 hundred to a thousand milliamps over here. You don't really want to draw more than an amp from a double a alkaline because of the ESR up here. you can come a gutter. Anyway, we've also got constant power curves.

Me, whatever. But bingo. Constant resistance curves. This is what we want.

And look, we've almost got one that matches the 47 Ohms. And here we've got 43 Ohms. Yeah, good enough for Australia right? So I could use the 47 Ohms on here. But unfortunately to discharge this thing like let's say, 90% of its capacity is lost at around that point.

Eight volts cut off voltage. So we're talking I Don't want to have to wait a hundred and one hundred and plus hours for these batteries to discharge. So I might just result of these. We say a 10 ohm resistor for example, don't make it too low, but a 10 ohm resistor and maybe constant resistance? Discharge them.

that'll take at say after 24 hours or maybe not 24. but say after 20. I'll come back after 20 hours and I'll check the voltage on the things and any that are like I really low I'll have to take them out or whatever. But yeah, I probably want to get it down to about naught point 8 volts or something like that.

I don't want to. definitely don't want to go beyond that. Basically, anywhere below naught point 5 volts is like there's no 99.9 percent of the capacity of the battery is gone, but not necessarily due to the electrochemistry. No, I don't have a battery.

Here's the battery holder. anyway. Imagine that's a battery due to the electrochemistry on there and depending on how you discharged it and at what rate you discharged them, you can actually recover some of that charge back into it after you've discharged it. So you might think you've discharged a hundred percent.

Then you take the load off. If you discharge this vigorously, then you might find that some of the charge like five percent or something. you mate. This still might be like five percent left in there and all.

I'm sure all the chemistry experts can explain why and that's the case. But anyway, um yeah, like I don't know I don't know. Maybe like discharge them to a vault. Maybe something like if I come back 20 hours later.

So I come back during the day tomorrow. it's what is it now. almost 3:30 So just after lunch tomorrow, I'll go around check all it, measure all the voltages on all the cells, and if they're around about a volt, I might actually take him out, say and then I might do more I might actually do another lot down 2.8% types. I don't really know.

This is kind of like trial and error here I don't know whether I know at what point I can discharge him like fast. this is fast, over 24 hours and then slowly discharging with 100k resistor for example. So yeah, I'm it's gonna suck it and see. and unfortunately I won't know the results for months.

Like many, many months. six months, nine months. again, something like that. So anyway, if you think I'm doing this wrong I can always.

If you like, don't just go. Or you should do this. like if you've got some other proof like technical documentation, research, whatever from various battery manufacturers or somebody else that shows that you're more likely to get leakage. If you do this particular discharge or whatever, then please let me know.

Leave it in the comments or send me an e-mail or whatever because I need to like I've had a look and really there's not I couldn't find anything so it's kind of just suck it and see here. Pretty much so. Yeah, Anyway, hopefully we'll get a result Murphy won't let me, but give it a go anyway. So basically yeah, anything beyond Eight Point Eight Volts I Think from memory it's like you know 95% of the capacity of the batteries gone at nah 20 volts.

So that's what the batteries are tried to do. The batteries are tried to tap into that extra like 5% and get 800 percent the battery the life of the dog. God, let's not go then again, yeah I died in the ass. this websites still there.

but but I don't think they sold many. they didn't get that big, came out of order they wanted. Yeah so let's just pull up some random curves here that I found. Double a alkaline cell is discharged at 1 amp.

Here we go. you know, 500 milliamps or whatever and you can see that basically once a point, five point, Six point, seven point, eight volts there you can see that there's very little area under the curve. we're talking. You know, like as I said like 5% tops and and as I said if the higher currents are for example, like let's go ahead and choose one amp.

you know it's a decent discharge and no print five. Oh, what's wrong. Six, seven, eight. Okay, so eight is in there, just the lines gone.

That's no, that's got a reasonable amount of capacity left. I don't know it's the area. Once again, you know you're not going to get much more than five percent of the capacity left at naught point. eight volts.

Some people might argue should discharge at two point Five. But yeah, no, no, I don't think so. I'd be happy with point Eight max. I might even stop it at a Volt for example.

So most of the capacity is gone and then we'll just slow discharge it with our huh decay resistor at 10 micro amps or so. Anyway, that's the plan. It's suck it and see. So I think the plan will be two batteries of each particular brand and type a discharge - one volt and then I will with the four with the 10 ohm resistor and then I'll put I'll replace the resistor with a hundred K and then I'll leave both of those two in series of the same brand and then I'll just leave it for months and months and months as it slowly it effectively oh, not quite self discharges, but like really ultra low load.

Like as I said, like sub 10 micro amps standby load and see what happens. But I think I'll also do another two sets of each brand and type battery and then just leave them with no load Or I might just yeah, might as well do two because I got a discharge two at a time. so I might just do two of each type. So therefore I'll have four batteries of each brand and types.

This is quite a large test it's not you know but ultimately to get data from something like this where probably we're really relying on lucky and we're relying on the characteristic manufacturing bell curve of the batteries to have like just outline faults in the pressure vents in them. and that's really what we're trying to determine. So really, you really need like hundreds of batteries at each type to get like to sort of guarantee a result that maybe you know like one of them is gonna leak or something. You know you've got to like a one percent failure rate so one in a hundred for example might leak.

So yeah, we're I don't know. but if we can say get both of a particular brand leaking through both of their negative terminals then that'll be I Like a really good result given the number of cells that were actually testing. that's and that'll be positive proof that that particular brand has not so good manufacturing on there in sales are for example. So anyway, that's the plan but that place your bets down below set up a pool on the Eevblog for about take your bets of how whether or not I'm going to get a result after six months Anyway, any updates will go on my secondary eevblog to channel.

If you're not subscribed eevblog to, there'll be a link down below and at the end. and you should subscribe to Eevblog too, because that's where I dump a whole bunch of videos which you'll never see otherwise. And don't forget to subscribe to my library channel because I might occasionally put some exclusive videos on there as well. If I'm going to make exclusive videos, I'm going to put them on my Patreon account, my supporter section, on the forum, and on my library channel as well.

That's the plan anyway. So and there's already an exclusive video I over there. you should go watch it, my library channel down below Anyway, fingers crossed. Catch you next time.