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 I've seen reports in the last couple of days about the Amazon Kindle 3. There's been a problem with it. People who have bought one of these uh, genuine Amazon covers for it, the one without the light.

They're saying that the things are resetting all the time. you. thousands of people apparently are getting this problem and Amazon um have actually been replacing them. They will replace it, but apparently sometimes it doesn't even fix the problem.

So there's something going on here and people have asked me to check it out. so let's do just that now. I Don't actually have one of these covers? Uh, that doesn't have the light. This is a friend of mine.

he uh loaned this to me and it's got the LED light on it now. Apparently this cover does not cause this uh, lockup problem but I figure it's a good enough uh represent presentation cuz I've looked at the photos and it looks like the Um attachments and everything which we'll go into are identical. um, except for the fact that they painted on this one. So um, on the non-light version.

So that's not going to stop me investigating not having the exact cover. I Think we can do some reasonable investigation without it. Now the first thing I want to look at are these contacts here because um I saw this on slash dot or via slash dot. Somebody had actually noticed uh, that on the case that this problem is, they're They're not the brass tabs like this or they're still brass, but they're actually covered.

uh, black. They're completely uh, painted over in black paint. They're not gold like this, but apart from that, they're actually identical. retention Clips This one up here slides and they're the same shape and everything, but they're painted black now.

Uh, this person noticed that it had after uh, continual insertion and removal of the Kindle it had started. Some of the paint had started to wear off and was exposing some of the metal. Now, they actually measured the resistance between these two Terminals and they actually uh, measured something. Well, that turned out to be incorrect.

It was actually 2 megga ohms they measured, but apparently um I think they were touching the probes at the time, so there's there was no Um I've heard that there is no issue on actual resistance between these two terminals on the cover in question, and even though this isn't the Cas in question just for completeness, I will actually measure the resistance between the two terminals. Uh, on here and as you can see on this version, it's um, it's open basically, so that's what it should be on the real version as well. So there's no electrical contact between here. So the uh prevailing theory is that these contacts are wearing away or and or they're touching something internally to the Kindle Now we'll actually have a look at this.

I've actually taken my Kindle apart and we'll actually investigate whether or not these contacts, if they're totally exposed, can actually short something inside the Kindle which can cause it to, uh, lock up or reset or something like that. Now just to show you up close what this contact is actually like, this is the bottom contact here and as you can see, it's a curved piece of brass like this and the case under question is an identical one, but it's painted black so that's the bottom contact and this one here is the top contact and as you can see, it's brass also, but it, um, it slides like that to accommodate the Uh. that actually allows it to lock the Kindle to lock in place so that it doesn't fall out. So these are identical on the other case, so let's see if these can, actually, uh, short out anything inside the Kindle at all.

Now let's actually take a look at the mechanism itself. Okay, the Kindle goes into this bottom slot down here at an angle like that and then it clips into through the two slots on the side of the Kindle into this top one and as you can see, it can't fall out at all. No problems. It's kind of a neat little design I don't mind it at all.

and then to release it, you simply slide that back and it pops out like that and let's flip it over and take a good look at these particular Uh hooks in some detail, shall we? Now here's the bottom hook down if I can get that to focus. Here is the bottom hook down here and as you can see, there are four contacts. Now these contacts are the Um serial interface and there they are in the order. The top pin is uh, the Uh TX, the next one is RX, then ground, and then there's a mystery pin which will find out what that is in a minute.

Now this is the Rs232, Um Uart serial interface to the Uh internal monitor program for the main Um microcontroller in the Kel and um, there's been a few hacks, people. Um, accessing that sort of data to try and hack the Kindle. But that's what we're concerned about, these four pins in here, and the theory that we're going to try and bust here. That or confirm that the Uh, that somehow the metal, this big metal contact down here shorts out, Um, several, one or several, or all of those pins in there, or contacts something else to reset the Kindle.

Now the other Um one up here is just a sing single contact. Now if we take a look at this, okay, let's get some focus on that. There's actually a flex Uh membrane with a single connection with a single screw and then a single contact like that. So this one's very simple.

We've only got one contact to worry about up there, so we got four down, the bottom, one, up the top. Let's find out what they do. Focus on the bottom one here and thankfully, four screws allow us to completely remove this top connector and as you can see, the contacts are actually um, staggered. These ones connect to the PCB but the other ones in there are staggered so that it actually connects the ground and the other pin first and then the other two later.

And that's a common Uh concept inside connectors. Here's the actual board with the four contacts. These are the pads that uh, make contact with the Uh pins we saw and as you can see, this one is actually uh, grounded. this the um.

second pin. From the bottom, there is ground, so all this exposed, uh, goldplated copper all along this outer Point here is all connected to the ground so we don't have to worry about that pin which comes through the the actual retention clip from the case shorting out. Uh, that ground because it's going to, you know, touching anything in there. As you can see, there's nothing else in there at all for it to short to.

So the only thing it can possibly short to is physically inside these two connectors here inside these four contacts in inside the connector. So let's actually plug that up. Now if you see, this will actually flip directly over like this and it will plug in like this, this is what it will actually go in like when you plug it in. So you put it in like that if I can get it and then it flips around and it looks like it does make contact with all four of those pins.

I've got the connector in here Now let's see which pins actually make contact with the outer lug. sh We The bottom one does no problems at all. the top one does and so does RX but it looks like that's it. It looks like when it's plugged in like that.

uh, the top one TX The transmit pin as you'd expect, does not make contact because uh, that can actually drain the battery if, um, that's transmitting. If that pin is transmitting something out, you don't want to be shorting that to ground. So um, but that could happen. So we'll try that as part of our experiment to see if that causes the problem.

but we know when you plug this in, three of those bottom pins are grounded. Now to make our experiment a bit easier, what I've done is I've attached a wire to this terminal up here. Just allow us to do some um, handre actual clipping onto that terminal and a ground Point here as well. Now I Prove that this one is actually a ground Point by uh, measuring between here and the second bottom pin which as you can see, it's 0.

Ohm, So it's actually that's part of the grounded circuit. And we'll start out by just measuring the base states of all these pins on this connector. Okay, this one down the bottom is uh, almost 1.8 Vols Okay, the second one is obviously ground cuz it's connected up there. We proven that and the third one's also 1.8 Vols and the top one is 1.8 Vols as well.

and the Kindle is on. By the way, it's actually switched on. There it is and from experiments other people have done I Do know that the top two uh pins up here are transmit and receive. As I've said transmit, receive and ground the Rs232 uart uh interface here.

So I'm not going to concerned with those pins, but this bottom pin um I has me, uh, concerned I Want to know what this pin actually is? My guess is this bottom pin down here is actually an input. Now the reason I suspect this is because I reckon this metal contact down here makes cont with these pins down here and outputs of voltage on this top pin cuz we know it has to drive this Led. Now the Kindle's turned on so it must output a voltage across here and here. But we've already measured this pin up here and we're getting nothing.

so there must be some scent circuit which switches that on and my guess is that it's the bottom pin down here. Let's check it out now. I'm measuring the top output here relative to ground and we'll short out the two pins down the bottom and let's see what we get. Get on our output, shall we look at that? Bingo 3.9 Vol disconnect and there you go.

So obviously this bottom pin is controlling the output on here to drive the LED So that proves my theory that this contact down here shorts um, the bottom pin and ground, which switches on the LED output and just for completeness, let's use the oscilloscope here to see what the output is like. Is it a D or is it a DC signal? So let's just short the two pins out here and bingo there it is. 1, 2, 3.9 odd Vols Exactly what we measured on the meter. So it is definitely a Uh DC signal and let's see what happens when we hook up a white LED uh, in this case it's a lumy leads uh 1 watt LED directly up with no dropper resistor straight on the terminals and let's switch it on and see what happens.

Yep, it work works, but take a look, we're getting 250 milliamps Huge. Well, okay, clearly the Um output here. The switched output controlled via the input is uh, not current limited cuz uh, clearly the LED they're running up here is not uh, 200 milliamps I Would have expected 20 milliamps or something like that, but let's see what the short circuit current is. So I've got this on the Uh 10 amp range and let's switch the output on, shall we? As you can see, it shuts down pretty quick.

Let's try that on the milliamp range and it jumps right up there. and it's it. Looks like it's output short circuit protected, so no problems at all. It won't switch on and as you can see, it's still functional.

The whole Kindle still works, so not a problem, so it doesn't matter. Even if you short out, um, this output here, it makes no difference at all. In fact, if you actually hold it on here and look at the output, it actually hiccups. there it is, It switches on, then off, and it self- protects itself.

No problems at all. And now that we've proven that it can survive a direct short across the Uh output when it's on, let's try just shorting out all of the pins down here on this bottom connector just in case it shorts out the TX pin and all sorts of things. Because that's what some people claim is that these connections are somehow shorting out inside to something. So let's short them all out.

I've got some Al foil here, by the way, so let's put that on there across all the pins. you can't see that. but I've actually uh, I've shorted out all the pins there and does it still operate? Yes, it does. It hasn't locked up at all and that's what.

uh, people claim it resets, locks up. It'll take ages to resets. clearly not doing that. So it survives.

Shorted out all those pins, and even if we connect this output here. okay, let's go to the top here, hold this on here, and short it to all the pins down the bottom. Here, short it down to everything. It still operates, no problems at all.

And if you try all combinations of inputs down here. um, it doesn't matter what you actually do what you short out, it makes no difference whatsoever. all the combinations and all, and it still works so clearly. there is no issue at all with uh, these contact pins shorting out anything inside this.

Kindle So there you have it. I Think we've conclusively busted the myth that it's uh, this problem is being caused by these metal tabs wearing off the black paint and somehow shorting something inside the inside. The Kindle Granted, uh, we've only had a Um sample size of one here, but from a technical point of view, I Can't see how shorting out any of these, uh, any of the external connectors in here can cause a problem. We, you saw it.

We drew 250 milliamps quarter over an amp from this sucker and it didn't reset. it, didn't lock up, it didn't do anything. So that was a pretty worst case scenario. I Think we shorted all the pins out and as you saw the Uh clearance around the Uh the keyhole connectors here means that there's no way that those connect can touch any of the other circuitry.

So myth busted. It's not that. so what is it? Well I don't know. My next best guess would be: uh ESD electrostatic discharge.

maybe caused by the case or something like that coupling into the Uh contact terminals, but why? it only happens on the non LED version. this is the LED funky little pop out one. apparently doesn't happen on this. There's but the identical version of this with the black PL painted tabs it happens on Why I don't know I've tried some ESD stuff.

can't kill it at all. Um, but I need to do further investigation on that to try and figure it out. But I don't know. it's definitely not the tabs.

so I hope that's cleared that up. All those uh comments on slash Dot and everything else people thinking it was that sorry I don't think so I'll keep you updated if I find anything further.