Slashdot Mirror
Dangerous Apple Power Adapters?
An anonymous reader writes "Even with all these exploding Dell notebooks and other notebook safety problems, Apple has seemed relatively immune. Every once in a while, some odd thing came along, but it seemed like relatively calm waters. Not anymore — Apple's notebook power adapters appear to be the source of some serious safety concerns. Every iBook and PowerBook user should read this and keep a close eye on their adapter — the adapters suffer from very poor design including wires that seem prone to short out and burn and zero short circuit protection."
I don't think he's qualified to say that his adapter has zero short-circuit protection.
Here's what he says: Meaning, I can short the adapter on the DC side, generate a spark, and repeat again and again without causing the adapter to power off or any circuit breaker/fuse/GFCI outlet to cut the power.
He's expecting the wrong results. Sure, shorting any supply with output capacitors will generate a spark -- that's typical good design. The spark doesn't last long and it isn't indicative of the total energy released.
Now, if his circuit breaker or fuse triggered, I'd be concerned. That means the adapter is shorting out the mains voltage -- very bad, very dangerous. But, it apparently is not. It's good that this doesn't happen, but the guy seems to think it should. And a GFCI wouldn't trigger due to a hot-neutral short -- he would have to throw the adapter in a bathtub to have a chance of it tripping.
I'm not saying there are no problems with the adapter, but his assertion is unsupported by his evidence. I suspect that the adapter has an internal short-circuit protection that kicks in milliseconds after the spark is seen. He would need to use a current meter to detect if the circuit exists.
(why, yes, I'm an electrical engineer)
HIV Crosses Species Barrier... into Muppets
Damn, this pisses me off. The correct tag is "!fud", RTFM.
Just a technical addendum. The white bricks are no longer produced in the 45w range. All new white bricks are 65w. This is most likely because the latter g3/4 laptops were all requiring more power than the early G3s. The 65's work fine on all the older equipment, (watts are drawn on demand, a 65 won't hurt a machine that took a 45 originally) so we just carry the 65's. I have replaced maybe five white packs that had the wire broken at the strain relief where this fellow had the problem, and as many more where the wire went at the DC jack end. Apple does need to improve the strain relief at both ends. I find OP's claim that there was "no visible damage before the fire" to be laughable. When I look at the picture I note immediately, the wire always comes straight out of the pack when it's new, and there is a good inch of cord needed to bend it 90 degrees without excessive force. But when you look at the picture, the wire is almost emerging at 90 degrees right out of the strain relief. Good money says he tends to plug the pack into the wall a long way from his ibook, and the cord is always being strained and pulled hard to the side, and was a direct cause of the cord damage and the fire.
Also of note, the "ufo" power adapters that shipped originally on the ibook G3s are much much worse. They are known for failure where the DC cord meets the computer plug and where the AC cord meets the connector that plugs into the pack. We have replaced many of them for failure at one of these two points. Though for all the macs I've worked on, I have yet to encounter a single apple pack that caught fire. This sounds like an isolated incident and someone trying to make a whole lot of noise, stomping about and shouting "defect, recall, save me!"
Given 50,000 production units of electronics, a couple of them are going to be bad. There is no escaping that. And yes, one of them might burn down your house. But a meteor might hit it first, and has roughly the same odds, OP needs to get over it.
Though I don't deny he needs to post about it, because this is how you find out about real issues. Now if we saw a dozen "me too" followups immediately we might want to look into this more, but right now we just have a blowhard.
I work for the Department of Redundancy Department.
It turns that Kensington makes a pretty slick one.
The bits on the bus go on and off... on and off... on and off...
Networking over serial I/O was a reasonable choice for the time. Zilog had a chip that would do serial I/O at 230 kbps and Ethernet hardware was still very expensive.
Mea navis aericumbens anguillis abundat
This sort of behaviour is fairly common with small switchmode supplies - in fact, it's more likely if the power supply is well-designed and over-rated.
;-)
It's to do with the current waveform. Any switchmode supply tends to have a very spiky current load, as it switches on an off to keep the output voltage stable. A cheap switcher, if it's lightly loaded, will draw huge spikes of current only in the early part of each half-cycle - so it's current load looks just like one or two noise spikes, which get absorbed by any output filtering &/or ignored the protection circuitry in the source UPS/inverter.
A better switcher, on the other hand, will spread that current draw over the each half-cycle - so it's current load looks like a continuous noise hash to the supply. Enough hash to get back past any output filtering on the UPS / inverter and trigger the protection circuitry.
Hence the reason any decent UPS or inverter has specific warnings and / or deratings when used with switchmode loads.
(Yup, that's a simplified explanation - but it's also basically correct...)
In your case, it would probably work better with a smaller inverter, or a cheaper & nastier one without such good protection circuitry
What part of "a well regulated militia" do you not understand?