Dan's Data letters #79Publication date: 11 December 2003.
Last modified 03-Dec-2011.
I just purchased a new laptop, and I plan on using it for quite a while. Now, I know that batteries age - their capacity decreases over time. I was wondering whether or not I should buy a replacement battery right now while they're still available for my laptop model. Of course, that wouldn't make any sense if batteries aged even while not being used. So, my question is: can I preserve the capacity of a battery by putting it on a shelf and letting it sit there unused, or will it still age and lose its capacity as the months pass?
Since it's probably a lithium ion battery, yes, it will age, quite quickly. You shouldn't buy LiI batteries for future use.
Here's some more info about this:
Isidor Buchmann's site is a very useful repository of all kinds of battery information.
Windows disappoints yet again. Sure, it's a small problem, but it's annoying nonetheless... Picture it, Windows 2000 SP4 plus all the updates, AMD Athlon XP 3000+ on an Asus A7N8X, three HDs offering ample storage space, 512Mb reasonable quality Samsung memory, ATI All In Wonder Radeon 8500, and all the latest drivers, some software odds and ends (Office XP, Norton Internet Security '04, Winamp, some games, I don't know, DivX, you know, the usual).
But something is odd. Despite unprecedented stability thanks to good cooling and, I think, a UPS every few days I find a "~" file on my desktop, and I can't figure out what it is. Deleting it causes no harm but also doesn't take care of the problem; within days it's back. So what's a man to do but ask Dan? Much thanks for any advice or help you can offer.
I'm currently looking into purchasing an UPS/surge protector. I had a read of this article, and also this one, which enlightened me a little bit, but not really enough to feel confident making decisions for myself, or recommendations about this to other people.
Here's my problem: If, as you say, a lightning strike contains around 10,000,000,000 joules of energy, how is it that a mid ranged surge protector (which can only provide protection for around 450 – 500 joules) or a UPS (that protects me to about 320 joules) is worthwhile?
It struck me that a measurement in joules could only be relevant in the immediate vicinity of the lightning strike, which would mean I'm totally confused. So joules aside...
If a lightning strike is capable of producing 100,000,000 (one hundred million) to 1,000,000,000 (one billion) volts, and 35,000 to 40,000 amps of current, how is it that every time a bolt of lightning strikes a power line (I don't know how often this happens) half the state isn't calling the insurance company trying to get money back for every electrical appliance in their home?
Obviously in the case of a lightning strike, a lot of it is dissipated amongst to 1000 odd people on my particular electrical grid (not sure how many people would be on a grid?), so how much of it actually makes it back to me and my UPS/surge protector?
As you suspect, a lightning strike that hits power lines doesn't transfer all of its energy to the ground via one path. If it just dumped all of its energy straight into the electricity grid then it would indeed, as you say, cause widespread havoc. But this isn't what happens.
Power poles are, generally, designed to dump lightning to earth; the insulators between the power wires and the pole might as well not be there as far as the lightning's concerned, and the pole probably has a lightning-rod conductor running down it.
One pole's lightning rod doesn't get rid of all of a lightning bolt's energy, but a few consecutive poles will all do their bit, and thus the area of the power grid that'll suffer a dangerous spike can be surprisingly small.
It still pays to have good power conditioning hardware, though, if you live somewhere with dodgy mains or a lot of lightning storms. A whole-house system installed by an electrician is likely to be much better than any number of plug-in power conditioners, and "surge suppressor" power boards are likely to be pretty much useless.
If lightning strikes the power pole outside your house then it's unlikely that any power conditioning hardware will save you from losing not just all of your appliances, but most of your household wiring - but at least your house won't be blown to bits, as it definitely would be if by some quirk of fate all of a lightning strike's energy were released there.
I'm 29, and I can remember when watching He-Man in the mid-80s there was a commercial for a video camera that used regular audio cassettes. If I remember correctly, it only recorded in black and white with no sound. It was just for kids to play around with. I haven't been able to find any information about it; just wondering if you ever heard of it.
I'm trying to think out some ideas for a case mod project, and one thing I would like to do is to integrate a Ethernet router into a case to cut down on clutter in my work area. Can I just run a +5V line and a ground from a spare Molex connector (or better yet the PSU's standby line) into the router for power? It takes 5V/2A in.
The standby rail would be great, since it's powered whenever the PSU has power from the wall and not just when the computer's on. But it only has a small current rating, so it wouldn't cut it. A regular +5V wire should work fine, though.
I was under the impression that LEDs were the most efficient light emitters available on planet Earth. Based on this fact (that I heard in a physics lecture) I have pursued the technology for years, and have built my own devices based on LEDs.
As an example, I have a "torch" that consumes just 0.66W of power, and emits a red light vastly superior to the 21W bulb usually used in a car brake light. Car manufacturers now use LEDs of course for various purposes.
I also have a green light that I use for mountain biking, which uses around 4W and I believe to be as bright as significantly more power-hungry halogen lamps (30-40W range). It is bright enough for me to use with confidence at any speed as though it were daytime. This PDF document from Agilent (HP) describes the LEDs that I have used in my cycle light (HMLP-CM15). It clearly states that the luminous efficacy of the LEDs is 540 lumens per watt, which seems very high compared with every other technology.
I have spent the last five years buying very expensive LEDs, batteries, and other parts for these devices, and wonder if I'm actually barking up the wrong tree, as you seem to have implied in one of your recent responses that LEDs are no more efficient than an average fluorescent tube.
No, LEDs aren't ultra-efficient, though there are several caveats involved; "efficiency" is not as easy to measure as you might at first think.
Actually, technically what we're talking about here is its "efficacy", the amount of light you get per unit of input power. Luminous efficiency is, when you're being picky, different; it's the ratio of useful (visible) light emitted by the device to the total amount of light its photon-spitting element generates. The more light is out of the visible spectrum (infra-red in particular, for incandescent globes), or absorbed by the packaging around the photon-spitter and never makes it out of the lamp, the lower the efficiency will be.
In general terms, different light sources line up the same way as regards both efficiency and efficacy. The usual way of measuring the efficacy of a light source is in lumens per watt. The lumen, however, is a unit that's derived from the candela, which is weighted to match human vision, which has a sensitivity peak in the green spectrum. Move away from green up or down the spectrum and the same number of photons give you fewer lumens, and thus fewer lumens per watt.
That said, household triphosphor fluorescent tubes can manage an easy 80 lumens per watt, and maybe as much as 100lpw. "Warm white" and "daylight" tubes aren't as efficacious or efficient, but it's hard to find any kind of fluoro that scores less than 65lpw. The very smallest compact fluorescents might dip below 60lpw.
For comparison, a normal nitrogen-filled tungsten-filament light bulb is unlikely to manage more than about 15lpw, and possibly less than 10. Linear halogen bulbs - the rod-type lamps used in floodlights - can push to a bit more than 22lpw.
High and low pressure sodium lights are the kings; they're capable of more than 100 and more than 150lpw respectively. But they're "quite orange" and "very orange", respectively (those weird pinkish-red streetlights are low pressure sodium lamps), and not useful for general purpose lighting.
I can believe that your red torch looks great, though, for two reasons.
One, automotive (and ordinary flashlight) bulbs generally have quite abysmal efficacy and efficiency, partly because they're small (generally speaking, efficacy falls with size in filament globes), and also because they're built for durability rather than light output. A heavily suspended, thick-filament low voltage bulb just can't have good effic-anything.
Then, in the case of a brake light, you take the yellowish light from the bulb and pass it through a red filter that eats almost all of the photons. The final watts-in to useful-watts-out efficiency of this process, even if you don't weight the light by its far-from-green spectral content, is awful.
There's a pretty wide range of efficacies in high intensity LEDs (old-fashioned coloured-lens low-intensity LEDs all score very low), but you can count on better than 30 lumens per watt for most coloured high intensity LEDs. White 5mm LEDs can manage about 20lpw, at best. "Super-LEDs" like the Luxeon Stars typically do rather better, but not amazingly so; maybe 25lpw for white LEDs.
(More info here.)
Car manufacturers like LEDs because they're naturally red or amber and don't need filters, and because they're shockproof (handy for bike lights!) and, practically speaking, never burn out. This is also why LEDs are being used in traffic lights.
Green LEDs like your bike light get a boost in the lumens per watt stakes, by virtue of being close to the 555 nanometre frequency that the candela unit's based on. And, of course, this does mean the light from them is easier to see - and the beam from most LEDs is a nice wide even pool, more useful for seeing where you're going than the uneven illumination from most bulbs in reflectors.
A halogen bike light, however, probably has a lumens per watt score about half that of the green LEDs, so four watts of LEDs is good for only about eight, maybe charitably 10, watts of halogen. Any further difference you see comes from light distribution, not actual brightness.
So what's the deal with the 540 lumen per watt Agilent LEDs? I can only presume that they're quoting figures for very low currents, at which many high intensity LEDs are more efficient than they are at their rated current (or when overdriven, as they often are in lighting applications).
I've been playing with electronics as a hobby for about eight years now, and have learnt that cathode is the negative terminal, and the anode is the positive terminal, for diodes, etc. I also am aware that electrons flow from negative to positive, which to me means from cathode to anode. We've started the electrochemistry topic in class now, and the teacher has told us that in a galvanic cell, the anode is negative, and the cathode is positive. This really confuses me, so I Googled around, and found all the chemistry sites say the anode is negative, but the physics and electronics sites say the anode is positive.
So I'm wondering, which is it? does it depend on the cell chemistry?
In a cell or battery that's delivering current, the cathode is where reduction is taking place, sucking up electrons and making the terminal positive, and the anode is where oxidation is taking place, spitting out electrons and making the terminal negative. Things that aren't delivering current are, as you say, labeled the other way around; positive anode, negative cathode.
The nomenclature was set up this way by Benjamin Franklin, specifically to make learning about redox and such in high school more fun.
You can read all about the fraught issue of current direction, which is what led to this silly situation, in various places. Like this Usenet thread, for instance.
While browsing for various torches and spotlights suitable for possum hunting, I came across the Megaray (homepage here). Once I got over the initial awe and controlled my drooling, I started thinking about the battery it uses. You don't often hear of lithium phosphate batteries, but they sound very impressive at less than a quarter the weight and volume of a similar capacity sealed lead acid battery, according to the Megaray review.
What sort of "civilian" equipment commonly uses these batteries, and can you buy them batteries separately for a sane price? I have read about tablet PCs that use them, but I was thinking of a 12 volt arrangement that would be happy to power a 100 watt halogen spotlight. It would have to be a huge improvement on the common 7 amp-hour SLA brick that most people use, especially if it was better suited to deep discharges, which the SLAs don't seem too happy about. Then again if they really are a quarter of the weight and volume, you could just have four times the capacity and only use 25% of it, I guess.
Lithium phosphate batteries are pretty new, and I think Valence Technology owns all of the relevant patents.
They've shown up in a few portable computers, as you say, but have yet to make any real impact in consumer devices. You can't buy them off the shelf, and I don't know whether they're touchy about charging or not.
As far as current delivery goes, lead acid is quite adequate, provided you don't mind lugging around a brick of a battery. 100W from a 12V gel cell means a theoretical 8.3 amps - in real life probably a bit less if the lamp's rated for 100W from 12V, but the battery's sagging a bit under load. Anyway, one 7Ah brick will deliver that without screaming too hard, especially if you're just using bursts of ultra-flashlight light; a couple of bricks in parallel will probably be quite happy, as long as you don't run 'em dead flat or charge them while they're still hot. Ideal current for cheap 7Ah bricks is more like one to two amps, but 4A each for a pair ain't horrible.
If lithium phosphate manages a quarter of the weight and volume of lead acid, then they're probably managing 160-odd watt-hours per kilogram, which is good (up there with lithium polymer), but not incredible. Since they're new tech, they're probably too expensive to compete for general applications yet.
Regarding running laptops on planes from lead-acid batteries - I have a feeling that airports don't look too kindly on people bringing highly corrosive acid that has the slight possibility of exploding onto their airplanes.
Doing a bit of digging, the Qantas Web site has a brief outline of what they consider dangerous goods here and here; they say nothing about lead acid batteries, but lithium batteries are allowed as long as you only take two. Cos, you know, I'm sure there's a Web site out there that tells you how to make a thermonuclear warhead out of three and exactly three Li-Ion batteries.
If you want more info, you can buy the book for $US60 here, but I'm understandably not THAT curious to know.
Airlines and airports ban all sorts of weird things these days. I wouldn't be surprised if SLA batteries were on the list for one or more of the rent-a-cop pseudo-security outfits, like the one that confiscated my girlfriend's minuscule sewing scissors on the Sydney-to-Canberra flight (that's a notoriously terrorism-prone corridor, of course...) that she took the other day.
Realistically speaking, though, even liquid sulfuric acid is not the horror-substance that people tend to think it is. You don't want to get concentrated H2SO4 on you, and you certainly don't want sulfuric of any significant concentration in your eyes - but if you do splash yourself with battery-acid H2SO4 (which is only about 35% concentration), you needn't run through the screen door without opening it on your way to wash the stuff off.
People with tough skin commonly fail to notice battery acid splashes entirely; the first they know of it is when they see the hole in their jeans.
And SLA acid is gelled, which makes it even less dangerous. It can't soak your clothes or flow down your leg.
Hydrochloric acid is another relatively non-terrifying compound, and they can't ban that from planes; everybody's got some in their stomach (despite modern medicine's best efforts), and it can be strong enough to eat a hole in the carpet at the scene of a stabbing.
I'm going to America in two weeks, and I am wondering what sort of power sockets they use, as I am from the UK. I need an adapter for my hair dryer.
You may find this site helpful.
Note that your 220/240V hair dryer will not work properly from American 110/120V (going the other way is dangerous; running a 220V gadget from 110V, at least, doesn't often result in a fire). To run the dryer, you'll need a step-up voltage converter. One of those capable of running a hair dryer will be big and heavy and probably cost you as much as a new hair dryer would, so I suggest you leave the dryer at home, buy one at your destination, then give it away when you leave.
Any appliance that doesn't actually have a sticker on it that says something like "100-240VAC 50-60Hz" will need a voltage converter to be used in a different country. This covers every cheap appliance, even things like corded electric razors that draw very little power.
All sorts of electronic gadgets these days are coming with multi-voltage power supplies; modern laptop computers and mobile phone chargers, for instance, are generally fine everywhere. But ordinary little appliances with motors and heaters in them work only from the voltage they were designed for.
If you were coming to Australia, where I live, you'd be fine with just a plug adapter; we've got the same AC voltage and line frequency as the UK uses. I doubt you're willing to change your travel plans on just these grounds, though!
No, they're useless.
Some people swear that fancy power cables work miracles, but some people also think that Elvis is alive, Keanu Reeves is a good actor, and squirting coffee up your bottom will cure cancer.
Fancy power cables are just one of many magic snake oil products created to milk more cash out of the more lunatic kind of audiophile. Some of the fancy power leads may have electrical properties significantly different from those of ordinary $2 leads, but those properties do not and can not have any effect on anything, unless the mains supply is very, very weird and/or the equipment power supply is very badly engineered. Needless to say, nutty audiophiles tend to use gear with power supplies that NASA would consider over-engineered, so the whole concept of super-special power cables to go with them is completely detached from reality.
And, in any case, there's nothing the fancy leads can possibly do that a plain lead made from heavy gauge cable with, possibly, a couple of filter components built into its plug (total parts cost: $US10, tops) wouldn't do.
In that picture of you at the top of the photo tutorial - does your t-shirt have the old Amiga Guru Meditation error message on it?
Yes. That shirt's an Errorwear product.
See also this.