Dan's Data letters #191
Publication date: 16-Oct-2007.Last modified 29-Sep-2012.
Tiny torches
Enough of your insanely informative and in-depth LED torch reviews with your power consumption graphs and your battery life calculations.
I just want one question answered:
What is the best, commercially-available LED-based equivalent to the Mag-Lite Solitaire?
I want a torch that's small enough to comfortably put in your pocket, bright enough to find the back door's keyhole/spot that thing you dropped behind the TV/read a map in a pinch/see your way up unfamiliar stairs/find your way to the generator when it's out of fuel and there's zombies bearing down from all sides, and feels like it could get dropped from a height without it really noticing.
And I want it to be a nice neat little cylinder I might mistake for my bullet pen in my pocket.
What say you?
Thom
(Frustrated burnt-out-bulb Mag-Lite Solitaire owner and fellow Australian.)
Answer:
Mag themselves finally started selling
LED flashlights this
year, but they don't as yet have a Solitaire equivalent. The smallest LED Mag so far
is a version of the very popular,
but a bit too big for pocket carrying, 2-AA-cell Mini Maglite.
You can, however, get third-party LED lamps that fit the tiny Solitaire. Here's one that sells for about $US20. If your Solitaire's OK except for the bulb (even when they're working exactly as designed, the tiny 1.5 volt Solitaire bulbs have never been very good), then that could be an idea.
M'verygoodfriends at qualitychinagoods.com, who provided the lights for my most recent review, currently have three single-AAA-cell lights. This one looks pretty neat, but it's bulkier than the Mag Solitaire types. Still, it's only $US21.28 including delivery.
Personally, I still use an ancient Arc-AAA. I think the current Arc-AAAs (the company went through a bit of bankruptcy and rebirth drama) are perfectly OK, and of course rather brighter than the old ones, thanks to the great strides white LED technology's made over the last few years.
The Arc Flashlight online store isn't very attractive to Australian shoppers, though, because it only offers stupidly expensive overseas shipping options. But you'll still be able to get one light for less than $US50 (or three for less than $US100...). That isn't too bad, seeing as the thing'll probably last forever.
Peak LED Solutions' current products look pretty neat, too; they, like Arc, sell lights that still look pretty much like the ones I reviewed years ago, but are now brighter and get better battery life.
Peak have a bunch of cool lights hiding behind their disastrous (and misspelled!) Flash site navigation. The "Matterhorn" is their Arc-alike single-AAA-cell light; I recommend you get the one-LED version rather than the 3-LED one if you don't intend to use a lithium or rechargeable AAA.
Peak don't have an Australian dealer either, but their site appears to allow you to pay a mere $US15 for international delivery of one light, which is OK.
Not compensating for anything
I read with interest your latest piece on LED flashlights, and was wondering what you thought the BIGGEST (light-wise) cheap-ish flashlight would be at the moment?
I'm not really after something super practical, I'd love to buy a Tesla-6 but $US160 is a little much. Lets say the brightest possible for less than $AU100.
Bryn
Answer:
I think it's hard to go past the "Ultrafire"-ish lights
at the moment. Qualitychinagoods.com
have several lights with ten to 18 watt power ratings, which are probably a bit on the
optimistic side (I'll get some for review and see...) but still really can't be beaten,
for well under $US50 including delivery.
Even if you have to get 18650 lithium-ion batteries and a charger as well, you'll still end up well under your $AU100 limit. You may be able to find the same things cheaper on eBay, but probably not much cheaper.
Alternatively, you could get a cheap "Dolphin"-type lantern (or Mag-type flashlight) and swap a high powered LED lamp into it. The usual lamp type in these kinds of lights is a "PR"-type flanged bulb, and you can now get LED replacements that can handle a range of battery voltages. You'll want a lamp that can handle six volts for a standard lantern battery, or nine volts if you get one of the giant six-cell Maglites.
The only problem with this approach is that high-powered LEDs can be hard to heat-sink through the meagre metal-to-metal contact of a PR bulb socket. Incandescent bulbs are meant to run hot, but LEDs don't like it. This means a lot of PR-type LED bulbs are pretty dim - they give you a ton of battery life, but less light than the stock filament globe.
It can still be done, though. Terralux, for instance, make a range of LED PR-type lamps (I reviewed a rather odd one here, and their old flagship lamp here). Their current 6-cell-capable Maglite bulb is the MiniStar5, which retails for only $US25.
(Incidentally, Elektro Lumens no longer sell the enormous Tesla-6 I reviewed back in 2005. As I write this their current flagship light is the far brighter Stunner-P4. It can be yours for a mere $US499.99!)
The PC-V1000's little friend
I have read your Lian Li PC-V1000 case review and thought it was excellent. I also read your PC-S80 review and thought that it was excellent, too. I decided to sell my own S80 in favour of a V1000 for the best cooling performance, and of course chic looks for my new system.
I'm also interested in the PC-V600, because I'm building a computer for my girlfriend and she liked the V1000 but said it was a bit big. I'm a little worried about the available space between the CPU and the PSU inside the V600, though.
Aaron
Answer:
The layout of the PC-V600 is indeed...
...peculiar. The fan on a bracket at the top of the case is hanging over the expansion slot area; the power supply mounts...
...over the top of the rest of the motherboard area, including the standard CPU socket location.
I'd deal with this by getting a low profile CPU cooler, and making sure the PSU you choose doesn't have a fan that'll fight the fan on the top of the CPU cooler.
This may be overkill, though. A stock CPU cooler actually ought to fit fine in the V600, and will of course also work fine if you're not planning on massive overclocking.
It seldom makes any significant difference whether you install the fan on a CPU cooler so that it blows or sucks. So if you get a PSU that has a fan facing the CPU cooler and sucking air out, just install the CPU cooler fan so it sucks air up into the PSU fan, and you'll be fine.
(You'd probably be fine anyway, but I'd rather not have fans fighting each other if it can be avoided.)
There's more discussion of this issue on the HardOCP forums here.
As they say in that thread, practically no after-market coolers will fit in this case, because they're almost all taller than the stock units.
Zalman make some decent slimline units, though. I think something like a CNPS8000 (which Australian shoppers can buy from Aus PC Market for $AU77 delivered) might do the trick, but I don't think you can flip its fan. The CNPS7000 has the same problem.
There are several other models of slimline CPU cooler out there, but most of them only perform about as well as a stock cooler. They're aimed at the rackmount server market, which needs to be able to cram ordinary levels of CPU cooling into a pizza-box form factor.
But, as I said, there's probably no reason to worry about fancy CPU coolers if you're not overclocking quite a lot. A stock cooler should fit and work fine.
Don't forget the tartan grilles
I was wondering if you could help me out with getting started on building speakers. I work in the AV industry and I have sold all kinds of speakers... Boston Acoustics, B&W, Aerial Acoustics, Definitive Technologies, Totem, ad nauseam. My favorites are the Aerials but I had to sell mine recently and I miss them dearly.
So I thought it would be fun to build some kick-ass high quality speakers like my Aerials but, you know, better because they're "hand made". Ha.
I have been having trouble finding suppliers and (more importantly) reviews of quality drivers. I know more or less what to look for in a driver but not from which companies.
Can you help?
Jonathan
Answer:
It's been a while since I've built any speakers from scratch (God, it must be more than
ten years now), but I built a few pairs back then.
The Big Secret Nobody Wants You To Know is that driver quality is really not terribly important. There are lots of simple honest paper- or polypropylene-cone Chinese bass drivers that are perfectly fine, and far better value than the European-made drivers. Even if you want something exotic like ribbon tweeters, you can often find 20-year-old New Old Stock units for good prices.
The problem with cheap drivers is consistency, and accuracy of specifications. The stuff on the spec sheet may be perfectly accurate, or it may be completely fictitious, or it may only match one driver from the first production run that was made in a whole different factory from the drivers you just bought.
Fortunately, it's not terribly difficult to figure out the Thiele-Small parameters for a given bass driver from first principles yourself (I'll wait while you go and buy the Loudspeaker Design Cookbook), and after that you can tweak woofer/tweeter balance by ear. Things get more complex if you're building a system with a midrange driver in it, but that's seldom necessary in these days of small bass drivers and relatively widerange tweeters.
If all this sounds Too Damn Painful, you can short-cut to probably-accurate specs by buying Euro-drivers from, say, one of the companies now under the Tymphany umbrella (ScanSpeak, Peerless and Vifa are all Tymphany brands now). You may get good accurate specs if you buy something like Parts Express's "Dayton" house-brand drivers; I don't know for sure.
As far as designs go, I recommend your first speakers be straightforward medium-to-large sealed box units, because those are easy to design.
If you're then looking for something more challenging, allow me to recommend a transmission line design. You don't see TLs much in commercial speakers, because they need a lot of wood. They're not terribly difficult to build at home, though, and can sound spectacular with very inexpensive drivers (or even better with a fancy Fostex widerange).
The Mantovani-only car stereo
My girlfriend has a 1999 Honda Civic with an aftermarket Sony CD/MP3 player.
She also has had 4 Sony Xplod speakers installed. They're a little bigger than a CD in diameter.
For some reason the audio has little to no low range. I've tried all the standard EQ preset settings like "rock", "dance", etc, but still cannot seem to get a semi decent bass sound out of the speakers. It's so bad you pretty much can't hear any instrument that would be considered bassy. It's like half the instruments in the song are not there.
I have the same speakers in my car's rear speaker pots, and I get quite acceptable sound right across the whole audio range.
My Panasonic CD/MP3 player says it outputs 200 watts. My girlfriend's Sony CD/MP3 player says it does 52W X 4, so it's pretty much the same in wattage output.
I'm curious if this could be caused by the speakers in her car having the wrong polarity. Would the wrong polarity cause an issue like this?
Another theory I've got is that my rear speakers bolt onto this thin but flexible carpeted chipboard-like material that covers the space between the back seats and the rear window on my sedan. In her hatchback, the speakers bolt onto plastic. Is it possible I'm getting extra low range resonance from the material spanning across the rear of my car?
Sam
Answer:
I don't know why this is, but I've got a few guesses. In decreasing order of probability:
Most likely: Perhaps some of the speakers do indeed have the wrong polarity. This is technically referred to as being "out of phase". If one side is connected with the red/striped/whatever wire going to the plus/red/whatever terminal on the speaker and the other side is wired the other way, the speakers will indeed cancel each other out at lower frequencies.
(If every speaker is connected with the same wires going to the same terminals, there'll be no problem whether you decide to go with striped-wire-plus or striped-wire-minus. They all just have to be the same.)
Out-of-phase speakers create a very distinctive sound, reminiscent of the way music sounds when you've got a heavy cold and one side of your head's more blocked up than the other. The acoustic environment in a car is usually so awful that this can be camouflaged, though - the fact that the listeners are usually sitting way off-centre doesn't help.
Less likely: Perhaps the speaker mounting locations are little sealed enclosures with very little volume.
The back of most car speakers face some random-shaped leaky enclosure, like the inside of the door assembly or part of the chassis frame. If the speakers are instead mounted in what is, effectively, a very small sealed box, then you won't get much bass, because the speakers will be trying to compress and expand the small springy air volume, which is hard work.
I don't know whether any cars are actually like this, but as the frame gets filled up with spaces for side intrusion bars and curtain airbags and so on, I suppose it's possible.
Even less likely: Perhaps the mount is too porous, so the speakers are behaving as if they're just hanging in the air, not really mounted on anything at all.
Again, this'll produce little bass, this time because the rear wave from the speaker will leak straight around to the front and cancel out lower frequencies.
It shouldn't be possible for this to happen, but I guess it could if, for instance, the speaker installer used screws that were too long, couldn't tighten them all the way down, and thus left the drivers hanging in front of the panel with a little gap all the way around the edge.
Less likely again: Perhaps the CD/MP3 head unit has an integrated crossover option, that deliberately filters out the bass for the regular speaker outputs and sends bass only to a subwoofer output, which in your case is not connected to anything. It shouldn't default to this configuration, though.
Virtually impossible: Perhaps the speakers themselves have crossovers on them, for a similar reason. These would probably be an obvious little circuit board on the back of the speaker. I'd be pretty surprised if this was the case, but it's not impossible.
Bugs bugs bugs bugs bugs bugs...
It appears that my landlord has been systematically spraying my apartment complex, with the intention of eradicating all insect life within. I say "with the intent" because in practice it has just herded all the buggers down into one apartment. MINE.
Now I have roaches everywhere, in everything. I know how to KILL them and get RID of them, but not the stinky oily residue that they exude into their favorite hiding spot, the one that's INSIDE my Macintosh and video game consoles.
How, oh how, can one go about cleaning bug excrement off of a computer mother board?
I've thought of taking it apart, methodically shorting each capacitor with a 1Mohm resistor then gathering a large amount of very low mineral content water and rinsing things off. But I have no ideas about using soap, or if this is even going to work without scrubbing the surface mount parts off the boards. Are there any computer-safe solvents I can use to clean things up, or am I just stuck?
BTW did you know that German cockroaches are small enough to hide 5 adults under the plastic of a PCI slot?
Echo
Answer:
It'd be pretty safe to squirt everything with
denatured alcohol ("methylated
spirit", here in the Commonwealth) from a pump spray bottle. Just keep an eye out for
things like stickers, which'll drift if soaked with alcohol. They usually don't end
up anywhere dangerous, but may be more important than you'd think. If you're cleaning
an old computer, for instance, it may contain UV-erasable
EPROMs, whose erase window is usually
covered with a sticker, which you should replace if it falls off.
There are lots of other PCB-cleaning chemicals, but some of them are ferocious plastic-eaters and very bad to inhale (like 1,1,1-trichloroethane, also known as the thinner used in old-style Liquid Paper). Other PCB cleaners arguably smell worse than cockroach doo-doo anyway. No purpose-made PCB cleaner is anything like as cheap as denatured alcohol, which you can get from any hardware store and most supermarkets.
Alcohol is hygroscopic (as I've mentioned before), and denatured alcohol always has some water in it anyway, so there'll always be a little bit of water left when the alcohol evaporates. This isn't a problem, though; just leave the computer parts somewhere warm and breezy (outdoors, if possible) until every cranny's dried out. And, duh, don't leave them near an ignition source.
Water - even plain tap water - with a few drops of detergent in it will do a pretty good job too. But alcohol evaporates faster, won't corrode tin- or copper-plated connectors, and cleans most oily crud better.
I wouldn't worry too much about manually discharging capacitors. Just wear rubber gloves if you're working on anything live, like a PSU. (Gloves aren't a bad idea for a job like this anyway, of course.)
The usual disclaimers apply regarding CRT monitors, though; see this RepairFAQ page regarding safely discharging those.
And they make that Jetsons-car noise, too
In two to three years we could all have non-toxic, non-heat producing betavoltaic batteries with run times on the order of years?
This sounds too good to be true.
Joel
Answer:
You've really got to wonder how hard and how often the person who wrote that Next Energy
News
piece was dropped on his or her head as a child.
All they had to do was look at Wikipedia's page about betavoltaics, and they'd realise that their strangely-capitalised-and-poorly-punctuated "Although betavoltaic batteries sound Nuclear they're not" claim is complete bunkum.
You can't get the beta particles that these sorts of "batteries" require without a decaying radioisotope. Or, if you're a nitpicker, some very dramatic electric charge movement, like lightning. Artificial generation of lightning in order to run a betavoltaic device and make electricity would be something of a W. Heath Robinson approach to the problem.
Betavoltaic devices usually run on tritium (the article seems to think they generate it...), which in quantities larger than those contained in the average luminous watch is a pretty highly regulated substance, on account of how it's your basic fusion bomb fuel.
And, like other "nuclear batteries", betavoltaic "batteries" cannot be switched off. Their radioisotope decays all the time whether you're using it or not, and the energy from that decay has to go somewhere. Where most of it goes, whether you're using the battery or not, is straight to heat, keeping the battery toasty warm all the time.
Nuclear batteries of one kind or another are tremendously useful devices. The established kinds - not betavoltaics - have spectacular energy density, orders of magnitude better than chemical batteries; that's why they're used to power things like space probes, and weather stations in remote areas. The plutonium-based radioisotope thermoelectric generator type was even used to power early pacemakers!
But I strongly doubt you'll ever see a nuclear battery of any significant size in a consumer product. Even if you could somehow pacify the various regulatory agencies, consumers are far too terrified of radiation to buy such a thing. I doubt many people would have even accepted a plutonium-battery pacemaker if the alternative hadn't been death.
Oh, and as Rupert Goodwins points out, betavoltaic materials currently have a number of other shortcomings besides the fact that, contrary to the Next Energy News piece, they do indeed run hot. They tend to have a rather short lifespan, for a start. And, apparently, the best energy density anybody's actually managed to get out of a betavoltaic battery so far is less than one seventieth of the energy density of current lithium ion batteries.
Oh, and as with other radioisotope generators, the output of a betavoltaic battery depends on the amount of radiation its beta source is producing. As the beta source decays away, the output of the battery falls. So it'll either no longer have enough juice to power your laptop after five or ten years, or start out producing rather too much power and running really hot as a result.
The Magic Water Stick™
Recently came across someone who was convinced that by stirring a glass of water using a metal stick it would somehow improve the taste and "quality" of the water. They mentioned that it was called the "Grander method" and after a bit of Google searching I found this site.
I started reading and it seemed a fairly classic example of a con job. Some of the stuff on that site is truly hilarious. They don't even try to explain how it's all is supposed to work.
I'm still giggling to myself over it a few hours later.
Worryingly enough, an Austrian holiday house purports to use this "invention".
Have you come across anything like this before?
Moreover, why oh god why are the Austrians being taken in by this? I was sure it'd be some of the more gullible Yanks!
Matt
Answer:
It's remarkable just how much quackery and pseudoscience is directly related
to water.
It serves as something of a marketing master-class. Long before people were being conned into paying for allegedly delicious springwater in plastic bottles, umpteen cranks and scammers were theorising frantically about "structured", "clustered", "magnetised" or otherwise "energised" water, which can command an even larger mark-up than Dasani.
Stephen Lower's Web site is an excellent resource for all of this. Grander is mentioned specifically on this page.
There isn't really much difference in gullibility between the nations of the world, as far as I can see. Different countries may get suckered by different kinds of scam - for instance, homeopathy is huge in Germany and I presume also Austria, and almost nobody outside Korea believes in the very hilarious "fan death" - but poor critical thinking is a worldwide problem.
