Dan's Data letters #110
Publication date: 11 June 2004.Last modified 03-Dec-2011.
Compaq concealment
I just read your review of the Lian Li PC-V1000 case and I was wondering: Can I take the motherboard and power supply out of my Compaq Evo W6000 workstation and install it in this case?
Sean
Answer:
I doubt it. One look at the shape of the motherboard in the
parts guide
makes clear that it's not at all ATX standard. This is to be expected, from a proprietary machine; manufacturers like
Compaq and Dell like to lock their customers in to using only their own expensive components.
Volts from nowhere?
I'm attempting to settle a dispute between a colleague and myself. The question presented is when a battery (various types), is discharged to a state where the appliance which uses the battery does not work, do chemical reactions/heat work to charge the battery?
For example, say I'm using a drill, and the battery runs flat, I come back an hour later, and push the trigger on the drill, the drill bit moves, but obviously slowly and not enough to do anything worthwhile. What causes this? If you were to leave the battery for, say, a decade, would it be charged?
Mat
Answer:
No, batteries don't charge magically when left alone. But batteries do recover terminal voltage when they're
unloaded after a significant load. The reasons why this happens are complex (translation: I couldn't be bothered looking
them all up), but it has to do with the basic chemical reaction involved (only so much of which can happen per unit
area of cell electrode), and the temperature of the battery (which will rise under heavy load; the warmer the battery
is above a generally quite temperate optimal point, the fewer volts it'll present to a given load).
When the battery's unloaded (or only lightly loaded), the chemical reaction gets time to recover the normal equilibrium state of the positive and negative electrodes, and the battery temperature falls. This is why a flashlight with "flat" batteries will give you a brief pulse of light if you turn it off for a while and then try it again, and it also explains your drill example.
This phenomenon is strictly limited. It will never charge the battery (in order to notice the phenomenon at all, the battery has to be pretty close to flat), and, in time, all electrochemical cells will self-discharge to nothing. This will take several years for quality alkalines, rather less time for NiMH and NiCd rechargeables, and rather more time for lithium cells, but it'll happen to every kind of battery, eventually.
Hi-V LEDs
I recently got hold of some no-name white and blue LEDs from an electronics shop in Pasar Road, Malaysia. The retailer claims that they're 5V models. I thought there's only 3.5V white LEDs. I hooked them directly to my PSU's 5V output without any resistors and they ran perfectly fine. I also found out these LEDs draw some 50mA, which stayed the same until I turned them off. No resistors, no heat sinks, very bright and costs 1.5 ringgits (40 US cents) each. Maybe I got ripped off.
Adam
Answer:
There's no such thing as a 5V LED, in the simple discrete LED market, as far as I know. There are, however, LEDs with
pre-attached external current limiting resistors; perhaps you can also get ones with internal resistors, and that's
what you bought. Ordinary blue and white LEDs are, as you say, rated for 3.5-ish volts.
You can also buy flashing and multicolour LEDs which have extra (exceedingly small!) electronics built in, and which run from a higher voltage than the dies inside them need. Those LEDs also don't need a current limiting resistor, if you just hook 'em up to their rated voltage directly.
50mA is on the high side for a single normal LED, but not amazingly so. A 3.6V blue LED running from 5V will draw quite a lot more than 50mA in the few seconds before it goes up in smoke.
Low cost lasers
I've always used inkjet printers at home because laser printers were too damn expensive. However I've noticed that it's now possible to get a laser printer for under 300 Australian dollars - what's the story with that? Are they real laser printers? How are they suddenly so cheap?
David
Answer:
Inkjets of all types are only cheap to buy, not to run. If you're printing a lot of pages, a laser will work
out far cheaper than inkjet (dot matrix is likely to be substantially cheaper again, but at a large price in speed,
quality and noise...).
There've been cheap laser printers for quite a while now - about ten years, actually. They're all "dumb" printers with practically no on-board processing power or memory, but that's how pretty much all consumer printers are these days.
The first of these laser "Winprinters", that lean on the host PC to do all their thinking for them, weren't great solutions. But the surfeit of processing power in modern PCs, plus the considerable data transfer advantages of ubiquitous USB, mean that modern dumb printers are only a problem if you're running an operating system for which there's no driver for them.
Some of the early cheap lasers also had excitingly expensive, low capacity toner cartridges, but that landmine's not as common these days either. You may get a reduced-capacity cartridge with the printer, though; check to see if that's the case, and factor in the price of a proper cartridge.
Cheap dumb printers are still a bad idea if you need a bad-ass workgroup laser that'll be pumping out pages all the live-long day, but for home use they're very definitely superior to inkjet printers, if all you need is black and white.
Fan fixing
I just wanted to drop you a line in thanks for your very handy article on fan maintenance. An old GeForce 3 of mine was just starting to buzz and I figured I'd have to replace the fan on it, but before I tried that I figured I'd follow your advice and it seems to have worked perfectly! It's sounding so much better now!
What I'm wondering though, is how long can I expect it to last? Has its previous lack of lubricant damaged the fan? Should I just keep an eye on it and hit it with a bit more 3-In-One when it starts acting up again? Or should I look at this as more of a "OK, now you've got another month or two, buy a new fan" thing?
Michael
Answer:
What state the fan's in now depends on how long it was running dry before. The bearings in small high speed fans like
the ones on video cards are likely to chatter themselves to death pretty quickly, but "pretty quickly" is in comparison
to larger fans that may buzz away at less than full power for months on end before actually seizing. There are several
hard-to-quantify variables involved, but you should be safe if you just keep an ear out for renewed buzzing. If the
fan fails completely, system crashes will probably alert you to the problem before permanent damage is done; modern
silicon that's lost its fan but still has a heat sink on it generally doesn't actually barbecue itself. It just gets
too hot, screws up, and causes a system hang that then allows it to cool down again.
Tank maintenance
I have a Tamiya tank, the Porsche turret King Tiger. The problem I have is that too much hard use over building sites has ruined the gears; the nylon gears have had their teeth compressed and now turn pretty freely. Where can I get just the gears themselves so as I can get it up and running again without the need to buy completely new gearboxes?
John
Answer:
I don't think you can get the gears separately - at least, not easily. According to the parts listing (in the back
of the manual, or online; this page has PDF parts listings
for lots of models, with the Porsche turret King Tiger here),
this model, like the other tanks, has the gearboxes as one lump.
This doesn't mean you couldn't find matching gears somehow somewhere else, but you'd have to hunt around. There may be a perfect match somewhere in the Stock Drive Products metric parts catalogue, but from my point of view, the issue of finding that match is surrounded by an SEP field.
Faster flashes?
I am the proud owner of a Canon EOS-10D digital SLR and a 550EX flash, but, as you may know, this thing positively EATS batteries, and as bad as it is, the MT-24EX is even worse.
So I thought it would be cool to acquire an AC adaptor for studio work.
But, what do you know, Canon makes no such beast. Furthermore, no one at Fred Miranda seems to know of such, and a fairly intense Google search has produced nothing.
Any thoughts? Heard of anything? Or could it be whipped up in the laboratory?
Christopher
Answer:
I haven't heard of any commercial mains power products for these flashes. There are commercial large-capacity
battery packs for all kinds of flashes, and Canon make a couple of packs of their own as well, but I think people
who want a mains-powered flash are expected to shell out for a separate
studio strobe. Which, to be fair, isn't a terrible idea; it's not
difficult to get studio strobes to sync to DSLRs these days, and there are lots of studio strobe packages that start
out rather cheaper than the 550EX but generally deliver more light and faster charge times.
But you've got your strobes already, and presumably are not champing at the bit to clutter your life with more of the darn things. So a bit of fooling around with my bench power supply and Ye Olde Pair Of Multimeters revealed that the 550EX will charge, eventually, from pretty miserable voltage, but unsurprisingly you need a low-resistance power supply delivering something like 4.8 volts to get a reasonable cycle time out of it.
With some PSU knob-tweaking to compensate for the extra resistance of my inline ammeter and gimcrack cabling arrangement, I figured out that the thing needs a maximum of almost three amps while charging - so, to have a safety margin, you'd want at least a four amp power supply, and probably five. A regulated 6V 5A PSU would probably give you very nice results, though you'd want to be careful about shooting like crazy at high power levels; flashes that're meant for battery powered use don't expect to be flashing a whole lot more frequently than they can from NiMH AAs.
You're unlikely to find a plugpack power supply that can deliver enough current for this application. A bench power supply could do it, but would be neither cheap nor elegant - if someone bumped the voltage knob you could blow up your flash. Personally, I'd strongly consider a dirt cheap six volt sealed lead acid battery or three, and an appropriate charger. That'd give you the voltage and current you need, and if you got a large capacity battery (12 amp-hour 6V batteries are commonly available), it'd probably last for a whole day of intensive shooting. And still be (relatively) portable.
Actually hooking the battery up to the flash(es) without ruining the battery compartment, I leave as an exercise for the reader. Many photo hobbyists have had success making dummy batteries out of dowel, with metal contacts on the end.
New wire price record?
Apart from dogs and those fancy pants so-called audiophiles, can it really be said that a human can notice the difference between your relatively expensive Monster - type audio cables and these babies from Transparent?
What's the real deal in charging over $AU10,000 for a couple of interconnects? Should I take a trip to Dick Smith and start buying the bits to build and sell my own?
Ed
Answer:
Package it right, and people will buy anything.
Transparent Cable have been selling these things (mainly cheaper versions, of course) for years now.
Their cables, like many other exotic cables these days, are supposed to contain a painstakingly hand-tuned "network", and their explanation of why the network is needed is typical audio snake oil mumbo-jumbo. Talk about very high frequency response cables as if they're something that a stereo system needs, say some stuff about noise that'll have anybody with an electrical engineering degree rolling on the floor laughing, and then finally get around to explaining that the "network" is just meant to attenuate very high frequency radio noise, which hi-fi gear can't respond to and doesn't reproduce anyway, so what the heck the problem is meant to be escapes me. They say RF noise and other things that're all at minus a zillion decibels over the whole audio spectrum have more effect on what you hear than you'd think. Other people say that it's worth paying good money to have someone allegedly beam healing energy to your cat. People say a lot of things. (Oh yeah!)
Transparent's claim that typical audio cables "resonate" at 1500 to 2500Hz is, by the way, laughable. Yes, you can make a speaker cable resonate, and some audiophiles who almost know what they're doing have made some pretty graphs to show the phenomenon.
The graph on that page, though, was created by a signal with a 35 nanosecond rise time (indicating a frequency way up in the megahertz), and there was no termination on the end of the cable. Put a speaker there and poof, there goes most of the resonance; feed it with an audio signal and there's nothing to resonate in the first place. Sure, the cable will pick up some RF noise, but at minuscule voltages compared with even line level voltage. Transparent claim that the RF can make a difference. Other people claim the government is spying on their brains with radar.
Here's another attempt to sell expensive cables based on claims of resonance. Yep, zip cord's more inductive than fancy-cable (or much cheaper off-the-shelf low inductance cable, like Belden 8214 coax...). This can matter, if your speaker is not inductive at high frequencies, but virtually all speakers are; if you've got a moving-coil tweeter in your speaker, then it is. Piezo and electrostatic and some other less common tweeters can benefit from lower cable inductance. A pair of cables to suit may cost as much as $US50 to make, if they're a bit long and you use very nice connectors.
The Transparent Cable people are, of course, adding inductance to their cable, unless all they're doing is putting a low value capacitor across the conductors. This search turns up their patents (as well as some others), so you can read the gory details for yourself. Yep, they put capacitors across digital cable conductors as well. In other news, it is apparently possible to tap the hollow Earth's internal Vril energy.
Unsurprisingly, not everyone is impressed by Transparent's alleged innovations. It also appears that some people don't believe that "Every thought and experience you've ever had in your life gets filtered through these chakra databases."
It's really easy to demonstrate the effectiveness, or otherwise, of a voodoo audio tweak. Blind testing, people. If you sell cables for thousands of dollars (and make a big deal of your magnificently equipped lab...), you should be able to afford to spend a weekend doing tests with some people who aren't on your payroll. These outfits never do that, though, for the same reason that psychic surgeons aren't big on independent testing of their claims.
The only places the word "blind" appears on the Transparent Cable site are in the review reprints here and here.
In both of them, we learn that Transparent Cable's own staff swear up and down that they've done blind tests and verified their claims, in a very nice listening room. Well, ain't that special. Benny Hinn's staff insist that he really does cure the lame, too, but apparently he never has a gap in his schedule to visit the National Institutes of Health and do it in controlled conditions. Fair enough; I guess he's got enough millions already.
People believe the voodoo audio hype just the same, of course, and buy the darn cables. Other people believe their badly behaved children are space aliens.
Honestly, sometimes I wonder how the heck we ever managed to invent the lever, much less land on the moon.
(And it goes on...)
