Dan's Data letters #175
Publication date: 11 October 2006.Last modified 03-Dec-2011.
Rust-bug repellent
I am the lucky buyer of a new car and my dealer is offering me the chance to buy a "Defense Pak Computerised Electronic Corrosion Inhibitor" (at a cost of $AU1000).
Only issue is, do these things actually work?
I've been trying to look around the net and can't find anything about them and was wondering if you could shed some light.
Lambert
Answer:
No, they don't work. At all. I mention them in passing in this column, as being even
less plausible than the electronic rust inhibitor doodad I'm complaining about there.
See also this page.
(The Car Talk guys don't know why these gadgets don't work - they have sacrificial-anode devices mixed up with electronic ones, and don't know that sacrificial-anode devices work because the thing they're protecting is immersed in an electrolyte, which cars aren't - but they do at least know that they don't.)
The Defense Pak Blah Blah, of course, does not explicitly claim that it's a cathodic or impressed-current system - nooooo, it "sends a continuous stream of silent electronic pulses throughout the body of the vehicle", which "act to inhibit the electro-chemical process".
Except, of course, they don't. This is pseudoscientific gibberish. A conductor with current flowing through it will rust exactly like an identical piece of metal that's just sitting there, unless it's part of a natural or forced electrochemical cell that shifts the corrosion somewhere else. Cathodic and impressed-current rust protection does actually work, when the thing being protected is surrounded by a conductive medium like water or earth. But It doesn't work on cars, because they're (mostly) surrounded by nothing but (mostly) insulative air.
The process the Defense Pak and most other "electronic" anti-rust gadgets claim to use is even further from usefulness. It is nonsense, and cannot work, ever.
I am, of course, not qualified to tell the universe what to do. Neither are all of the world's electrochemists and physicists. The proof is in the pudding; if your explanation for your antigravity machine is nonsense but the thing's obviously hanging there in the air, you still get your Nobel Prize.
But these products fit a well-established pattern which makes clear to me that if they're not scams, they're doing a very good job of pretending to be.
Umpteen companies have been selling electronic rust-stoppers for many years. Many billion-dollar industries would love it if the devices worked. And yet the devices are still being sold one by one to regular Joes, just like miracle fuel-economy potions and magical plug-in "power savers".
If electronic rust-preventers actually worked, the people who sold them would have made more money than God in a week and retired, and most if not all vehicles would come with one built in.
Does Caterpillar include an electronic de-ruster on every bulldozer? Does the US Army put 'em on their vehicles?
Kenworth?
Toyota?
Note that you certainly can use a sacrificial-anode or impressed-current system on any part of a land vehicle where water tends to accumulate - if there's some nook or cranny that always gets damp and starts rusting, you can bolt a lump of magnesium in there and protect that particular spot just fine. Or, better yet, you can give the whole car's steel panels a coating of a suitable other metal like, oh, I don't know, zinc.
This process is, of course, known as "galvanising", and it's normal in the car industry these days. I don't think it's even possible to buy a steel car that doesn't have galvanised panels, any more (though there are still plenty of cars that aren't fully galvanised). I think there used to be problems with spot welding such panels, or something, and that was one of the leading reasons why old cars rusted so rapidly and so badly - they had to be made out of ungalvanised steel, or the galvanising was damaged by the welding and they rusted at the welds, which is of course the very last place you want rust to start. But modern mainstream cars are well-galvanised from bumper to bumper.
Modern cars are far more rust resistant than older cars, mainly because of galvanising but also because of the wider use of plastic for some panels and bumpers, and possibly also because of better paint. People who remember how fast their old Cortina was eaten by the rust bugs may, therefore, buy a new Corolla, pay extra for some shady dealer to bolt a useless electro-gizmo onto it (or, more traditionally, pay for greatly overpriced "undercoating" to be sprayed all over the car's belly - though at least that stuff probably does genuinely provide some protection, especially in places where they salt the roads in winter), and be amazed to see that there's no rust to speak of after ten years.
Except that that's normal these days, if you don't live by the seaside, drive on salted roads in some cold-country winter, or keep scraping your car on concrete pillars in the supermarket parking lot and exposing bare metal.
Oh well. I suppose it's nice, at least, to see a non-functional automotive gadget that's not supposed to do anything to your fuel economy.
Vintage digicams
Love your work! Thanks to this and this, I am now, via eBay, the confused owner of both an Olympus C-2500L and a Sony DSC F505.
Being the simpleton that I am, I have combined the logic in both your articles to conclude that, having spent about $AU200 on eBay buying these two cameras, I now have products that are, for the purposes of stunning 4 by 6 prints, about as good as or better than any current $AU500 or up prosumer digital camera. I say $500 or up because they don't seem to sell $AU2200 non-DSLR cameras anymore, so I figure anything that looks like a C-2500L today must be in roughly the same class, and these seem to start from that price point.
Have I got it right? Does $AU5000 (in year 2000 dollars) worth of gear beat a, say, Canon S3 IS hands down, for picture quality?
I want to believe! Tell me it is so!
Larry
Answer:
I think I, personally, would have bought just one of the cameras from my old comparison.
Still, the more the merrier.
In answer to your question - yes, those old cameras certainly can still cut it today against new models that cost significantly more.
There are still limitations, chief among them low light performance; the C-2500L and DSC-F505 are both crappy if you want more than ISO 100 out of them, though both cameras go to f2.8 at full wide angle, which lets in a decent amount of light.
But, thanks to the megapixel arms race, no modern cheap camera is significantly better. If you scaled down the output from a good modern $US300 camera to 1600-by-1200-ish then you could probably get a somewhat better result than these 2000-vintage cameras manage, but it's really not a huge difference. You still have to buy something with a bigger sensor - which probably means a DSLR - to get really good high-ISO or long-exposure performance.
The Canon PowerShot S3 IS is (...is is...) indeed an excellent example of a modern mid-range compact, and it's got some features that the old cameras can't touch. Huge zoom range (without much distortion or chromatic aberration at either end, and without a giant lens barrel either; in the older cameras' day, super-zoom cams looked like this thing...), image stabiliser (essential with that much zoom), and a decent movie mode too. And it's got enough resolution that even if you don't much care about huge prints, you can crop your shots heavily and still have enough resolution for a sharp normal photo print.
All that for around $US400 (or maybe $AU550) delivered with a proper memory card (I do not consider Canon's stock 16Mb card to be "a proper memory card") is a great deal. The S3 IS is well worth buying.
But it still isn't any good at more than ISO 200 (even that is stretching it), and its lens is only f2.7-f3.5 at wide and telephoto, respectively (though that telephoto f-number is very good, given that it's at a 432mm-equivalent focal length). So, compared with either of the quarter-the-price old cameras, the S3 IS really isn't all that thrilling.
The old cameras don't beat the S3 in any important respects (except perhaps for how sad you'll feel if your camera is lost/smashed/drowned - oh, and the C-2500L has that rather useful IR remote control!), and they definitely don't beat it for telephoto applications.
(Though you may well be able to find screw-on tele-adaptors for them cheaply on eBay; those accessories were expensive when the cameras were new and exciting, but now there's no demand for them, and for other specialised stuff like Sony's old add-on flashes - the down side of this is that the F505's lithium ion battery remains non-cheap.)
But they definitely are better value for money, for most people's purposes.
You got two perfectly good cameras for little more than a third of the price of one new one, after all!
Glowy things
I saw these and went "I want..."
Hendo
Answer:
Yep, I saw them too, and considered trying to scam a couple for review - then I saw that the contact info was just
some dude's e-mail address, which does not suggest a large production run.
Note that the Pak-Lite is basically the same thing, but without a diffuser on top. Make some frosted plastic matchboxes to sit on top of some Pak-Lites and you'd be in business, and be able to use the lights as proper flashlights as well. The basic Pak-Lite goes for about $US10 plus delivery, battery not included.
(A reader's now pointed out to me that bike tyre repair kits are often sold in containers that look very much like "frosted plastic matchboxes". So your local bike shop could provide a short cut to success in this project.)
It wouldn't, of course, be tremendously difficult to make your own version of these lights from scratch, either. A single 5mm LED would be bright enough for the purpose (sand off the tip of the lens to diffuse the beam), and you wouldn't necessarily even need a switch - just unsnap the battery to turn the thing off.
With a simple series resistor (270 ohm for a 3.6V white LED from 9V at 20mA; this calculator is handy) you'd still get decent battery life. A 9V alkaline can deliver 20mA for at least 30 hours, and would actually last quite a bit longer as its voltage fell and the lamp dimmed. You'd actually get perfectly good performance at an initial 20mA from cheap 9V carbon-zinc batteries - they'd last a couple of nights, no worries.
Or do it the real man way, and reinvent the product to sit on a six volt lantern battery!
(This Instructable uses a flashlight bulb, which'll flatten the battery very quickly {or just explode, if you pick a bulb that expects to run from two or three cells}. This design's much neater.)
M cro hone pr bl ms
I've been having problems with headsets lately. It seems that the past two headsets I bought, the microphone breaks after two weeks of use ($35 each). The idea of spending $70 a month just to have a working microphone is a bit unsettling. How it breaks is odd, because the microphone when positioned a certain way will crackle and pop and eventually work if I hold it in a very specific position. That is far too tedious to make use out of it; I need it to be convenient for it to help me. (I play World of Warcraft and need a microphone to direct orders/report things via Ventrilo.)
Please note that I live with a family that can treat headphones rather harshly, so durability is a big pro I'm looking for in a headset. I'm wondering if you could explain a few things to me.
What is wrong with my microphone? Can I fix it (I have no warranty on it)? If not, what is a good headset (with microphone) that is highly durable and decent quality? By durable, I mean having to live through family members slinging it across the desk when they're done with it, surviving a chair rolling over it, etc.
Brett
Answer:
I'd place a small wager that the symptoms you report indicate a fractured wire going to the mic. Cheap headphones
commonly go intermittent if they're treated at all roughly (which is pretty much inevitable; computer setups generally
guarantee a yanked headphone cable from time to time), and it's the same deal with the mics. Sometimes the breaks
can be fixed by someone mildly handy with a soldering iron; sometimes cat-hair-fine wires and cheap plastics make
it very hard. It's seldom worth the effort, for really cheap 'phones.
Note that some cheap 'phones come with an excellent warranty. Koss is a particularly good example; that's their US warranty, and an Australian reader assures me that the same deal applies in Australia, though here you have to pay $AU15 for a no-questions-asked repair or replacement, rather than $US6. Koss have a few headset offerings that you might like to consider. I suggest the SB40 set, mentioned in one of my letters columns here.
Really high quality headsets can get very expensive. A more realistic option is to buy a nice pair of headphones (or a cheap and indestructible pair, with or without lifetime warranty) and use a separate mic. A clip-on lapel mic is one more thing to remove and/or unplug every time you get up from your computer (unless you clip it onto your headphone cable, which is not a great idea from a sound quality point of view, but it'd work OK for games - a reader pointed out this Zalman product to me), but basic ones are dirt cheap and there are many very high quality models available, as well.
For gaming and Skype and such you should be fine with any $4-delivered clip-on mic. There are lots of options that don't cost a great deal more and provide good enough quality for you to record your bedroom-studio top 40 single, too. The magic search keyword for tiny-mics is "lavalier", by the way.
Or, of course, you can stick any number of other kinds of mic on a stand on your desk, or hang 'em off your monitor, or whatever. If you're using headphones, feedback won't be a problem. Then you can insist that you just can't play properly without your Shure 55SH Series II Elvis mic!
Requirements: 80-column card, floppy drive
I came across an ancient review of yours of the Jaton Magic DVD card. I got one from a friend and would like to use it for an XP media center. Everything is in place except... the drivers, of course. I've been looking for them all night now (on the Web, not in my friends home :-) and they are just nowhere to be found. Also mailed Jaton, but got no reply.
Do you still have them in an accessible place? If so, could you send them to me, pretty please :-)?
Robert
Answer:
I'm sure there's a terribly good reason why you're determined to use a hardware DVD decoder from the
days before CPUs were fast enough to do it
themselves, so I'm not even going to ask.
Surprisingly enough, it is still possible to get drivers for these old cards. Not from Jaton, but the Magic DVD board uses the RealMagic Hollywood+ chipset from Sigma Designs, and they've kept making drivers for it, including ones for Windows XP.
The drivers I got with my review unit would, of course, be useless to you, since at the time the board only came with drivers for Win95/98. Win98SE hadn't even come out yet.
Have you ruled out acetylene?
I can imagine you probably get electrical questions by the bushel every day, but I hope you can answer this one, this one time.
I recently dusted off an old dual lamp Niterider bicycle lighting system after years of neglect. While this posed no issues with the lamp housing and other rubber and plastic bits, you probably can imagine what happened to the 13.2V NiCad battery pack. Not too bad actually; two dead cells and most of the others had crystallization upon them (probably from the bleeding dead cells).
To the point - like many (particularly older) halogen-based lights for bikes, these use MR-11 12V halogen bulbs with integral glass covers. So in effect to boost efficiency at the cost of bulb life they are being driven, idealistically? optimally?, by an additional 1.2 volts. In the battery pack for this system it was 11 1.2V sub-C cells.
I decided to replace the dead-ish pack with something similar, but as is typical with me I found that some people were successfully driving the MR-11's with 14.4V NiMH packs with additional corresponding decrease in life. But these bulbs are designed for thousands of hours of life, from what I have read, so a few extra volts would still leave at least a couple hundred hours of burn time, right?
Story doesn't end there, however. I had my eyes on a 14.4V NiMH pack and was about to buy when a little further probing revealed that a 14.8V lithium pack was way lighter, with 4000 mAh capacity compared to 3200 or so for the NiMH! So I bought the 14.8V pack and it is now charging in my Maha universal charger. I know your feelings on lithium batteries, but honestly for what I paid and if the system works the way I am hoping it will, I don't mind replacing it in two years.
Delving a bit into the 14.8V information, I found a reference that it can put out 16.2V peak, which has me a bit worried and comprises the first of many questions... but first:
The system I have has a soft on/off for the lights. After charging the 14.8V pack for about 30 minutes I decided to do a quick test of how bright I might expect the bulbs to be, but the results were not what I was hoping for. I hoped the battery simply hadn't enough of a charge yet, because all I got were the filaments barely lighting and then fading to black. I'll give it another shot after a good eight hour charge (it is after all only an 800mAh rapid charger going into a 4000 mAh battery, which has never been charged).
Currently I have a 12 watt and a 30 watt MR-11 lamp in the unit.
So, should I worry about the 16V+ peak voltage? For that matter, what about 14.8 volts? Is it simply going to be too much, or only allow the bulbs a few hours of life at almost a 25% overvolt? Would it be smarter using lower or higher wattage bulbs to increase life? Melting might be an issue at some point. Could a resistor placed somewhere help? What about something called a PWM regulator, which I have no clue how to use or implement?
Answer:
Lots of cheap video lights, and a lot of homebrew bike lights, also use those 30mm MR-11 lamps. Homebrewers also like
the 50mm MR-16s normally used as downlights - cheap, reasonably tough, tons of output, decent beam shape for a bike
light.
You're right about the tradeoff between efficiency (or, at least, brightness) and lamp life that's being made, here. Increasing the input voltage from a nominal 12 volts to a nominal 13.2 means a theoretical power boost of 21% (power increasing with the square of the input voltage for a simple resistive load), but a bit less in real life, since the resistance of the filament increases with its temperature. You also probably won't get quite the full nominal voltage out of the pack, since it's pretty heavily loaded, so the actual "overclock" probably isn't very great.
Running the lamps hotter will make the light bluer, as well as just brighter, though. So you'll get more of an illumination improvement than you'd expect from whatever the actual power boost turns out to be.
Going to 14.4V nominal is indeed getting a bit extreme, though. I wouldn't be surprised if that was a genuine 30% power boost, or even closer to the theoretical 44%.
Bulbs differ in their ability to withstand higher currents, but you're really pushing it if you ask for 35% more output from most filament globes.
There's no particular relationship between a globe's wattage and its ability to withstand higher power operation, beyond the fact that the higher the power a globe of a given size is running at, the hotter it'll get, and excess heat can lead a bulb to various non-typical failure modes - exploding when water hits it, melting any plastic around it, melting the solder on its contacts, settin' yo' ass on fire, et cetera. Halogen lamps are meant to run hotter than ordinary incandescents, but there are limits.
And, yes, the higher terminal voltage of the lithium ion/polymer pack could be a problem.
All rechargeable chemistries can deliver more than their nominal voltage when freshly charged, but most of them drop down close to their nominal rating very quickly under load. Rechargeable lithium batteries, though, can deliver more than their rated 3.6 or 3.7V (lithium has higher voltage per cell than NiCd or NiMH - a LiI pack of a given voltage will have a third as many cells as a NiMH equivalent) for quite a bit of their discharge curve, even under a fairly heavy load. You can expect the first 20% of the discharge curve to remain above 3.75V, which is 15V from a four-cell pack like the one you've got. Even that pack's "14.8V" nominal rating (other people sell the same thing as a 14.4V-nominal pack - there's no clear way to specify nominal voltages) is pretty darn hot for the lamps you're using.
On the plus side, only the lamps (and maybe their housing...) should be getting barbecued, here. Modern LiI packs with as much capacity as yours are all fine at several-amp discharge currents, and many are good for a lot more. A total of 42 watts of MR-11s, even if they're running at close to 60 watts, should be no big deal for such a pack to drive.
(I still wouldn't build the battery into my bike seat if I were you, though.)
Regarding the light's brief pulse of illumination when you tested it, if the thing has a soft on/off circuit and not just a simple switch, then it's possible that the new high voltage pack has popped something. Probably just a fuse, but anything's possible. The pack should have had more than enough charge for a proper test after half an hour on your charger.
(If the charger wasn't actually charging for some reason, though, you'd see the same symptoms.)
MR-11 lamps aren't very expensive, so as long as the new battery isn't frying the light's control hardware then you're not in any particular danger here.
Personally, though, I'd stick with the standard 11 cell pack, and just replace the old NiCds with NiMH cells.
(Doing that now, of course, will leave you with a surplus LiI pack on your hands. If you want to unload it and you have a local R/C hobby store, try calling them - a lot of model plane people are using LiI cells these days, and are used to paying too much for them.)
You can buy NiMH cells loose, but soldering them together can be a pain for a newbie - as I explain in my ancient camera battery piece, you have to scuff up the cell surface and use an iron with a broad tip, but not a very hot iron, and also not long soldering time, or you'll boil the electrolyte. Plastic battery holders are a simple alternative that can work OK for medium current applications, but you can also get pre-connected batteries by buying model car "stick packs" or dismantling cordless drill batteries. Big-brand cordless tools these days often have quite good cells in their batteries.
Eleven cells is an inconvenient number, but since pre-connected cells can usually be had in multiples of six (7.2V stick packs, 14.4V drill batteries), you'll probably only end up wasting one cell.
This can work out better than you might think, since 14.4V drill batteries often have a neat block of eleven cells, with the last one inside the contact doodad that sticks up out of the pack. Delete that cell and you've got a nice little eleven cell module.
Pee-wee planes
I wanted to let you know about the Air Hogs Aero Ace biplane.
Everyone is raving about these things. Looks like a lot of fun for $US30!
I think you should do the very hard "work" of reviewing this "gadget" and let us know where we can get them in Oz!
Michael
Answer:
Yes, it does, but as you say, no bugger seems to sell them here in Australia. Air Hogs also seem to no longer be making
that product, possibly because of their enthusiasm for the commercialised version of the "Hydro-Foam",
videos of a prototype of which were freaking everyone out a few months ago.
There are a few other options in the little-foam-plane-with-pager-motors department, though. Umpteen Chinese companies are pumping out indestructible trainers in the small-to-medium size categories, and there are lots of littler version that're around the Air Hogs Aero Ace price point.
Some of them are quite scale, some aren't.
I never really got into R/C planes - too nerve-wracking - but the little indoor fliers are pretty much bombproof and, as you say, cheap.
Unfortunately, the whole idea's pretty much a non-starter, since I have three cats...
And now, an opinion
What can be done? Frequency modulators seem to have at least 20+ patents: Their in medical scanners, air conditioning units, wind turbines and HAARP...
I was told that some 8 years ago by someone in the air conditioning industry that they knew their equipment was non-complient to EMC law. They use a inverter drive system to act as a transformer... Except they don't work like so call transformers. This is supposed to add to power effeicency in both wind turbines and air conditioning units.
Yet the local company who runs the National grid (central networks) their EMF officer Ian Brookes, seem to let slip that the national grid did not use inverters, because in his words that inverters were not power effeicient.. Someone's telling lies?
So called alternative power or renewable energy is being falsely applied to these devices... Its been convenently lost as to what and how they work/hence how much radiation they give off????
I have been told that inverter/rectifiers produce a standing wave; where that energy does not get less over distance. Yet experts who install microwave transmitters and mobile phone transmitters really have not a clue how they work. In that most of the power is in non-linear or T waves/inverted waves.
If you would like to discuss this further could you please write?
Yours
Faithfully
(Mrs) Carol [surname redacted]
Answer:
I didn't reply to Carol.
Should I have?
I really don't know what I could have said.
Perhaps this piece set her off, or maybe one of my pieces about UPSes.
I don't know.
I don't think I know about anything, any more.
I'm going to have a little lie down now.
