Dan's Data letters #60
Publication date: September 2003.Last modified 08-Mar-2015.
Mit Spitzensparken, Again
A few minutes ago I did a very stupid thing.
I've recently returned from a stint at the University of California, over which time I bought a Samsung ML-1430 laser printer. Needing a printer back home in Sydney, I shipped it back at the end of my exchange, and it arrived today. In my haste and need to print a document I grabbed a spare power cord to replace the American one, plugged it into the wall, and some sparks, smoke and a tripped circuit breaker later realised that I probably should have looked at the back of the printer. It wasn't a dual 110/220 volt device.
So I obviously screwed up. The two concerns I have are: In your opinion, is my printer dead, fixable or perhaps miraculously fine? And, if the printer is returned to a workable state, is it feasible to perform some sort of voltage conversion, given that it was a very nice but fairly cheap laser printer to begin with?
James
Answer:
The printer's probably toast. Damage from doubling the input voltage to
power supplies not expecting it may be restricted to the PSU, but
even that may cost you more than the price of a new printer to replace -
especially if you're trying to buy a 110V PSU in a 220V country, or a 220V
PSU to put into a previously-110V device. Standard ATX computer PSUs are
easily interchangeable, but the ones in other devices aren't, necessarily.
And yes, you can run a laser printer from a step-down (or -up) transformer, but it's probably not worth the effort. Laser printers draw a lot of power when they're heating up their fuser - 500 watts or more, commonly. A 500 watt 220-to-110V step-down transformer is a pretty hefty chunk of iron. As I mention here, though, odd-brand step transformers can be had quite cheaply, if you shop around.
I really think you'd be better off just getting a new printer, though.
Double-click of death
Under Win98 I used to be able to make a shut down (or reboot) shortcut, by making "C:\WINDOWS\RUNDLL.EXE User.exe,exitwindows" the path. (You could also do "exitwindowsexec", for a reboot.)
Well in Win2k or XP that obviously won't work, so I tried to change it to "%windir%\system32\RUNDLL32.EXE User.exe,exitwindows", but it still won't work. I tried using userinit.exe instead, to no avail. Is there any way to create a shortcut that will shut down your system under Win2k?
By the way, if you're wondering why I would ever want to do this, it's quite handy - I can set the task scheduler to shut down my PC, I can sneak one of these shortcuts into an unsuspecting friend's Startup folder...
Alex
Answer:
Try this utility.
WinXP, by the way, has a command line shutdown utility built in as standard; it's inventively called "shutdown". It's documented in XP's help; you can also type "shutdown /?" at the command line for options.
Shutdown.exe isn't included with Win2000, but it seems to work fine if you just copy it over from an XP system. You can find it in the windows/system32 directory.
To PFC or not to PFC?
I recently bought a shiny new GA-7NNXP and have had annoying little problems with it that you may have heard of before. Generally, on turning on my computer, either the display doesn't come up and it doesn't sound like it's actually booting, or it will get to around the RAID detection stage and restart by itself. Rather annoying problem that I've been told can be solved by using an Active PFC power supply. They look rather costly, though.
What's going on? Is it the power supply, or should I take the bastard back?
Krunchy
Answer:
Your problem may indeed be caused by a lousy PSU - varying flakiness like
this is a classic bad-PSU symptom-set - but there's nothing about Power
Factor Correction (PFC) that will, by itself, make any difference to this.
Different kinds of PFC make a PSU look different to the mains supply, but
they all look the same to the computer, all other things being equal.
Any quality 300 watt or higher PSU should solve your problem, if the problem is indeed PSU-related. In situations like this, I always recommend you buy the new PSU and give it a go, since the worst case scenario is that the problem turns out to be something else and you're stuck with a spare PSU. A spare PSU is a useful thing to have on the shelf, since PSUs always seem to drop dead at the start of a long period when you have no opportunity to buy another one.
Also, use both sides of the toilet paper
Firstly I'll say that I'm a student in the Level 5 Diploma in Information Technology at the TAFE in Warrnambool. It's a great course, everyone should check it out </shameless unpaid promotion>.
Recently at the yearly budget review, the TAFE administration decided that too much money was being spent on electricity, and the blame was laid on the IT department, primarily about the fact that there are about 150 CRT monitors (15" or 17") turned on for 10 hours per day, not counting night classes.
In a bid to reduce the power consumption of the institution, the IT department have decided simply that people should turn off their monitors when they are not in use, or when the user is at the computer. Stickers have been put in place next to the power button on every monitor to indicate this should be done.
But they must be turned off. Standby mode is simply not acceptable, as apparently the standby LEDs on the monitors draw "more power than you would think", if the campus newsletter is correct.
Fair enough that this could reduce the amount of wasted power used by idle CRTs and nasty, power hungry LEDs, but it can also lead to the monitors being turned on and off around 5 to 10 times per class (three hours), perhaps being powered on and off 30 times per day.
I'll be the first to admit my meagre knowledge of CRT construction and internal operation, but I'm fairly certain that the act of turning a monitor on draws a large amount of current for a brief time before settling down into a relatively lower consumption mode. Would these repeated surges actually create a higher kilowatt-hour usage for the monitor than leaving the screen on for the entire day? It certainly couldn't be healthy for the monitor unit itself. The usable lifespan might be noticeably decreased.
I'm interested to hear your views on this (perhaps wrongly implemented) energy conservation plan.
John
Answer:
The turn-on degauss surges don't draw an important amount of power. I've
addressed this before, here; even a bad-ass
degauss circuit won't draw enough power to run the monitor for more than
a minute. The turn-on surges may draw enough for as much as an hour of standby
operation, but it's more likely to be ten or twenty minutes' worth.
Power-cycling a monitor or PC a couple of times a day doesn't generally do anything quantifiable to its lifespan, but doing it thirty times a day could. So yes, this could be very false economy indeed.
You're not saving that much power by turning stood-by screens off, anyway; a modern monitor that's properly in standby (as opposed to just displaying a black screen) should draw less than five watts.
I've got some more rambling on this subject here and, more relevantly, here.
Purity test
Right, unshielded speakers distort (at close range) monitor pictures. I have nice, unshielded speakers. They sound good, and were cheap, and the only disadvantage is that I have to keep them reasonably far away from my monitor. This isn't a problem. What I do want to know is if there is a method of calculating the size (in terms of distance from magnet) of the field my speakers would be outputting, and what that method would be. Other than, of course, moving the speakers about until I have a nice square picture again, and then measuring the result.
Guy
Answer:
The size of the field, technically, is infinite; it just drops off rather
rapidly as you get further away from the magnet, and for magnets of domestic
dimensions (as opposed to, say, neutron stars) you don't have to take many
steps before the field strength is below that of the planet's own field.
To a first approximation, magnetic field strength drops off as the inverse cube of the distance from the middle of the magnet. So if you double the distance, you should see one-eighth as much perturbation of the screen image.
PPP over telepathy
I don't know what I did for this to happen, but I just found that I was connected to the net via my USB ADSL modem, yet Win2k had no network connection visible.
There I was, browsing the Web, when I noticed there was no little network symbol down in the taskbar. Having a look at Network Connections in Control Panel told me that none were "available". Yet my usual online game server query was pinging happily away at a long list of servers, and I could connect to any of them. So, here I am connected to God-knows-what, yet Windows (and I) really have no knowledge of it.
Can you explain how this could happen?
Scott
Answer:
I don't know what your particular PC's doing in detail, but various versions
of Windows are perfectly capable of getting screwed up in this way, sometimes
permanently - or, at least, until a good solid reinstall.
I've got an unimportant Win2000 box that's broken like this at the moment. I got around to putting SP4 on it the other day (should have left well enough alone, since it's behind a firewall and was working just fine with minimally-patched Win2000 SP1...), and now whenever it boots, svchost.exe fails shortly afterwards, and stuff gets weird. Network connections work just fine, but you can't view properties for them or for anything else. You get no taskbar icons for various things. And special folders, like the network connections one, now contain either basic icons or nothing at all.
(In case you're wondering: No, it doesn't have the Blaster worm.)
Reinstalling Windows over the top of itself may or may not cure this problem.
Big batteries
I've decided to make a battery pack for my camera [as per my elderly tutorial here] out of five 7.5Ah NiMH D-sized Sanyos. $US18 each. Now I'm faced with the issue of charging these beasts. Sanyo doesn't even list the model - D7500NM-WT, and the only comparable NiMH that Sanyo DOES list is 200mAh under and doesn't list charging info. Searching for this model, or even a partial name (the WT apparently denotes solder tabs) turns up nada.
Fortunately, the spec sheets for Sanyo's other batteries all clearly show photos of the battery prominently displaying the "Standard charge" details.
Also, despite being hell bent on spending $US100 on the actual batteries, I don't want to spend much on the charger. Not to mention the fact that I can't find a hobby charger anywhere that will push anywhere near 7.5 amps; best I can find is 3 amps. What about using a voltage step-down circuit attached to the 5V (32 amp) rail of an ATX power supply? I'm familiar with how to construct the step-down circuit, but I have not the slightest idea about how to charge batteries. I wouldn't even begin to know what voltage to step down to, or even how the PSU's limited amperage would affect the charging process.
And once the things are charging, how would I know when they're done? I assume the first time I could easily check with a multimeter to see when the batts are done, note the time, and then put the PSU onto a timer so the AC just cuts when the charging period is done, but I don't plan on EVER running these things flat, so there would be a lot of extra time sitting on the charger, which I believe is dangerous. So, what on earth should I do?
Tony
Answer:
7.5 amps? Yee-ow! You don't want to try to throw full capacity into these
things in little more than an hour, which is what a 7.5A charge would do.
The very fastest NiMH chargers can manage one hour charges without beating
the batteries to death too fast, but cells should not generally be
treated so roughly.
Three amps is plenty. A full, from-flat three amp charge should take about three and a half hours (since you have to put more energy into the batteries to charge them than you'll get out on discharge; the rest is wasted as heat), and you shouldn't expect to be able to charge them any faster than that. Given the monstrous run time the pack is likely to have, it's probably not going to be very near flat most of the times you charge it, so a quite modest little charger may give you perfectly acceptable charge times.
Now is not the time to penny-pinch; buy a NiMH-capable hobby charger. You'll need one that works with five cell packs (basic chargers are likely to expect six or seven cells), but the capacity of the battery isn't likely to be a problem. Most chargers don't "sanity check" the battery capacity and freak out if it seems to be very high.
You can throw some charge into any NiMH pack with a simple constant-voltage circuit, but picking the end of the charge cycle is harder with NiMH than it is with NiCd. NiCd pack voltage drops a tad at the end of the charge cycle, after a long rise (so you can pick the peak by hand with a multimeter, if you're alert), and they also get hot when they're full. NiMH voltage just plateaus off at the end of the charge, and the cells get hot as a matter of course when fast-charged, so you can't use temperature monitoring to pick the charge end either.
So: Much better to buy a pre-built charger.
Duratrax seem to have some decent options that're selling for non-ridiculous prices; see here, here and here, for instance.
Baby-cam timings
I just purchased an Aiptek Pocket DV2 on eBay after reading your great detailed review. You mention that slower CompactFlash cards will make the camera take longer than two seconds per picture.
Will slower cards affect video frame rate?
Supposing I had the slowest card in the world, about how much longer than 2 seconds could it take to save a still photo?
Joey
Answer:
The Pocket DV2 (and Pocket DV 3100/CoolDV 350)
video data rate is only about a tenth of a megabyte per second, which should
be slow enough for any card. If a card couldn't even accept data that
slowly, the camera would probably abort the video or seize up completely;
I doubt it'd be smart enough to drop frames.
I don't think even the oldest and crustiest of CompactFlash cards would push the shot-to-shot time of the Pocket DV2 above three seconds.
3.5-inch archaeology
I'm confused. Dean was saying the floppy drives in his cases were not working because of the bezel on the case partially ejecting the disk. Ok, that is certainly possible, but the picture shows two floppy drives, one with a regular bezel and one with the case bezel covering the drive. This means that one should have worked fine, while the other didn't work at all when installed in the case.
So what I'm confused about is, why does he need two floppy drives?
Aaron
Answer:
From the "Caleb" logo on the front of the other drive, I figure it's a UHD144
(mentioned
here). The UHD144 was the cheapest of the small crop of high-capacity
floppy-compatible drives that got beaten to death by the original 100Mb
Zip drive, in those halcyon days when CD writers were things Hugh Hefner
couldn't afford, and Iomega's stock price shot through the ceiling.
(By the way, Ziff-Davis' five year old poll is still at least pretending to accept input, for some of the options!)
Anyway, the UHD144 is an IDE device, and so you can't boot from it (OK, maybe with some special BIOS, but not as a plug-and-go option). These sorts of problems were part of what doomed the UHD144 and the more popular, and more expensive, LS-120; they were floppy-compatible, but not floppy-compatible enough. Reasonably current computers today can all boot from LS-120. Nobody much cares any more, though.
Dean's gotten back to me, now, and said that yes, that's an UHD144. He paid $US20 for two of them, with disks. The one in the picture isn't even hooked up, because he found it worked about equally well whether or not he connected the cable.
Capacity mystery
The current NiMH AA cells in my local store are rated 1850mAH. They also sell AAAs rated at significantly less (somewhere around 600 mAh, if I remember correctly), and C and D cells. The C and D cells are both rated 2200mAh.
Since C cells have something like four times the volume of an AA cell, shouldn't their capacity be significantly higher than the AAs, instead of merely 20% higher? And even more confusingly, since the D cell has roughly double the volume of a C cell, how can it possibly have the same capacity as the C?
Charles
Answer:
First up: C sized rechargeables are probably not quite full. The standard
C-ish size is the "Sub-C"; full Cs are probably a Sub-C in an overcoat.
The volume of AA, sub-C, C and D sized cells is about 0.5, 1.1, 1.6 and 3.2 cubic inches, respectively. This is assuming they're perfect cylinders, which they aren't; they've got end-nipples and cans of non-zero thickness. But it's in the right ballpark.
The strangely high capacity of AA NiMH cells compared with sub-C and C sized ones comes largely from the much higher demand for high-cap AAs. Far more of today's bumper crop of portable consumer gadgets run from AA cells than from larger sizes, and people want long run times from them, so that's where development's been focused.
Mass market AA cells have been creeping up in capacity by at least a hundred milliamp-hours per year, but C and D NiMHs already had good enough capacity for most purposes when they were first introduced. You don't want to use NiMH or NiCd cells to power C and D cell things like standby flashlights and boom-boxes that're going to sit on the shelf for weeks, anyway; the high self-discharge of these rechargeable chemistries will leave the batteries practically flat when you need them. So the market for the bigger cells just isn't that big.
What's the deal with the identical-capacity C and D cells? Those D cells are C cells, or more probably sub-C cells. They're just in a bigger casing. If you cut them open, you'd find some airspace or filler material around a smaller cell.
This has been going on for quite some time; it's how you make cheap D-size NiCd or NiMH cells. Plates and electrolyte are a lot more expensive than air or filler.
These cheap low capacity cells have always sold perfectly well to people who don't know (or who know, but don't care) about capacity, and just need cheap rechargeables that fit in a D-battery device.
The empty-space issue is not restricted to rechargeable batteries. Suspiciously cheap (and light) six volt lantern batteries, for instance, are also likely to contain some not-very-fresh air.
Here's an excellent page of current rechargeable cells. Note the nine amp-hour NiMH D cells. Those suckers are full!
Note also that there's a trade-off between capacity and current delivery ability. More solidly built cells with lower capacity may be happier with a faster discharge, and a faster charge as well. This is why the NiCd battery chemistry is still going strong in the radio control world; cheap NiCds can survive monstrous discharge rates and horrifying charge speeds (four minute discharge, 15 minutes charge? Bring it on!) better than the best NiMH cells.
