SoylentNews is people
SoylentNews
Switching supply units used today are of light weight and compact design, but also susceptible to errors due to the incorporated electrolyte capacitors. Film capacitors would have much longer service lives. However, they need up to ten times more space. Scientists of KIT's[*] Light Technology Institute (LTI) have now developed a digital control method for use of film capacitors that need slightly more space only.
The control method runs on a microprocessor integrated in the supply unit and detects disturbing environmental impacts, such that e.g. higher voltage fluctuations can be balanced. Hence, storage capacitors of reduced capacity are sufficient. Michael Heidinger, LTI, summarizes the advantages: "Use of these film capacitors eliminates the main cause of failure of power supplies, i.e. electrolyte capacitors. Depending on the design, service life may be increased by a factor of up to three." The result is a much reduced maintenance expenditure.
This one is digital.
[*] Karlsruhe Institute of Technology.
(Score: 2, Disagree) by GreatOutdoors on Thursday May 16 2019, @10:00PM (13 children)
Not only does the writer lack a basic understanding of the English language, but they also have no clue what they are talking about. Capacitors do not have errors, they are passive components. Complicating power supplies with more silicon based solutions is not a way to make them more efficient and reliable.
Yes, I did make a logical argument there. You should post a logical response.
(Score: 0) by Anonymous Coward on Thursday May 16 2019, @10:06PM
phys.org strikes out again
(Score: 0) by Anonymous Coward on Thursday May 16 2019, @10:38PM
electrolyte capacitors, new from Brawndo
(Score: 2) by RS3 on Thursday May 16 2019, @11:44PM (8 children)
It's of German origin, so maybe Google translate is to blame?
The word "error" is twice in the article, and in neither case does the article state or infer that capacitors have errors.
It says:
Certainly the wording is awkward, but I understand what they're trying to say.
I admit I have some issues with the article, but I don't have a problem with tighter (much faster) regulation, and thereby eliminating electrolytic caps. We've been doing microprocessor control of motors and other electromagnetics, sensing and analyzing winding current, phase, "back-emf, etc., so why not power supplies? And I think they understate the service life extension. If the design is conservative enough (over-rate caps, transistors, etc.) the things could last 50 years.
(Score: 0) by Anonymous Coward on Thursday May 16 2019, @11:52PM (4 children)
No, they won't.
The code in the microcontroller will suffer bit rot before then.
(Score: 2) by RS3 on Friday May 17 2019, @12:03AM (3 children)
Not necessarily. Not sure what you mean by "bit rot". If you mean FLASH wear, it should run in RAM and RAM disk, with FLASH for IPL and maybe some long-term saved variables only.
(Score: 0) by Anonymous Coward on Friday May 17 2019, @02:26AM (2 children)
Leakage of charge from a floating-gate storage cell, or from the storage cell getting hit by a passing cosmic ray.
All "non-volatile" floating gate semiconductor storage has this problem, and modern microprocessors
are filled with them, including "programmed at the factory" undocumented configuration bits which
WILL discharge over time, bricking the part.
(Score: 2) by RS3 on Friday May 17 2019, @05:09AM
So we make sure they have tin-foil hats.
(Score: 2) by RS3 on Friday May 17 2019, @05:26AM
I’m curious, what evidence brought you to that conclusion?
(Score: 2) by inertnet on Friday May 17 2019, @12:14AM (2 children)
This seems to be the original German text: (PDF) [kit.edu].
The original of your quote is: "Die Netzteile sind fehleranfällig, was auch die Lebensdauer der Endgeräte verkürzt."
I'm not a native German speaker, but I'd translate that almost the same, although "Lebensdauer" is "lifespan" and I'd translate "Endgeräte" in this case to "consumer products".
(Score: 2) by RS3 on Friday May 17 2019, @01:42AM (1 child)
Thanks for that pdf. I have several German friends. I won't bore / annoy them by sending it to them, but if I see any of them, I'll print it out and see what they think.
Yeah, and I'm seeing "Endgeräte" come up as end-point, and I'm taking it as "destination". Maybe it was written with an Enigma machine.
(Score: 3, Interesting) by Rich on Friday May 17 2019, @12:44PM
That's indeed correct. As in "Datenendgerät":"data terminal device". "Endgeräte" in the context is supposed to mean "devices where the power supplies end up", or "End-use devices". The German word use by them is not completely correct and a bit of an anglicism, but there is no german word for expressing it in an easier way, so it's forgivable.
(Score: 5, Informative) by bzipitidoo on Friday May 17 2019, @12:05AM (1 child)
Okay, talk of "capacitor errors" was weird, but it's true that electrolytic capacitors are often the weakest link in a power supply, and electronics in general. A power supply going bad is one of the top hardware failures that afflict computers and the capacitor is the most likely component of a power supply to fail and thereby cause the failure of the power supply. And it's not just the explicitly separately boxed components that make up a power supply, it's also a lot of capacitors and electronics on the motherboard. The other top hardware failures are fans and hard drives, essentially anything else that is not solid state, and has mechanics. Optical media drives are another high failure rate piece of hardware-- and as weak a link as they are, I am baffled by this custom of having an additional completely unnecessary motor in there to eject and retract the tray.
As for silicon based solutions to analog energy problems, don't sneer. They work. They work great. Such electronics are why we now have inverters the size of a 3.5 hard drive for a $100 or so, instead of the size of a cube refrigerator for over $1000. An inverter is a crucial device for such things as electric cars, essential for varying the frequency of the AC electricity that is supplied to AC motors, in order to make them run at different speeds. It's also essential for home solar panel installations, with its ability to convert DC into AC, which is quite a bit harder than converting the other direction.
There's really not been much alternative to the electrolytic capacitor. There are dry capacitors, but as mentioned in the summary, they take a whole lot more room, so much they really are impractical. No one wants to quadruple or worse the size of a desktop computer to make room for dry capacitors.
One other thing worth a mention is the capacitor plague of the early 2000s. Thanks to a bad formulation, a lot of capacitors manufactured in those years had much shorter lives than even their already short expected lifespans. They began to fail in as little as 3 years. If you looked at these capacitors, they would show telltale bulges at the ends, and even some electrolyte would leak out and leave a brown stain on the board underneath it. Electronics manufacturers were saved from having to do massive recalls and replacements by technological advancement. By the time a computer manufactured in the first half of the 2000s was beginning to spontaneously reboot every hour or so because its power supply was failing thanks to those defective capacitors, the computer was obsolete anyway. Most people simply upgraded to a new computer a little earlier.
(Score: 2) by RS3 on Friday May 17 2019, @12:25AM
Great post! Way back in the day (80s) I had gotten a little involved with 4-phase step (stepper) motors. Very simple drive circuits, torque dropped off quickly as RPM rose, etc. Then mid-90s was asked to evaluate and implement a Parker-Hanfin (sp?) "CompuMotor" step motor controller. They used very high voltage (200+) but pulse-width modulated, to drive the motor to insane speeds, rock-solid torque, no missed steps, and "micro-stepping" - holding armature between magnets by varying coil currents. All done by microprocessor control, sensing and analyzing winding amps, back-EMF, etc. Geeky, but I was quite impressed. (To get the really insane speeds you had to run an "autotune" routine, and they warned you it might tear up an attached mechanism.)
Much more recently I installed some industrial motor controllers (slang term: "freq drive" or just "drive"). They take 120 VAC, voltage-double rectify, filter, and 3-phase "H-bridge" drive 3-phase AC motors, all using a microprocessor, sensing winding current, back-EMF, etc. These were for 3 HP pumps. By default they'll only drive the motor up to 60 Hz (USA), but you can easily reprogram them to go much higher and run the motor much faster than stock, if the load can take it. I was stunned at the size of the thing- maybe 2" x 3" x 6", and no discernible heat. So it's not a stretch to control a switching supply transformer this way too.
And they're using this for residential water well pump motors- doing away with the storage tanks (big wet storage capacitor!) and maintaining pressure. (I'm not sure I like it- gotta see the power used in standby / maintain pressure...)