I've never heard of Vloss either, and I've been at it since Leyden was making capacitors with glass jars.

Reforming caps: This applies only to electrolytic capacitors. You can find a good deal of info on the web about how to do it, but by all means ask here if you are left wondering about some details. I have a little box with some resistors, switches and a current meter that I use together with a variable AC supply (rectifier in the box) just for this purpose, but my method is rather ad-hoc based on what I expect in terms of normal leakage currents. The higher the voltage rating and/or the larger the physical size of the cap, the more important it can be to reform it if it has be unused for a long time. This is really mostly just because that combination is more likely to lead to very unhappy events if the cap needs reforming.


My preference for values is to avoid a leading decimal if possible so, for example, 270 pF rather than 0.27 nF, 100 nF rather than 0.1 µF. On schematics I omit the space between the quantity and the unit, which is SI proper practice, to conserve valuable space; if I must use a leading decimal I always use a zero before it, space issues be damned 0.1R (I normally use the multiplier instead of a decimal, so 0R1, 4k7, etc).

The use of microfarads instead of millifarads for large values is just a "what is usually done" thing - rather like expressing something as 40 dB instead of 4 B.

pF v, µµF - I suspect the latter persisted in amateur radio circles for some time after it was replaced by pF in industrial electronics. I don't recall ever using µµF on a schematic I drew, and I started drawing schematics for things other than hobby bits over 40 years ago. My 1986 ARRL Handbook uses pF, as does my 1970 Electronic Experimenter's Handbook.

In something posted at AAC very recently there was a schematic where all the resistors and capacitors were marked in digit-digit-mulitplier fashion e.g. 104 for 100 nF, 101 for 100 ohms. Weird and most definitely not standard practice anywhere, as far as I know. I've certainly never seen it before.

 

I recall people using "pufs" and μμF but that is from old Brits and military service guys before the 1950's and 1960's.
nano farads has been in regular usage after 1960's.

To find out common usage, look at circuits in "Sourcebook of Electronic Circuits" by John Markus (1968).
(You might have a point to make here. I could not find a single usage of nF in the above text, with over 3000 circuits. I did not check them all.

 

Vloss: http://www.foxcomputer.se/Capacitor voltage loss Vloss.pdf

Anyway, Angstroms have gone out of favor too for wavelength.

Yep, the metric system is cool, but you can;t use body parts to estimate anything. In the metric system the numbers are too big. Forget 100 mm. That should be 10 cm. The latter I can "better" visualize.

Now, we can't have an Angstrometer output in units of nm, can we?

Thousanths of an inch or mills work when machinining. Metric and English works when the tools are in the same units.

Now, IC packages have gone metric or should I say, their packages have metric equlivelents, but they are the same part.

Adding insults, I can buy a 7/8" copper fitting at one store and it will be identical to a 3/4" fitting at another. How weird is that?

 

I looked at the "vloss" doc at the link at #23. Over the years I've put a considerable effort into learning about capacitors but I haven't a clue what that document is trying to say. It looks to me like someone just made up a term to apply to something he didn't understand. For example, "ceramic" and "foil" are both quite meaningless without additional details. My suspicion is that the phenomenon so badly described is related to dielectric absorption, but it may have as much to do with poor quality instruments.

If we are complaining about units, I would like to register my discontent with the continued use of "footage" for video. I do realize it is easier to say than nanometrage.

 

If we are complaining about units, I would like to register my discontent with the continued use of "footage" for video. I do realize it is easier to say than nanometrage.

Whoa! I never thought about the origin of that term. Did "footage" used to be a reference to the length (in feet) of film that were used to capture the images? Fun stuff!

 

Whoa! I never thought about the origin of that term. Did "footage" used to be a reference to the length (in feet) of film that were used to capture the images? Fun stuff!

Yup.

 

I looked at the "vloss" doc at the link at #23. Over the years I've put a considerable effort into learning about capacitors but I haven't a clue what that document is trying to say.

Good to see I'm not the only one who thought it was gibberish.

It looks to me like someone just made up a term to apply to something he didn't understand.

That was my impression, also.

 

Good to see I'm not the only one who thought it was gibberish.


That was my impression, also.

Also, it has numerous typos and misspellings. It looks like it was written by someone who's birth language is other than English.
So maybe the terminology is made up, or it's a literal translation (which is always a mistake) of the other language's original terminology.

 

Hello All,
I recently bought one of those generic "component testers'" on Amazon or eBay, and I must say I am impressed. For less then 20 bucks this thing can identify and value test most passives, and show discreet semiconductor pinouts and generic family (NPN/PNP BJT, Triac, diode, etc.). I especially like the rapid test for capacitor value and overall condition. However, I felt like a bit of a dinosaur when I tested my first cap. I was told it was a 100nF cap with X % of VLOSS and had an ESR of x.xΩ.

I have been retired for about 7 years, spent the last 20 years of my career as a video systems engineer, and for 25 years before that I was a bench tech/field service eng./installer. My point is I have been away from "components" for quite a while, which may explain the questions coming next.

When did nano farad become a thing?

Most of my electronics training was through the U.S. Air Force, and we mostly talked in "puffs & muffs" (pF and µF). Of course there are now super capacitors, and we actually have caps measured in Farads, but this nF thing caught me off guard. And when did VLoss become popular? More importantly, are go/no go values for VLoss published anywhere (similar to ESR charts)? My last days as a bench tech introduced me to ESR. We often had to shotgun entire boards because those little SMD caps didn't have much of a life (has that improved at all?). One last comment - I used to be an electronic "tinkerer" in the late late '70's through about the mid '80's.

About a year ago I got the bug again and I am amazed at what is available. My home "lab" was stocked mostly with Heathkit stuff back in the day, but now DSO's, DDS generators and all manner of other goodies are now within the reach of the hobbyist. Add to that the Arduinos, Pi's and plethora of shields and software and this is a great hobby (with apologies to those of you that actually make a living at this stuff - I used to do that too, but find being a hobbyist is much more fun). Comments?

It's not that nanofarads is a hard thing to remember...or figure out....it's just that it's a very UNNATURAL value. Low frequency/power/audio circuits use values on the order of microfarads, while RF circuitry is generally in picofarads. Not many circuits NATURALLY fall into the nanofarad range. In addition, I have NEVER seen a real capacitor of any value labeled in nanofarads.

And it's certainly not a metric vs English "thing." As an old school radio guy, I've always measured wavelength in meters.....one of the few areas where metric IS more convenient. If we all of a sudden started measuring wavelength in fathoms, it would be very awkward. (Incidentally my own personal cubit happens to be PRECISELY half a meter).

Here's one place where nano IS a fairly natural division.....radio waves travel approximately a foot per nanosecond.

Eric

 

Philips ceramic plate capacitors use n markings.

n10 = 100pF
n15 = 150pF
n27 = 270 pF

1n0 = 1.0nF
2n2 = 2.2nF
4n7 = 4.7nF

 

I've seen many many capacitors marked in nanofarads. It is common for the types of film caps I design with in the three decades where it is appropriate (1 nF to "999" nF). It has certainly become more common over the years, but I still have remnants of a stacked polyester film capacitor engineering kit that is probably 30 years old and all except the 1 µF are marked in nanofarads. I've designed caps like those into all sorts of industrial electronics and instrumentation.

 

I think Vloss is undefined unless you have the circuit to which it refers. Every capacitor will have a Vloss at a range of frequencies compared to the impedance it is attached to. Without the circuit and the frequency, Vloss is useless.

I did find that audiophools had a point in saying film caps sound, "better". Metalized film capacitors have the lowest dissipation loss in the audio frequencies. As a sound tech and relative Guru to some musicians, I became a believer right after I did about 6 hours of research. (You can do it faster now that the Internet has been invented.)

 

I went to school to be an "audio engineer" although it turned out that the school I went to taught me audio production techniques and a reasonable amount about acoustics, but nothing at all about electronics... I still like to think of myself as an audio engineer, but it's hardly an accurate term.

Anyway, as I've started learning about electronics in the last few years, I've tried to sort through which audio claims are real and which are bogus.

Seems like film caps are one of the (few?) legit ones for a variety of reasons. Here's a short discussion about non linear voltage coefficients:
https://electronics.stackexchange.c...itor-which-one-is-preferred-in-audio-circuits

And here's a more thorough app note:
https://www.maximintegrated.com/en/app-notes/index.mvp/id/3171

 

"Film" is far to generic. There are numerous plastic films used in capacitor manufacture - polyester ("Mylar" trade name), polypropylene, polysulfone, polycarbonate, polystyrene (no longer being made), and others. Each has different characteristics. One of the major differences is dielectric absorption. DA can cause distortion and errors in signals, but even then the effects tend to vary with specifics of the signal. Polyester is quite bad for DA and unsuitable for things like integrators and sample and hold circuits. But it is cheap and compact. Polypropylene is much better in terms of DA but is more expensive, in spite of the fact that PP is dirt cheap, and the caps are bigger (lower dielectric constant). PP seems to be the darling in audio, and for sound reasons.

oops - as I type ebeowulf's post popped up - that Maxim note is quite a good summary. I think it may have been that paper that first clued me in to how horrible the voltage coefficient of capacitance of some modern ceramic types is

 

Vloss: http://www.foxcomputer.se/Capacitor voltage loss Vloss.pdf

Anyway, Angstroms have gone out of favor too for wavelength.

Yep, the metric system is cool, but you can;t use body parts to estimate anything. In the metric system the numbers are too big. Forget 100 mm. That should be 10 cm. The latter I can "better" visualize.

Now, we can't have an Angstrometer output in units of nm, can we?

Thousanths of an inch or mills work when machinining. Metric and English works when the tools are in the same units.

Now, IC packages have gone metric or should I say, their packages have metric equlivelents, but they are the same part.

Adding insults, I can buy a 7/8" copper fitting at one store and it will be identical to a 3/4" fitting at another. How weird is that?

Computers are not a part of electronics.

 

Computers are not a part of electronics.

it rings true ... but they're its grandchildren ...