Nanofarads?

This can turn into a huge political debate... I once saw an "office memo" announcing that officially everything was the closest engineering unit

*except* 1nF-9.99nF would still be labeled in pF. Sheesh.

I use the closest engineering unit on everything I do... For your examples, I'd label them as 100nF and 1nF; no one born after about 1960 seems to object to this. :-) As far as I can tell, nF is becoming more and more common (and happily "mmf" for pF seems to have died!).

On the other hand, the schematic capture package we use will let you label, e.g., a dozen bypass capacitors, all 100nF, as C[1-12] and then just generates the 12 separate instances in the netlist when you go to lay out a PCB, and I've heard people complain about the use of such practices. Opinions? In my view it's just a degenerate form of hierarchy and should be perfectly legitimate. (But I've also heard people complain about hierarchy... I did a design once that contained 16 100% identical copies of a circuit, and there was insistance that this had to end up as 16 schematic pages, copied and pasted, rather than 1 schematic page with the circuit and one other page instantiating the first page's circuit as a block 16 times over.)

All of this falls into "don't sweat the small stuff," though... one should be thankful if these are the only problems you need to deal with at work!

---Joel

"mmf" died in the mid-1970s, if I recall correctly, and "0.1 uF" still seems more common than "100 nF" although I may just be an old fogey.

Capacitance meters read in nF, which helps to promote the nanofarad.

Right... Today I had to label a cabinet for storing capacitors, knowing it would be used by students who are real newbies and who, for the most part, have a capacitance meter in hand to help them identify the capacitors.

A heirarichal design is great for, say, an FPGA, but every part on a pc board is a distinct physical entity that really needs its own reference designator, so the 16 schematic pages are actually useful. Besides, it makes the block i/o connections (decode selects, connector pins, whatever) easier to find.

John

I usually write uF or pF out of habit, if I'm feeling lazy or am short on space I might write nF, or mF also. Which, unfortunately, creates confusion, because some jackass decided a) to call "MFD" or "mf" as microdfarads, and another jackass declared that "m" be mili, not micro.

(I always write "µF" on schematics when used, so it should be obvious enough to see "mF", as opposed to "µF" everywhere else, that I mean the 1/1000th "m"...)

Tim

--
Deep Fryer: a very philosophical monk.
Website: http://webpages.charter.net/dawill/tmoranwms

"mc"  wrote in message 
news:Ppmkg.14235$gv2.7925@bignews3.bellsouth.net...
> What\'s your personal preference -- do you use nanofarads as a unit of 
> measurement?  100 nF or 0.1 uF?  1 nF or 1000 pF or 0.001 uF?
>
> I can remember when it was mf and mmf.
>
>
>

The best compromise is to use millimicrofarads.

-- Many thanks,

Don Lancaster voice phone: (928)428-4073 Synergetics 3860 West First Street Box 809 Thatcher, AZ 85552 rss:

email: snipped-for-privacy@tinaja.com

Please visit my GURU's LAIR web site at

Hierarchical design *is* very useful for I/C chips. I have some that run 4-5 layers deep.

But a block need only be drawn once though it is replicated multiple times in the implementation.

Also makes it easy on the layout guy... create a cell, then use it repeatedly.

...Jim Thompson

--
|  James E.Thompson, P.E.                           |    mens     |
|  Analog Innovations, Inc.                         |     et      |
|  Analog/Mixed-Signal ASIC\'s and Discrete Systems  |    manus    |
|  Phoenix, Arizona            Voice:(480)460-2350  |             |
|  E-mail Address at Website     Fax:(480)460-2142  |  Brass Rat  |
|       http://www.analog-innovations.com           |    1962     |
             
I love to cook with wine.      Sometimes I even put it in the food.

I thought that Leiden flask condensers were measured in liters...

--
Regards, Joerg

http://www.analogconsultants.com

Hi John,

The 16 "modules" are labeled something like U1-U16, and then the components within them get labeled as, e.g., U1.R2, U12.C6, etc.; hence the refdeses do end up being unique. (In fact, I prefer this sort of reference designator on the silkscreen, since it tells you more than just "R178" does.)

But I take your point...

---Joel

Egoist!!! It all depends on which side of the fence you are and what IS the importance of your "laziness". The end user is a 'moron' and has to be fed with a small spoon. I am sure that you had a "simplified" drawing of some repetitive circuit thrown on your table with "Find what's wrong with it, and we need it yesterday!" and spent time trying to make heads or tails of the overall connectivity. On the other side the worst part of designing is to produce the paper work for those 'morons', things that are self-explanatory to the designer are "UFO science" to the user techs, whatever level of electronics degree they hold.

Been there, done it!

Have fun

Stanislaw Slack user from Ulladulla.

One rule is to use no more than 2 digits before the decimal.

So, rather than 100n it should actually be 0u1. Use of the decimal point is deprecated since it's prone to introducing errros.

One value that always sticks in my mind for some reason is 4n7.

Then again I'd rather go for 820pF than 0n82 !

Take your pick.

Graham

m never was micro in the civilised world.

Graham

Gawd, who uses decimal points any more?, it's 0u1 thankyou! :-P

I'm personally happy with 100nF or 0u1 or 0.1uF and use all three depending upon my mood. If it's below 100nF though I always use nF, so

47nF is never 0.047uF. Anyone who writes 47,000pF these days is an idiot. 1000pF is just barely acceptable, but I'd still give them a serve. Generally we should all standardise on nF, it's a proper engineering unit and deserves equal billing! I'm a fan of the nF.

Dave :)

I've sometimes felt a good interview question for would-be EEs is to sit them in front of one of those European schematics of yours and see how long it takes them to figure out what's going on. (I remember first seeling that in high school, after having gotten ahold of an Elektor magazine.)

As far as I was ever informed, the reason for deprecating decimal points was because they were prone to being lost permanently after going through multiple photocopies. To a large extent, that seems to be a non-problem these days since usually multiple copies are generated one by one from the design file directly to a printer.

---Joel

Are we voting? NOT beying in the 40% of male humans on this planet, who ARE partially or completely color blind, I vote for the long forgotten 6 color designation of capacitors, with proper translation from the German, Dutch, Japanese, French and Yankee coloring books. Especially as nothing is written on the component to identify the country of origin.

So, lets have fun!

Stanislaw Slack user from Ulladulla.

There's an old unit of capacitance-- 'jars', 1 jar ~= 1nF

Best regards, Spehro Pefhany

--
"it\'s the network..."                          "The Journey is the reward"
speff@interlog.com             Info for manufacturers: http://www.trexon.com
Embedded software/hardware/analog  Info for designers:  http://www.speff.com

Hello Spehro,

Ok, then we have it: 0.1uF is 0.1 kilojars.

--
Regards, Joerg

http://www.analogconsultants.com

Most experienced electronics people accept or use either fractional uF or nF without thinking about it too much. When you see fractional uF you simply move the decimal point 3 places to the right to convert to nF and vice-versa.

I'm old, but not *that* old! :)

Even worse is 47000 pF or 1000000 pF, no commas, in small print. Having to count all those zeroes is a real PITA.

Mark