Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss

H&H AoE -- advice re: component value notation

45 views
Skip to first unread message

Winfield Hill

unread,
Feb 15, 2004, 5:14:10 PM2/15/04
to
We are redrafting the figures for the next edition of H&H AoE, and
we're thinking about changing the notation for component values (to
the European scheme) in some of them. For example, we would write 4k7
instead of 4.7k, for a resistor; or 100n instead of 0.1uF, for a
capacitor. We're planning on sticking with the US symbols, however,
eg a zigzag (not a box) for a resistor.

Advantages: less chance of error, avoids tiny decimal points (by
eliminating them altogether), familiarizes readers with both schemes.
Disadvantages: things like 100n look weird to us, we are used to 0.1
or 0.1uF, etc., and 0u1 looks ever weirder!

Questions: What do y'all think of this idea? If we do it, should we
just pick some subset of the figures to change, or maybe just a whole
chapter, or what?

PS: don't hold your breath, on the new editon!

Jim Thompson

unread,
Feb 15, 2004, 5:36:05 PM2/15/04
to
On 15 Feb 2004 14:14:10 -0800, wh...@picovolt.com (Winfield Hill)
wrote:

Why would you use 0u1? What's wrong with 100n? Do the Euro's use
0R1?

...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.

John Larkin

unread,
Feb 15, 2004, 5:59:41 PM2/15/04
to
On 15 Feb 2004 14:14:10 -0800, wh...@picovolt.com (Winfield Hill)
wrote:

>We are redrafting the figures for the next edition of H&H AoE, and


Please, please don't do it! Why should electrical engineering not use
proper scientific notation like all other engineering and scientific
disciplines do? Not even plumbers record measurements this way! Why
would we want to look like a bunch of uneducated freaks?

This is a silly european audio hobbyist thing. What's wrong with
decimal points? All the scientific journals seem to be able to print
them just fine. I don't see 22kg4 anywhere in The review of Scientific
Instruments. I work with scientists a lot, and they would think I'm
nuts if I had to explain this notation to them.

And besides, any time you see 4 7M it should be a clue to put on
your reading glasses.

But personally, I don't mind things like nanofarads; they're handy,
formally correct, and not hard to get used to. We use 2.7nF instead of
0.0027uF. But please, let's use proper scientific notation! If you
don't want to use the ohms or farads symbols, add a note somewhere
"all resistances in Ohms; all inductances in Farads" etc. That's
usually understood either way.

You might mention the notation somewhere, just to help people who come
across it reading Glass Audio or something.


John



Tom Del Rosso

unread,
Feb 15, 2004, 6:00:57 PM2/15/04
to
In news:3e852f41.04021...@posting.google.com,
Winfield Hill typed:

> We are redrafting the figures for the next edition of H&H AoE, and
> we're thinking about changing the notation for component values (to
> the European scheme) in some of them. For example, we would write 4k7
> instead of 4.7k, for a resistor; or 100n instead of 0.1uF, for a
> capacitor. We're planning on sticking with the US symbols, however,
> eg a zigzag (not a box) for a resistor.

The box thing is ridiculous. Why not boxes for caps and transistors
too? Op amps could be shown as rectangular function blocks like most
other ICs.

I don't like the look of 4k7 but it's not so bad. I never have trouble
seeing the decimal points though, except when the print is so bad I
can't read anything.


> Advantages: less chance of error, avoids tiny decimal points (by
> eliminating them altogether), familiarizes readers with both schemes.
> Disadvantages: things like 100n look weird to us, we are used to 0.1
> or 0.1uF, etc., and 0u1 looks ever weirder!

A scaling factor without a unit certainly does look weird. Why not
100nF? I've seen that in popular literature for ages. Only resistors
can get away with having no unit, since a special exception was made for
the omega symbol a long time ago.


--
-Reply in group, but if emailing add 2 more zeros-
-and remove the obvious-


Tim Auton

unread,
Feb 15, 2004, 6:22:46 PM2/15/04
to
wh...@picovolt.com (Winfield Hill) wrote:

>We are redrafting the figures for the next edition of H&H AoE, and
>we're thinking about changing the notation for component values (to
>the European scheme) in some of them. For example, we would write 4k7
>instead of 4.7k, for a resistor; or 100n instead of 0.1uF, for a
>capacitor. We're planning on sticking with the US symbols, however,
>eg a zigzag (not a box) for a resistor.
>
>Advantages:

Fewer syllables: four-kay-seven vs four-point-seven-kay

I suppose it comes down to how many of your readers are outside the US
and how small the font needs to be (the european system is a winner
with really small type).


Tim
--
Love is a travelator.

Gareth

unread,
Feb 15, 2004, 6:59:11 PM2/15/04
to

I think you should certainly mention the 4k7 style notation as many of
your readers are likely to come across it sooner or later, but I don't
think it matters which notation you use as, in my opinion, both are
quite clear.

It should be obvious to anyone with even the most basic scientific
knowledge that 0.1uF = 100nF. I personally prefer to write 100nF as it
avoids the decimal point and the use of a Greek letter, but I find
either perfectly clear. However when you get to smaller values I think
nF is definitely preferable, e.g. in my opinion 1nF is strongly
preferable to 0.001uF. Surely the purpose of these prefixes is to avoid
too many zeros?

I think 0u1 looks strange, though if written next to a capacitor symbol
it would quite obvious what it meant. I find 4u7 for 4.7uF perfectly
acceptable though.

Gareth.

--
-----------------------------------------------------------------------
To reply to me directly:

Replace privacy.net with: totalise DOT co DOT uk and replace me with
gareth.harris

Dan Mills

unread,
Feb 15, 2004, 1:52:01 AM2/15/04
to
Jim Thompson wrote:

<Snip>

> Why would you use 0u1? What's wrong with 100n? Do the Euro's use
> 0R1?

Yep 0.1ohms, fairly standard. A fairer comparison would be asking if
european practice was 0K1 = 100 ohms, which of course you never see.

Regards, Dan.
--
And on the evening of the first day, the lord said.... LX1, Go!
And there was light.
The email address *IS* valid, do not remove the spamblock.

Spehro Pefhany <Spehro Pefhany >

unread,
Feb 15, 2004, 7:07:16 PM2/15/04
to
On Sun, 15 Feb 2004 23:59:11 +0000, the renowned Gareth
<m...@privacy.net> wrote:

>I think you should certainly mention the 4k7 style notation as many of
>your readers are likely to come across it sooner or later, but I don't
>think it matters which notation you use as, in my opinion, both are
>quite clear.
>
>It should be obvious to anyone with even the most basic scientific
>knowledge that 0.1uF = 100nF. I personally prefer to write 100nF as it
>avoids the decimal point and the use of a Greek letter, but I find
>either perfectly clear. However when you get to smaller values I think
>nF is definitely preferable, e.g. in my opinion 1nF is strongly
>preferable to 0.001uF. Surely the purpose of these prefixes is to avoid
>too many zeros?
>
>I think 0u1 looks strange, though if written next to a capacitor symbol
>it would quite obvious what it meant. I find 4u7 for 4.7uF perfectly
>acceptable though.
>Gareth.

I agree. I prefer the squiggle resistors with the Euro notation, and
definely use nF. 100n by a capacitor symbol is perfectly clear, as is
100nF.

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

Frank Bemelman

unread,
Feb 15, 2004, 7:09:35 PM2/15/04
to

"Winfield Hill" <wh...@picovolt.com> schreef in bericht
news:3e852f41.04021...@posting.google.com...

0u1 looks silly here too. 100n(F) is better. I like the 4k7 notation.
Boxes for resistors - I don't mind, the zigsags are all right. In Europe
we also were a bit used to a horizontal line for ground and power. I prefer
the american style of triangles, up for +V and down for -V or ground.

This answer from someone who knows.

;-)


--
Thanks, Frank.
(remove 'x' and 'invalid' when replying by email)


Winfield Hill

unread,
Feb 15, 2004, 6:36:09 PM2/15/04
to
Tim Auton wrote...

>
> I suppose it comes down to how many of your readers are outside
> the US and how small the font needs to be (the european system
> is a winner with really small type).

Our readership is about 50:50. Or publisher is based in England,
but has a huge U.S. presence, with massive offices in New York.

There's no font issue. It's a matter of usage and professionalism.

Thanks,
- Win

whill_at_picovolt-dot-com

Frank Bemelman

unread,
Feb 15, 2004, 7:12:24 PM2/15/04
to
"Spehro Pefhany >" <Spehro Pefhany <spef...@interlogDOTyou.knowwhat>
schreef in bericht news:6n2030tko860sojlv...@4ax.com...

> On Sun, 15 Feb 2004 23:59:11 +0000, the renowned Gareth
> <m...@privacy.net> wrote:
>
> >I think you should certainly mention the 4k7 style notation as many of
> >your readers are likely to come across it sooner or later, but I don't
> >think it matters which notation you use as, in my opinion, both are
> >quite clear.
> >
> >It should be obvious to anyone with even the most basic scientific
> >knowledge that 0.1uF = 100nF. I personally prefer to write 100nF as it
> >avoids the decimal point and the use of a Greek letter, but I find
> >either perfectly clear. However when you get to smaller values I think
> >nF is definitely preferable, e.g. in my opinion 1nF is strongly
> >preferable to 0.001uF. Surely the purpose of these prefixes is to avoid
> >too many zeros?
> >
> >I think 0u1 looks strange, though if written next to a capacitor symbol
> >it would quite obvious what it meant. I find 4u7 for 4.7uF perfectly
> >acceptable though.
> >Gareth.
>
> I agree. I prefer the squiggle resistors with the Euro notation, and
> definely use nF. 100n by a capacitor symbol is perfectly clear, as is
> 100nF.

I could not agree more.

Jan Panteltje

unread,
Feb 15, 2004, 7:47:14 PM2/15/04
to
On a sunny day (15 Feb 2004 14:14:10 -0800) it happened wh...@picovolt.com
(Winfield Hill) wrote in <3e852f41.04021...@posting.google.com>:

>PS: don't hold your breath, on the new editon!

Good idea.
JP

Al Borowski

unread,
Feb 15, 2004, 8:18:31 AM2/15/04
to

> I agree. I prefer the squiggle resistors with the Euro notation, and
> definely use nF. 100n by a capacitor symbol is perfectly clear, as is
> 100nF.
>
> Best regards,
> Spehro Pefhany


Seconded.

Al

red rover

unread,
Feb 15, 2004, 8:10:59 PM2/15/04
to
I think I am in agreement with Gareth, Frank and
Spehro in general. No problems with 4k7 (also 1R1 = 1.1 Ohms)
and much prefer nF. What is the history with that? Why
were nF ignored in the past?

I prefer the zig-zag resistor.

Frank's last comment threw me.

> In Europe we also were a bit used to a horizontal line
> for ground and power. I prefer the american style of triangles,
> up for +V and down for -V or ground.

I currently work for large multi-national. In all my years of working
with different CAD systems and libraries I only recently
encountered an upside down triangle and that was from one
of our European units. I figured it was a European thing. I have
only seen triangles with the point down and only representing
ground, never V- or V+.

The horizontal line I've seen used frequently. Frank, did you get
get American/European backwards in your statement?

Steve

Gary Richardson

unread,
Feb 15, 2004, 9:59:31 PM2/15/04
to

"John Larkin" <jjla...@highlandSNIPtechTHISnologyPLEASE.com> wrote in
message news:d5tv2016kgj6gqiop...@4ax.com...
Well said!


Roy McCammon

unread,
Feb 16, 2004, 2:13:53 AM2/16/04
to

Well, extra decimal points used to be a problem when we used a blue
print process to duplicate drawings. It would pick up specs that
were about the size of a decimal point. So 47K could become 4.7K

But the problem doesn't really exist anymore.

But, 4k7 is elegant, has fewer characters, and is unambiguous.

Heck, do both.


Kevin McMurtrie

unread,
Feb 16, 2004, 3:27:18 AM2/16/04
to
In article <3e852f41.04021...@posting.google.com>,
wh...@picovolt.com (Winfield Hill) wrote:

I think the 4k7 and 0R1 notation would add ugliness to a book that will
be read by many people starting out in electronics. It's one more bit
of unique non-intuitive jargon to learn before getting started on the
basics. 4.7K and 0.1 are much easier to understand. Eliminating the
'.' was a hack to prevent printing errors. It's obsolete and I think
it's time for the practice to go away. (Avoiding 4-way junctions bugs
me too - it makes traces harder to follow.)

As for caps, I don't think it matters where you switch between micro and
nano. I've seen schematics where a section of electronics will stick
with a single notation for a wide range of values, presumably for fast
visual comparisons. Just don't do 0u1!!!!

I'm glad you're going with the zig-zag resistors. It much better
represents what a resistor does than a rectangle.

John Woodgate

unread,
Feb 16, 2004, 3:46:45 AM2/16/04
to
I read in sci.electronics.design that Winfield Hill <wh...@picovolt.com>
wrote (in <3e852f41.04021...@posting.google.com>) about 'H&H
AoE -- advice re: component value notation', on Sun, 15 Feb 2004:

>Questions: What do y'all think of this idea? If we do it, should we
>just pick some subset of the figures to change, or maybe just a whole
>chapter, or what?

I'll accept the wiggly resistor if you accept the nanofarad. I agree
that eliminating the decimal point is a good thing for avoiding
misreading, and you also please the half of the electronics world that
uses the comma instead of the point.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

John Woodgate

unread,
Feb 16, 2004, 3:49:32 AM2/16/04
to
I read in sci.electronics.design that John Larkin <jjlarkin@highlandSNIP
techTHISnologyPLEASE.com> wrote (in <d5tv2016kgj6gqiopmonu2ihgqb9sjnks2@
4ax.com>) about 'H&H AoE -- advice re: component value notation', on
Sun, 15 Feb 2004:

>This is a silly european audio hobbyist thing.

It certainly isn't. I can't cite the official backing for it at present,
but it exists.

>What's wrong with decimal
>points?

They don't reproduce well. Have you EVER seen a baby decimal point?

John Woodgate

unread,
Feb 16, 2004, 3:53:02 AM2/16/04
to
I read in sci.electronics.design that Tom Del Rosso
<tdne...@att.net.invalid> wrote (in <JSSXb.27020$hR.703336@bgtnsc05-new
s.ops.worldnet.att.net>) about 'H&H AoE -- advice re: component value

notation', on Sun, 15 Feb 2004:

>The box thing is ridiculous.

No, it's not ridiculous. It was MUCH easier to draw when circuit
diagrams were hand-drawn.

Please (everyone, not just Tom) at least try to stop slagging off
everything that wasn't invented in USA.

Bill Sloman

unread,
Feb 16, 2004, 4:47:33 AM2/16/04
to
wh...@picovolt.com (Winfield Hill) wrote in message news:<3e852f41.04021...@posting.google.com>...

I like the European notation for resistors. I prefer 100nF to 100n for
capacitors, but 100n and 100p are unambiguous in practice.

I agree with Spehro that any decent printer can print a perfectly
legible decimal point on paper, but once a circuit diagram has been
lying about in the lab for a day or two, "decimal points" seem to
start showing up everywhere :-)
Surface mount components really demand the European-style notation.

Zigzag's for resistors are fine - boxes were easier and quicker to
draw, but that isn't an issue nowdays, and you can put the resistance
value or the part number inside the box, which saves space on the
drawing (which is still an issue).

Given a free choice, I'd recommend going over to the European
convention throughout the book, only retaining a couple of circuits
drawn American style to inform the students about the alternative
style - I certainly never had any trouble interpreting either sort of
circuit diagram. Clearly, just converting a few circuits to European
style would be equally acceptable.

-------
Bill Sloman, Nijmegen

Terry Pinnell

unread,
Feb 16, 2004, 5:27:10 AM2/16/04
to
wh...@picovolt.com (Winfield Hill) wrote:

>We are redrafting the figures for the next edition of H&H AoE, and
>we're thinking about changing the notation for component values (to
>the European scheme) in some of them. For example, we would write 4k7
>instead of 4.7k, for a resistor; or 100n instead of 0.1uF, for a
>capacitor.

Yes please, excellent idea. And would encourage more widespread use.
Decimal points are hard to read reliably even in professionally
printed sources. And scanned images based on sources using the
conventional notation can be very misleading.

>We're planning on sticking with the US symbols, however,
>eg a zigzag (not a box) for a resistor.

That's a great pity. Why? Boxes are much better looking and easier to
distinguish from other components like inductors.

>Advantages: less chance of error, avoids tiny decimal points (by
>eliminating them altogether), familiarizes readers with both schemes.
>Disadvantages: things like 100n look weird to us, we are used to 0.1
>or 0.1uF, etc., and 0u1 looks ever weirder!

I use both. Never quite settled on one or the other. But 100n is
undoubtedly clearer, by avoiding that pesky decimal point again.

>Questions: What do y'all think of this idea? If we do it, should we
>just pick some subset of the figures to change, or maybe just a whole
>chapter, or what?
>
>PS: don't hold your breath, on the new editon!

--
Terry Pinnell
Hobbyist, West Sussex, UK

Stefan Heinzmann

unread,
Feb 16, 2004, 6:02:21 AM2/16/04
to

Since the world at large does not agree on what style to use, I think
there's no need to stick to one particular style in your book. I
wouldn't complain either way, though. You are US authors and thus you
are entitled to do it the US way. An electronics engineer will come
across all different styles eventually, and will have to understand
their meaning.

If you use the US way, however, translated versions of your book will
have to decide individually whether they adapt the style to the local
custom. This may mean that the decimal point gets replaced with a comma,
some schematic symbols need changing (resistors are just one example),
not to speak of symbols for logic gates (the unified international
standard was a failure, unfortunately). If you are worried about that,
you may want to mix styles, in order to make the reader aware of all of
them. This can be kept in translated versions. If you go for it, make it
clear in the introduction, and help the reader to make the connection
between the alternatives, for example in an appendix.

I don't understand why some people have such strong preferences with
resistor symbols or component value style. They're all just conventions,
and equally valid as such. To state that one is more "natural" than the
other is ridiculous, I think. The goal is clarity and conciseness.
Writing 4k7 is both. The only problem is that you may be unaccustomed to
it. But any engineer worth his salt should have no trouble decoding it.
Heck, you can decode the resistor colour code! Is there anything natural
in that? AFAIK it is completely arbitrary. It has been fixed in
prehistoric times and it is now what it is. Luckily the world agrees on
a single scheme.

I am tempted to add that non-native English speakers learn a complete
foreign language in pursuit of their electronics education, so hearing
people complain about having to learn even trivial bits of notation
seems somewhat cheeky.

> PS: don't hold your breath, on the new editon!

I did until I had a blue face. I finally gave up and bought the second
edition.

--
Cheers
Stefan

John Woodgate

unread,
Feb 16, 2004, 6:28:46 AM2/16/04
to
I read in sci.electronics.design that Stefan Heinzmann
<stefan_h...@yahoo.com> wrote (in <c0q7nh$1id$04$1@news.t-
online.com>) about 'H&H AoE -- advice re: component value notation', on
Mon, 16 Feb 2004:

>(the unified international
>standard was a failure, unfortunately).

Well, it was just ridiculously over-complicated. So is the standard for
symbols on controls. It would be at least as easy (and clearly more
useful!) to learn the Chinese characters for the functions, instead of
the contrived symbols in IEC 60417.

Syd Rumpo

unread,
Feb 16, 2004, 6:43:09 AM2/16/04
to
On Sun, 15 Feb 2004 14:59:41 -0800, John Larkin
<jjla...@highlandSNIPtechTHISnologyPLEASE.com> wrote:

[snip]

>This is a silly european audio hobbyist thing.

Nonsense. It's universal here among professionals except some older
people. One good reason is to avoid confusion between the period and
the comma which can have different usages in different countries.

Definitely squiggles for resistors, though.

--
Syd

Kevin Aylward

unread,
Feb 16, 2004, 6:48:39 AM2/16/04
to
Kevin McMurtrie wrote:
> In article <3e852f41.04021...@posting.google.com>,
> wh...@picovolt.com (Winfield Hill) wrote:
>
>> We are redrafting the figures for the next edition of H&H AoE, and
>> we're thinking about changing the notation for component values (to
>> the European scheme) in some of them. For example, we would write
>> 4k7 instead of 4.7k, for a resistor; or 100n instead of 0.1uF, for a
>> capacitor. We're planning on sticking with the US symbols, however,
>> eg a zigzag (not a box) for a resistor.
>>
>> Advantages: less chance of error, avoids tiny decimal points (by
>> eliminating them altogether), familiarizes readers with both schemes.
>> Disadvantages: things like 100n look weird to us, we are used to 0.1
>> or 0.1uF, etc., and 0u1 looks ever weirder!
>>
>> Questions: What do y'all think of this idea? If we do it, should we
>> just pick some subset of the figures to change, or maybe just a whole
>> chapter, or what?
>>
>> PS: don't hold your breath, on the new editon!
>
> I think the 4k7 and 0R1 notation would add ugliness to a book that
> will be read by many people starting out in electronics.

I dont. Your welcome to your opinons though.

> It's one
> more bit of unique non-intuitive jargon to learn before getting
> started on the basics. 4.7K and 0.1 are much easier to understand.

The multiplier for the decimal point is trivial to learn. It takes 1
sec. I can remember when I fist came across it many years ago. I
recognised *immediately* how wonderful it was.

> Eliminating the '.' was a hack to prevent printing errors.

It applies to schematics on computer screens as well. It *is* difficult
to see if there is a decimal point or not on a screen schematic.
Actually *go* and *try* it sometime and it will be trivially obvious. In
the worst case the grid dots on the schematic make it impossible to
determine if there is a decimal point or not.

>It's
> obsolete and I think it's time for the practice to go away.

Nonsense. Its about the next best thing to the invention of sliced
bread.

Kevin Aylward
salesE...@anasoft.co.uk
http://www.anasoft.co.uk
SuperSpice, a very affordable Mixed-Mode
Windows Simulator with Schematic Capture,
Waveform Display, FFT's and Filter Design.

"That which is mostly observed, is that which replicates the most"
http://www.anasoft.co.uk/replicators/index.html

"quotes with no meaning, are meaningless" - Kevin Aylward.


Tim Shoppa

unread,
Feb 16, 2004, 7:18:48 AM2/16/04
to
wh...@picovolt.com (Winfield Hill) wrote in message news:<3e852f41.04021...@posting.google.com>...
> Disadvantages: things like 100n look weird to us, we are used to 0.1
> or 0.1uF, etc., and 0u1 looks ever weirder!

I still think in mF and mmF, so don't ask me about modern notation!

But... isn't it most conventional to not have leading zeroes in
engineering notation? Thus 0u1 *should* look weird, and 100n should
look more normal?

Tim.

Tim Shoppa

unread,
Feb 16, 2004, 7:36:07 AM2/16/04
to
"red rover" <NOnatpr...@sympatico.ca> wrote in message news:<sMUXb.1784$w65.1...@news20.bellglobal.com>...

> I think I am in agreement with Gareth, Frank and
> Spehro in general. No problems with 4k7 (also 1R1 = 1.1 Ohms)
> and much prefer nF. What is the history with that? Why
> were nF ignored in the past?

In at least US electrical engineering, the microfarad was traditionally
denoted the "mF" and the picofarad was called the "mmF" (micromicroFarad).

This was before there were generally accepted terms for "pico" and the
greek mu wasn't yet adopted to mean micro.

Anyway, in the "mF" and "mmF" system there is no easy way to write
10^-9F, unless it is 0.001mF or 1000mmF. When the mu and pico came, they
just replaced "m" with "u" and "mm" with "p". "n" was also well defined, but
didn't catch on right away, and the effects obviously still linger.

It's not that electrical engineering is slow... there are other fields
where the "m" prefix still commonly means micro. I'm surprised it hasn't
killed some patients.

The ignorance of certain SI/metric prefixes is widespread. The "metric tonne"
is still in wide use, despite the fact that "megagram" sounds so much
cooler to me :-). Some scientific journals are catching on to my tastes,
finally.

Tim.

JeffM

unread,
Feb 16, 2004, 8:04:46 AM2/16/04
to
>squiggle resistors with the Euro notation, and...use nF.

I'm with Speff too.

Winfield Hill

unread,
Feb 16, 2004, 7:35:51 AM2/16/04
to
Roy McCammon wrote...

>
> But, 4k7 is elegant, has fewer characters, and is unambiguous.
> Heck, do both.

That was our plan, "If we do it, should we just pick some subset


of the figures to change, or maybe just a whole chapter, or what?"

And that was our question. So, every other figure? Nah. Every
other chapter? Maybe. Groups of figures? Sounds good. Some of
the more complex schematics here and there? Perhaps. Opinions?

Thanks,
- Win

whill_at_picovolt-dot-com

John Woodgate

unread,
Feb 16, 2004, 8:12:24 AM2/16/04
to
I read in sci.electronics.design that Tim Shoppa <shoppa@trailing-
edge.com> wrote (in <bec993c8.04021...@posting.google.com>)

about 'H&H AoE -- advice re: component value notation', on Mon, 16 Feb
2004:
>It's not that electrical engineering is slow... there are other fields
>where the "m" prefix still commonly means micro. I'm surprised it
>hasn't killed some patients.

The medics just went crazy. They adopted the silly unit 'milli-
centigram' mcg. The CIPM does not approve of double multiplier prefixes,
and indeed there is no need for them. The medics can write things like
'Recipe Tabs Potassi Cyan. 5 mg BP t.i.d per rectum' (;-), so surely
they could cope with 'μg' or 'ug' for 'microgram'.

Terry Pinnell

unread,
Feb 16, 2004, 8:33:04 AM2/16/04
to
Winfield Hill <Winfiel...@newsguy.com> wrote:

OK - universally 4k7 would be my strong preference, for reasons I gave
earlier.

Stefan Heinzmann

unread,
Feb 16, 2004, 8:41:24 AM2/16/04
to
Winfield Hill wrote:

> That was our plan, "If we do it, should we just pick some subset
> of the figures to change, or maybe just a whole chapter, or what?"
>
> And that was our question. So, every other figure? Nah. Every
> other chapter? Maybe. Groups of figures? Sounds good. Some of
> the more complex schematics here and there? Perhaps. Opinions?

I haven't got a strong opinion on this, but I worry that if you
alternate randomly you're bound to confuse somebody who comes to think
that there's a clever reason behind this.

What about the following idea:
You stick to the way you use in the current edition. You have got enough
control over the printing quality to not worry about legibility of
decimal points. Since that does not apply to the schematics drawn by the
readers - or anyone else - why don't you present the alternatives in the
appendix on drawing schematics, maybe along with some rationale as
discussed here in the thread?

When you reprint schematics, as on page 362f, you reprint the convention
used by whoever created the drawing, and that's fine. The appendix will
serve to clear up any confusion from this.

I am in favour of using nF, but personally I'd hesitate to use mF,
because the latter can be confused with the old usage of m to mean
micro. There are people who started using mF in earnest, however, as it
is compliant with SI. So people will have to deal with it anyhow.

--
Cheers
Stefan

Stefan Heinzmann

unread,
Feb 16, 2004, 9:09:27 AM2/16/04
to
Frank Bemelman wrote:

> In Europe
> we also were a bit used to a horizontal line for ground and power. I prefer
> the american style of triangles, up for +V and down for -V or ground.

This matter of different ground symbols has confused me many times. Is
there a clear rule for when to use which ground symbol? I assumed the
following meaning so far, am I right?

|
----- Signal Ground American Style
---
-


|
----- Signal Ground European Style


|
----- Chassis Ground (or Earth)
/ / /


|
--- Special purpose ground or reference potential
\ / (i.e. Analog vs. Digital)
V


At least in Germany it seems that a lot of people use the first symbol
as a symbol for Earth. In particular, you can see it encircled to mean
Protective Earth. So I guess there is some confusion, after all. How
does everybody else deal with this?

--
Cheers
Stefan

John Woodgate

unread,
Feb 16, 2004, 9:26:42 AM2/16/04
to
I read in sci.electronics.design that Stefan Heinzmann
<stefan_h...@yahoo.com> wrote (in <c0qimd$kt$00$1@news.t-
online.com>) about 'H&H AoE -- advice re: component value notation', on
Mon, 16 Feb 2004:

>At least in Germany it seems that a lot of people use the first symbol

>as a symbol for Earth. In particular, you can see it encircled to mean
>Protective Earth. So I guess there is some confusion, after all. How
>does everybody else deal with this?

By examining the diagram to see what the symbol(s) need to mean to make
sense.

Winfield Hill

unread,
Feb 16, 2004, 9:21:55 AM2/16/04
to
Stefan Heinzmann wrote...

>
> I am in favour of using nF, but personally I'd hesitate to use mF,
> because the latter can be confused with the old usage of m to mean
> micro. There are people who started using mF in earnest, however,
> as it is compliant with SI.

I've used mF, where limited space required. But I've always had
a nagging feeling that 22mF simply doesn't do such a large part
justice, whereas 22,000uF looks like the serious part it is. :>)

Thanks,
- Win

whill_at_picovolt-dot-com

Syd Rumpo

unread,
Feb 16, 2004, 10:13:09 AM2/16/04
to
On 16 Feb 2004 06:21:55 -0800, Winfield Hill
<Winfiel...@newsguy.com> wrote:

> I've used mF, where limited space required. But I've always had
> a nagging feeling that 22mF simply doesn't do such a large part
> justice, whereas 22,000uF looks like the serious part it is. :>)

But to many Yerpeens, that looks like a *very* accurate 22uF :-)


--
Syd

Jeroen

unread,
Feb 16, 2004, 9:57:42 AM2/16/04
to
Stefan Heinzmann wrote:
> [...] There are people who started using mF in earnest, however, as it
> is compliant with SI. So people will have to deal with it anyhow.
>

That reminds me, when I was a young lad, I walked into the local
parts shop:
- Can I have a 4.7 millifarad capacitor please?
- Sorry, we don't have those.
As I walked to the door, the thought struck me. He wouldn't be so ign...
Turning around, I asked:
- Ehm, do you happen to have 4700 microfarad capacitors?
- Sure, how many do you want? What voltage?
- Just one. 40 Volts. Thank you!

Oh well.

Jeroen

Roy McCammon

unread,
Feb 16, 2004, 10:08:47 AM2/16/04
to

The goal ought to be 50%, but make it easy on yourself; do all
the new stuff in the new way. If that's not enough, I'd want
to see examples in each chapter; it maximizes the opportunity
to see both.


--
local optimization seldom leads to global optimization

my e-mail address is: <my first name> <my last name> AT mmm DOT com

Frank Bemelman

unread,
Feb 16, 2004, 10:33:06 AM2/16/04
to
"red rover" <NOnatpr...@sympatico.ca> schreef in bericht
news:sMUXb.1784$w65.1...@news20.bellglobal.com...

> I think I am in agreement with Gareth, Frank and
> Spehro in general. No problems with 4k7 (also 1R1 = 1.1 Ohms)
> and much prefer nF. What is the history with that? Why
> were nF ignored in the past?
>
> I prefer the zig-zag resistor.
>
> Frank's last comment threw me.

>
> > In Europe we also were a bit used to a horizontal line
> > for ground and power. I prefer the american style of triangles,
> > up for +V and down for -V or ground.
>
> I currently work for large multi-national. In all my years of working
> with different CAD systems and libraries I only recently
> encountered an upside down triangle and that was from one
> of our European units. I figured it was a European thing. I have
> only seen triangles with the point down and only representing
> ground, never V- or V+.
>
> The horizontal line I've seen used frequently. Frank, did you get
> get American/European backwards in your statement?

No, I didn't think so. I guess there are always opposite examples.
In AOE-2 you often see multiple GND symbols, not a large horizontal
line from left to right.

Here's a schematic that I classify as 'European style':

http://www.vego.nl/3/2/3/3_2_3.htm


--
Thanks, Frank.
(remove 'x' and 'invalid' when replying by email)


Frank Bemelman

unread,
Feb 16, 2004, 10:41:04 AM2/16/04
to
"Winfield Hill" <Winfiel...@newsguy.com> schreef in bericht
news:c0qdf...@drn.newsguy.com...

You need one of these:

http://comparestoreprices.co.uk/novelty-gifts/unbranded-executive-decision-maker.asp


;-)

Tom Del Rosso

unread,
Feb 16, 2004, 10:50:11 AM2/16/04
to
In news:36C4YLCuTIMAFw$M...@jmwa.demon.co.uk,
John Woodgate typed:

> I read in sci.electronics.design that Tom Del Rosso
> <tdne...@att.net.invalid> wrote (in
> <JSSXb.27020$hR.703336@bgtnsc05-new s.ops.worldnet.att.net>) about
> 'H&H AoE -- advice re: component value notation', on Sun, 15 Feb 2004:
>
>> The box thing is ridiculous.
>
> No, it's not ridiculous. It was MUCH easier to draw when circuit
> diagrams were hand-drawn.
>
> Please (everyone, not just Tom) at least try to stop slagging off
> everything that wasn't invented in USA.

I didn't mean to demean British industry standards. :)

And I also didn't mean that it never had an advantage. I just don't
think it has one now. I'm sure even the published drawings that are
done without CAD are done on computers, so complex and frequently used
shapes can be inserted easily.


--
-Reply in group, but if emailing add 2 more zeros-
-and remove the obvious-


John Popelish

unread,
Feb 16, 2004, 10:58:26 AM2/16/04
to

A few years ago the nano farads and letter as decimal points jarred me
but I have not only gotten used to them, but have grown fond of them,
and have adopted them for my own schematics.

--
John Popelish

Harry Dellamano

unread,
Feb 16, 2004, 11:01:12 AM2/16/04
to

"Kevin McMurtrie" <mcmu...@dslextreme.com> wrote in message
news:mcmurtri-9AF3D3...@corp-radius.supernews.com...

> In article <3e852f41.04021...@posting.google.com>,
> wh...@picovolt.com (Winfield Hill) wrote:
>
> it's time for the practice to go away. (Avoiding 4-way junctions bugs
> me too - it makes traces harder to follow.)

Avoiding 4-way junctions also bugs me. It is still a good idea with pencil
drawn schematics but CAD drawings have clear dots for connections and 4 way
junctions make the drawing easer (IMHO) to follow. I do all my drawings with
4Ways and some customer, especially the military will not except them due to
some stone age convention.
Also, getting rid of decimal points or commas sounds great. All values need
units except resistors (Ohms). 4nF7 is not pretty but effective.

Regards
Harry

John Larkin

unread,
Feb 16, 2004, 11:00:16 AM2/16/04
to
On Mon, 16 Feb 2004 08:49:32 +0000, John Woodgate
<j...@jmwa.demon.contraspam.yuk> wrote:

>I read in sci.electronics.design that John Larkin <jjlarkin@highlandSNIP
>techTHISnologyPLEASE.com> wrote (in <d5tv2016kgj6gqiopmonu2ihgqb9sjnks2@
>4ax.com>) about 'H&H AoE -- advice re: component value notation', on
>Sun, 15 Feb 2004:
>


>>This is a silly european audio hobbyist thing.
>

>It certainly isn't. I can't cite the official backing for it at present,
>but it exists.
>
>>What's wrong with decimal
>>points?
>
>They don't reproduce well. Have you EVER seen a baby decimal point?

Out West, we have huge herds of robust, hardy decimal points. Maybe
those effite inbred European dp's are harder to raise.

John

John Popelish

unread,
Feb 16, 2004, 11:06:48 AM2/16/04
to

I vote for an early discussion of the history of the various
notations, with cross referenced examples that go back to mmf
capacitor labels and alternate symbologies, but then a select one of
the current standard approaches (whatever pleases your eye) for use in
all the rest of the text, except for reprints of other people's
schematics from data sheets, etc. Consistency makes the book an
easier read.

--
John Popelish

John Popelish

unread,
Feb 16, 2004, 11:09:08 AM2/16/04
to
Winfield Hill wrote:

> I've used mF, where limited space required. But I've always had
> a nagging feeling that 22mF simply doesn't do such a large part
> justice, whereas 22,000uF looks like the serious part it is. :>)

I am also tempted to double the size of the capacitor symbol for such
devices, too. I also often use a smaller than typical symbol for less
than 100 pf capacitors. It just helps me to keep the function in mind
when looking at the schematic.

--
John Popelish

Stefan Heinzmann

unread,
Feb 16, 2004, 11:17:57 AM2/16/04
to
John Woodgate wrote:
> I read in sci.electronics.design that Stefan Heinzmann
> <stefan_h...@yahoo.com> wrote (in <c0qimd$kt$00$1@news.t-
> online.com>) about 'H&H AoE -- advice re: component value notation', on
> Mon, 16 Feb 2004:
>
>
>>At least in Germany it seems that a lot of people use the first symbol
>>as a symbol for Earth. In particular, you can see it encircled to mean
>>Protective Earth. So I guess there is some confusion, after all. How
>>does everybody else deal with this?
>
>
> By examining the diagram to see what the symbol(s) need to mean to make
> sense.

That leaves the cases that don't make sense whatever the symbols mean :-)

But I'm relieved. It means whichever meaning I choose is probably right.

--
Cheers
Stefan

Stefan Heinzmann

unread,
Feb 16, 2004, 11:22:57 AM2/16/04
to
John Popelish wrote:

So how does that scale up to Farads then? Goldcaps with 1F aren't that
big, really.

Maybe you just don't use them...

--
Cheers
Stefan

Mike

unread,
Feb 16, 2004, 11:38:14 AM2/16/04
to
On Mon, 16 Feb 2004 16:01:12 GMT, Harry Dellamano wrote:

>> In article <3e852f41.04021...@posting.google.com>,
>> wh...@picovolt.com (Winfield Hill) wrote:
>>
>> it's time for the practice to go away. (Avoiding 4-way junctions bugs
>> me too - it makes traces harder to follow.)
>
> Avoiding 4-way junctions also bugs me. It is still a good idea with pencil
> drawn schematics but CAD drawings have clear dots for connections and 4 way
> junctions make the drawing easer (IMHO) to follow. I do all my drawings with
> 4Ways and some customer, especially the military will not except them due to
> some stone age convention.

I wouldn't have liked that convention in the stone age, but today I do.

I can't count the number of times that one of the following has happened:

1. I receive a schematic by fax in which every point where two
wires cross looks like a 4-way tie.

2. I receive a schematic that has been reduced and copied several
times, making every point where two wires cross look like a
4-way tie.

3. I receive a schematic in which some well-meaning aesthetically-
inclined engineer has decided that the dots on 4-way ties are
too big, and makes them so small that you can't tell if there's
a dot or not.

4. A C or D size sheet gets printed out on A-size paper, effectively
reducing all the connection dots to the point that they are
invisible.

5. Somebody intended to connect wires at a 4-way tie, but neglected
to do so. Because of 1-4 above, the error remains invisible to
everyone until many dollars are wasted and silicon is received.

I have never worked for the military, I've just been hit too many times by
errors due to unintentional ambiguity caused by 4-way ties.

-- Mike --

Daniel Haude

unread,
Feb 16, 2004, 11:49:29 AM2/16/04
to
On 15 Feb 2004 14:14:10 -0800,
Winfield Hill <wh...@picovolt.com> wrote
in Msg. <3e852f41.04021...@posting.google.com>

Actually I don't think it matters a damn. Although I prefer the 4k7
notation, and I use nanofarads and boxes for resistors, I couldn't tell
you which notation was used in the schematic I read half an hour ago. It
seems to be something that one gets used to no matter what.

> PS: don't hold your breath, on the new editon!

Sometimes I think that all you do on the book is coming up with random
detail questions for this newsgroup to make us all think you're *really*
*hard* at work on AoE3 ;-)

--Daniel

--
"With me is nothing wrong! And with you?" (from r.a.m.p)

Jim Thompson

unread,
Feb 16, 2004, 11:59:19 AM2/16/04
to

I agree with 1-4, BUT #5 shouldn't happen... you should simulate and
do LVS from the same netlist.

...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.

Ben Bradley

unread,
Feb 16, 2004, 11:59:37 AM2/16/04
to
In sci.electronics.design, sho...@trailing-edge.com (Tim Shoppa)
wrote:

>wh...@picovolt.com (Winfield Hill) wrote in message news:<3e852f41.04021...@posting.google.com>...
>> Disadvantages: things like 100n look weird to us, we are used to 0.1
>> or 0.1uF, etc., and 0u1 looks ever weirder!
>
>I still think in mF and mmF, so don't ask me about modern notation!

Whoa, I'm sure that little m means micro, not milli. These changing
standards over generations (see "22mF" elsewhere in this thread) can
really trip up someone who's not aware of them. At least I'm "old"
enough to remember "mmF" trimmer caps, only because I started reading
my dad's ARRL technical books around age 10.
But what's three or six orders of magnitude between engineers? :)

>But... isn't it most conventional to not have leading zeroes in
>engineering notation? Thus 0u1 *should* look weird, and 100n should
>look more normal?

Someone already mentioned that electronics doesn't follow either
scientific nor standard engineering notation...

>Tim.

-----
http://mindspring.com/~benbradley

John Larkin

unread,
Feb 16, 2004, 12:40:53 PM2/16/04
to
On 16 Feb 2004 04:36:07 -0800, sho...@trailing-edge.com (Tim Shoppa)
wrote:

>"red rover" <NOnatpr...@sympatico.ca> wrote in message news:<sMUXb.1784$w65.1...@news20.bellglobal.com>...


>> I think I am in agreement with Gareth, Frank and
>> Spehro in general. No problems with 4k7 (also 1R1 = 1.1 Ohms)
>> and much prefer nF. What is the history with that? Why
>> were nF ignored in the past?
>

>In at least US electrical engineering, the microfarad was traditionally
>denoted the "mF" and the picofarad was called the "mmF" (micromicroFarad).
>

I have a modest collection of electronics books and mags going back to
the early 1920's. The pre-1940 stuff mixes MFD, mfd, ufd, and uf,
sometimes in the same book. Simetimes the u is a proper 'mu',
especially if hand-drawn. The MIT RadLab books (1946) are
inconsistant... sometimes 1000uuf, sometimes just .001

The Radiotron books (Australian) used uF and uuF, with real 'mu'
symbols.

My earliest book using pF is British, Lewis and Wells' wonderful
"Millimicrosecond Pulse Techniques" of 1959.

Just my 0$02 worth.

John

John Larkin

unread,
Feb 16, 2004, 1:03:28 PM2/16/04
to
On 16 Feb 2004 04:18:48 -0800, sho...@trailing-edge.com (Tim Shoppa)
wrote:

>wh...@picovolt.com (Winfield Hill) wrote in message news:<3e852f41.04021...@posting.google.com>...


>> Disadvantages: things like 100n look weird to us, we are used to 0.1
>> or 0.1uF, etc., and 0u1 looks ever weirder!
>
>I still think in mF and mmF, so don't ask me about modern notation!
>

>But... isn't it most conventional to not have leading zeroes in
>engineering notation? Thus 0u1 *should* look weird, and 100n should
>look more normal?
>

>Tim.

Conventional scientific notation never begins a numeric value with a
decimal point. 0.99 Hz not .99 Hz

In real life, it's almost impossible to lose a decimal point. And
anyhow, in a real manufacturing environment, schematics are just
reference drawings... nothing is built from schematics. Since we build
various versions ("dash numbers") of our products, and component
values can change after a schematic is released, our schematics are
not even expected to have "correct" component values. Only the bill of
materials of a given assembly is expected to be correct.

Why doesn't AoE (or virtually any other engineering book, or any
engineering school) address the realities of engineering
documentation? After all, engineers don't purchase parts or pull kits
or stuff boards: they produce documents.

We need another book, BoE; the Business of Electronics.

John

Terry Pinnell

unread,
Feb 16, 2004, 1:05:14 PM2/16/04
to
Winfield Hill <Winfiel...@newsguy.com> wrote:

>Roy McCammon wrote...
>>
>> But, 4k7 is elegant, has fewer characters, and is unambiguous.
>> Heck, do both.
>
> That was our plan, "If we do it, should we just pick some subset
> of the figures to change, or maybe just a whole chapter, or what?"
>
> And that was our question. So, every other figure? Nah. Every
> other chapter? Maybe. Groups of figures? Sounds good. Some of
> the more complex schematics here and there? Perhaps. Opinions?


Win: You might be interested in this reprint by Ray Marston from an
article he published in 'Nuts & Volts' magazine, 1998, as it makes
some comparisons of the different styles and notations:

http://www.nutsvolts.com/PDF_Files/circuit.pdf

Costas Vlachos

unread,
Feb 16, 2004, 1:06:23 PM2/16/04
to
"Winfield Hill" <Winfiel...@newsguy.com> wrote in message
news:c0qdf...@drn.newsguy.com...

> Roy McCammon wrote...
> >
> > But, 4k7 is elegant, has fewer characters, and is unambiguous.
> > Heck, do both.
>
> That was our plan, "If we do it, should we just pick some subset
> of the figures to change, or maybe just a whole chapter, or what?"
>
> And that was our question. So, every other figure? Nah. Every
> other chapter? Maybe. Groups of figures? Sounds good. Some of
> the more complex schematics here and there? Perhaps. Opinions?
>
> Thanks,
> - Win
>
> whill_at_picovolt-dot-com


I don't think it's a good idea to use mixed notation/style in different
parts of the book. It wouldn't be consistent that way. I'd decide on a good
notation/style (perhaps a mixture of the good parts of the US and European
styles) and stick to it throughout the book. Exceptions to this could be any
schematics that are reprinted with permission, such as the Tandy switching
PSU (AoE, 2nd Ed., pp. 362-363), where readers can see the alternative
styles the manufacturers use.

You could add a small section in the beginning of the book explaining the
different possible styles (US/European, etc.). And please keep using the
real Greek symbols for micro, Ohms, etc., wherever possible. Also, try to
avoid using too many zeros in component values. For example, in AoE (2nd
Ed.), in the circuit on p. 1058, the value of C1 as "0.001" is not very
nice. You could use 1nF instead.

cheers,
Costas
_________________________________________________
Costas Vlachos Email: c-X-v...@hot-X-mail.com
SPAM-TRAPPED: Please remove "-X-" before replying


Klaus Vestergaard Kragelund

unread,
Feb 16, 2004, 2:00:59 PM2/16/04
to
"Gareth" <m...@privacy.net> wrote in message
news:c0p14n$1a45aa$1...@ID-211380.news.uni-berlin.de...
> Winfield Hill wrote:
> > We are redrafting the figures for the next edition of H&H AoE, and
> > we're thinking about changing the notation for component values (to
> > the European scheme) in some of them. For example, we would write 4k7
> > instead of 4.7k, for a resistor; or 100n instead of 0.1uF, for a
> > capacitor. We're planning on sticking with the US symbols, however,
> > eg a zigzag (not a box) for a resistor.
> >
> > Advantages: less chance of error, avoids tiny decimal points (by
> > eliminating them altogether), familiarizes readers with both schemes.

> > Disadvantages: things like 100n look weird to us, we are used to 0.1
> > or 0.1uF, etc., and 0u1 looks ever weirder!
> >
> > Questions: What do y'all think of this idea? If we do it, should we

> > just pick some subset of the figures to change, or maybe just a whole
> > chapter, or what?
>
> I think you should certainly mention the 4k7 style notation as many of
> your readers are likely to come across it sooner or later, but I don't
> think it matters which notation you use as, in my opinion, both are
> quite clear.
>
> It should be obvious to anyone with even the most basic scientific
> knowledge that 0.1uF = 100nF. I personally prefer to write 100nF as it
> avoids the decimal point and the use of a Greek letter, but I find
> either perfectly clear. However when you get to smaller values I think
> nF is definitely preferable, e.g. in my opinion 1nF is strongly
> preferable to 0.001uF. Surely the purpose of these prefixes is to avoid
> too many zeros?
>
> I think 0u1 looks strange, though if written next to a capacitor symbol
> it would quite obvious what it meant. I find 4u7 for 4.7uF perfectly
> acceptable though.
>
> Gareth.
>
> --

I second that :-)

Cheers

Klaus


Klaus Vestergaard Kragelund

unread,
Feb 16, 2004, 2:04:49 PM2/16/04
to
"John Woodgate" <j...@jmwa.demon.contraspam.yuk> wrote in message
news:36C4YLCuTIMAFw$M...@jmwa.demon.co.uk...

> I read in sci.electronics.design that Tom Del Rosso
> <tdne...@att.net.invalid> wrote (in <JSSXb.27020$hR.703336@bgtnsc05-new
> s.ops.worldnet.att.net>) about 'H&H AoE -- advice re: component value

> notation', on Sun, 15 Feb 2004:
>
> >The box thing is ridiculous.
>
> No, it's not ridiculous. It was MUCH easier to draw when circuit
> diagrams were hand-drawn.
>
> Please (everyone, not just Tom) at least try to stop slagging off
> everything that wasn't invented in USA.
> --

I like the box too - from my days when I was a kid. I often saw audio amp
schematics were the resistor looked like a line. Due to bad copy and bad
practice of the drawer.

And otherwise well said, John :-)

Cheers

Klaus


John Fields

unread,
Feb 16, 2004, 2:31:40 PM2/16/04
to
On Mon, 16 Feb 2004 16:01:12 GMT, "Harry Dellamano"
<har...@tdsystems.org> wrote:

>
>"Kevin McMurtrie" <mcmu...@dslextreme.com> wrote in message
>news:mcmurtri-9AF3D3...@corp-radius.supernews.com...
>> In article <3e852f41.04021...@posting.google.com>,
>> wh...@picovolt.com (Winfield Hill) wrote:
>>
>> it's time for the practice to go away. (Avoiding 4-way junctions bugs
>> me too - it makes traces harder to follow.)
>
> Avoiding 4-way junctions also bugs me. It is still a good idea with pencil
>drawn schematics but CAD drawings have clear dots for connections and 4 way
>junctions make the drawing easer (IMHO) to follow. I do all my drawings with
>4Ways and some customer, especially the military will not except them due to
>some stone age convention.

---
The stone age convention is based on doing the most possible to keep
from getting one's own forces killed, and is based, rightfully, on the
premise that draftsmen don't know what they're doing. When two
conductors cross and there's no dot at the crosspoint, if 4-way
connections are allowed there'll always be the possibility that someone
forgot the dot and the resultant confusion from the ambiguity of "Is
this really a junction or are the wires only crossing?". Worse, if a
draftsman sees the wires crossing and, for some ungodly reason, thinks
they're supposed to be connected, well... you can guess the rest. If
4-way connections are prohibited, however, then they can only be wires
crossing without being connected, and if there's a dot there then it's
easily picked up as an error and fixed with out much ado about anything.
---

> Also, getting rid of decimal points or commas sounds great. All values need
>units except resistors (Ohms). 4nF7 is not pretty but effective.

---
I mostly agree, except that if it's clearly a cap then the 'F' probably
isn't needed and 4n7 will identify the value nicely. After all, we
don't write 4kR7...

--
John Fields

JeffM

unread,
Feb 16, 2004, 3:29:15 PM2/16/04
to
>Avoiding 4-way junctions also bugs me.
>...and 4 way junctions make the drawing easer (IMHO) to follow.
>Harry Dellamano

I couldn't disagree more. I say the least ambiguous is best.
Avoiding dots altogether gets you there.


>CAD drawings have clear dots for connections

There ought to be a law.
Terry Pinnell already mentioned difficulty with scanned-in schematics
and Kevin Aylward mentioned dots disappearing on some screens.

Zak

unread,
Feb 16, 2004, 3:52:52 PM2/16/04
to
Tom Del Rosso wrote:

> I don't like the look of 4k7 but it's not so bad. I never have trouble
> seeing the decimal points though, except when the print is so bad I
> can't read anything.

It is also nice to read aloud. Four Kah Seven, instead of 4 dot seven
kilo-ohms. But that may depend on what one is used to.

It is not ambiguous, and saves some space.


Thomas

John Woodgate

unread,
Feb 16, 2004, 3:46:22 PM2/16/04
to
I read in sci.electronics.design that Winfield Hill
<Winfiel...@newsguy.com> wrote (in <c0qjm...@drn.newsguy.com>)
about 'H&H AoE -- advice re: component value notation', on Mon, 16 Feb
2004:

>I've used mF, where limited space required. But I've always had
> a nagging feeling that 22mF simply doesn't do such a large part
> justice, whereas 22,000uF looks like the serious part it is. :>)

Yes. Not yet available in smaller than 1210, unfortunately.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

John Woodgate

unread,
Feb 16, 2004, 3:48:36 PM2/16/04
to
I read in sci.electronics.design that Roy McCammon <rbmcc...@mmm.com>
wrote (in <4030DCFF...@mmm.com>) about 'H&H AoE -- advice re:

component value notation', on Mon, 16 Feb 2004:

>The goal ought to be 50%, but make it easy on yourself; do all


>the new stuff in the new way. If that's not enough, I'd want
>to see examples in each chapter; it maximizes the opportunity
>to see both.

Why not have them dynamically re-configurable, using electronic ink?
(;-)

Guy Macon

unread,
Feb 16, 2004, 4:25:11 PM2/16/04
to

Harry Dellamano <har...@tdsystems.org> says...

>[Avoiding 4-way junctions] is still a good idea with pencil drawn
>schematics

When I draw a schematic by pencil (pen, actually; I draw schematics
the way mozart wrote sheet music...) I avoid 4-way junctions, put
dots on 3-way junctions, and draw little half circle jumpovers when
two lines cross. The time wasted by having a technician get it wrong
just once is larger than the time used by a lifetime of putting in
redundant dots and jumpovers.

(Unlike Mozart, I do own own a white-out pen, just in case I need
to change a schematic because of someone elses error <grin>.)


--
Guy Macon, Electronics Engineer & Project Manager for hire.
Remember Doc Brown from the _Back to the Future_ movies? Do you
have an "impossible" engineering project that only someone like
Doc Brown can solve? My resume is at http://www.guymacon.com/

Guy Macon

unread,
Feb 16, 2004, 4:55:44 PM2/16/04
to

Stefan Heinzmann <stefan_h...@yahoo.com> says...

>I don't understand why some people have such strong preferences with
>resistor symbols or component value style. They're all just conventions,
>and equally valid as such. To state that one is more "natural" than the
>other is ridiculous, I think. The goal is clarity and conciseness.
>Writing 4k7 is both. The only problem is that you may be unaccustomed to
>it. But any engineer worth his salt should have no trouble decoding it.

The words we use are also "just conventions", and of course other
countries use other conventions, but still, it hinders communication
if we don't use the same words. Yes, you can decide to use non-standard
fleemishes and the reader can still gloork the meaning from the context,
but there ix a limit; If too many ot the vleeps are changed, it becomes
harder and qixer to fllf what the wethcz is blorping, and evenually izs
is bkb possible to ghilred frok at wifx. Dnighth? Ngfipht yk ur! Uvq
the hhvd or hnnngh. Blorgk! Blorgk! Blorgkity-blorgk!!!!

Guy Macon

unread,
Feb 16, 2004, 5:00:50 PM2/16/04
to

Kevin Aylward <kevindotayl...@anasoft.co.uk> says...

>It applies to schematics on computer screens as well. It *is* difficult
>to see if there is a decimal point or not on a screen schematic.
>Actually *go* and *try* it sometime and it will be trivially obvious. In
>the worst case the grid dots on the schematic make it impossible to
>determine if there is a decimal point or not.

I have no problem telling "10K" and "1 0K" apart, even if I can't
see the decimal point.

Mike

unread,
Feb 16, 2004, 5:00:51 PM2/16/04
to

We do... the cases where it fails are both relatively rare, but costly:

a) the MOS gates that were supposed to be connected to the bias point
floated to a point that things appeared to work in simulation. They never
did that in silicon...

b) a just-in-time change was made as the wafers reached metal. We were
working furiously on the next chip, even before the current silicon was
received. The change seemed so simple, that we just taped out the new mask
layers and went back to work on the next chip...

In the first case, we seemingly did everything right. If someone had
checked every voltage in the circuit, the problem might have been caught,
but they probably would have looked at the bias node, and everything would
have looked okay. Without knowing that the gates were disconnected from the
bias node, there's no way to know that they are now a different net.

I think there's an option in HSPICE to hang a large resistance (teraohms or
so) from every node to ground. If that had been done, the gates might not
have been biased to some other voltage during DC convergence. We usually
don't use that option, though.

-- Mike --

Terry Pinnell

unread,
Feb 16, 2004, 5:16:27 PM2/16/04
to
Guy Macon <http://www.guymacon.com> wrote:

>
>Stefan Heinzmann <stefan_h...@yahoo.com> says...
>
>>I don't understand why some people have such strong preferences with
>>resistor symbols or component value style. They're all just conventions,
>>and equally valid as such. To state that one is more "natural" than the
>>other is ridiculous, I think. The goal is clarity and conciseness.
>>Writing 4k7 is both. The only problem is that you may be unaccustomed to
>>it. But any engineer worth his salt should have no trouble decoding it.
>
>The words we use are also "just conventions", and of course other
>countries use other conventions, but still, it hinders communication
>if we don't use the same words. Yes, you can decide to use non-standard
>fleemishes and the reader can still gloork the meaning from the context,
>but there ix a limit; If too many ot the vleeps are changed, it becomes
>harder and qixer to fllf what the wethcz is blorping, and evenually izs
>is bkb possible to ghilred frok at wifx. Dnighth? Ngfipht yk ur! Uvq
>the hhvd or hnnngh. Blorgk! Blorgk! Blorgkity-blorgk!!!!

LXL!

Winfield Hill

unread,
Feb 16, 2004, 4:57:25 PM2/16/04
to
Terry Pinnell wrote...

>
> Win: You might be interested in this reprint by Ray Marston from an
> article he published in 'Nuts & Volts' magazine, 1998, as it makes
> some comparisons of the different styles and notations:
>
> http://www.nutsvolts.com/PDF_Files/circuit.pdf

That's a good one!

Thanks,
- Win

whill_at_picovolt-dot-com

Winfield Hill

unread,
Feb 16, 2004, 4:54:37 PM2/16/04
to
John Popelish wrote...

Right, and whereas a resistor is simply --/\/\--
one feels a serious
250W power resistor //\\ //\\
should be something ====// \\ // \\====
more like this, right? \\//

Thanks,
- Win

whill_at_picovolt-dot-com

Stefan Heinzmann

unread,
Feb 16, 2004, 5:39:45 PM2/16/04
to
Guy Macon wrote:

> Stefan Heinzmann <stefan_h...@yahoo.com> says...
>
>
>>I don't understand why some people have such strong preferences with
>>resistor symbols or component value style. They're all just conventions,
>>and equally valid as such. To state that one is more "natural" than the
>>other is ridiculous, I think. The goal is clarity and conciseness.
>>Writing 4k7 is both. The only problem is that you may be unaccustomed to
>>it. But any engineer worth his salt should have no trouble decoding it.
>
>
> The words we use are also "just conventions", and of course other
> countries use other conventions, but still, it hinders communication
> if we don't use the same words. Yes, you can decide to use non-standard
> fleemishes and the reader can still gloork the meaning from the context,
> but there ix a limit; If too many ot the vleeps are changed, it becomes
> harder and qixer to fllf what the wethcz is blorping, and evenually izs
> is bkb possible to ghilred frok at wifx. Dnighth? Ngfipht yk ur! Uvq
> the hhvd or hnnngh. Blorgk! Blorgk! Blorgkity-blorgk!!!!

Xifo zpv tbz opo-tuboebse, zpv sfbmmz nfbo opo-VT-tuboebse, sjhiu?

--
Cheers
Stefan

John Larkin

unread,
Feb 16, 2004, 6:01:19 PM2/16/04
to
On 16 Feb 2004 13:57:25 -0800, Winfield Hill
<Winfiel...@newsguy.com> wrote:

While England is a cute little island, how did the opinion of the
British Standards Organization come to be The International Style?

John

Guy Macon

unread,
Feb 16, 2004, 6:02:08 PM2/16/04
to

The world of Science uses the following prefixes along
with units such as Newton, Ohm, Candela, and Hertz:
X(enna), Y(otta), Z(etta), E(xa), P(eta), T(era),
G(iga), M(ega), k(ilo), m(illi), µ(micro), n(ano),
p(ico), f(emto), a(tto), z(epto), y(octo), and x(enno).
Number then prefix then unit is the standard.

_The Art of Electronics_ is an extremely influential book
among engineers and technicians. As such, it should
strive to move engineers and technicians in the direction
of naming things the way scientists name things. (It's
O.K if you keep the name instead of changing it to
_The Science of Electronics_... <grin>)

In addition, engineers and technicians do not, as a
rule, use terms such as "4k7" in conversation, when they
write technical documents, or when they post to USENET.
They use "4.7K." _The Art of Electronics_ should strive
to move engineers and technicians in the direction of
using the same terminology in schematics, speech, and
written communication.

Even in schematics, the use of the number+prefix+unit
standard is used for units such as Hertz or Amperes.
Nobody writes 4M77Hz next to a crystal or 3m3A next to
a constant current diode. They write 4.77Mhz and
3.3mA.

Guy Macon

unread,
Feb 16, 2004, 6:08:45 PM2/16/04
to

John Popelish <jpop...@rica.net> says...

>I am also tempted to double the size of the capacitor symbol for such
>devices, too. I also often use a smaller than typical symbol for less
>than 100 pf capacitors. It just helps me to keep the function in mind
>when looking at the schematic.

Ah, but do you make 10 Meg resistors longer and 10 Ohm resistors shorter?

John Popelish

unread,
Feb 16, 2004, 6:19:52 PM2/16/04
to

And with little wavey lines above it representing heat convecting
away. ;-)

--
John Popelish

John Popelish

unread,
Feb 16, 2004, 6:20:39 PM2/16/04
to
Guy Macon wrote:
>
> John Popelish <jpop...@rica.net> says...
>
> >I am also tempted to double the size of the capacitor symbol for such
> >devices, too. I also often use a smaller than typical symbol for less
> >than 100 pf capacitors. It just helps me to keep the function in mind
> >when looking at the schematic.
>
> Ah, but do you make 10 Meg resistors longer and 10 Ohm resistors shorter?

Don't do that till 100 meg. ;-)

--
John Popelish

Ken Finney

unread,
Feb 16, 2004, 7:01:39 PM2/16/04
to

"Winfield Hill" <wh...@picovolt.com> wrote in message
news:3e852f41.04021...@posting.google.com...

> We are redrafting the figures for the next edition of H&H AoE, and
> we're thinking about changing the notation for component values (to
> the European scheme) in some of them. For example, we would write 4k7
> instead of 4.7k, for a resistor; or 100n instead of 0.1uF, for a
> capacitor. We're planning on sticking with the US symbols, however,
> eg a zigzag (not a box) for a resistor.
>
> Advantages: less chance of error, avoids tiny decimal points (by
> eliminating them altogether), familiarizes readers with both schemes.
> Disadvantages: things like 100n look weird to us, we are used to 0.1
> or 0.1uF, etc., and 0u1 looks ever weirder!
>
> Questions: What do y'all think of this idea? If we do it, should we
> just pick some subset of the figures to change, or maybe just a whole
> chapter, or what?
>
> PS: don't hold your breath, on the new editon!

Whatever you decide to do, please coordinate it with JEDEC and
ANSI. I'd hate to have the have money held back on a contract
because some new hires followed the AoE convention, and the
contract required us to follow JEDEC and ANSI conventions.


Jan Panteltje

unread,
Feb 16, 2004, 7:26:56 PM2/16/04
to
On a sunny day (Mon, 16 Feb 2004 21:52:52 +0100) it happened Zak
<sp...@jutezak.invalid> wrote in <F4aYb.522$EV2.3636@amstwist00>:

Why don't we just draw the colored bands in the diagrams ;-)?
These days with inkjet color printers no problem.

Andreas Hadler

unread,
Feb 16, 2004, 7:41:01 PM2/16/04
to
Guy Macon <http://www.guymacon.com> wrote:

>In addition, engineers and technicians do not, as a
>rule, use terms such as "4k7" in conversation, when they
>write technical documents, or when they post to USENET.

AS A RULE ?

Am I missing a smiley?

All engineers and technician I've worked with, nearly all european I
admit, prefer 4k7, even in technical documentation, except when
talking to or writing for the non-initiated, e.g. sales or users ;-).
Even the unit is included in these official cases.

May be, the rule is yours only. Please feel free to stay with it, but
don't try to tell me I've only worked with people disobeying The Rule.

>They use "4.7K." _The Art of Electronics_ should strive
>to move engineers and technicians in the direction of
>using the same terminology in schematics, speech, and
>written communication.

AOE should strive for technical clearness and common use, IMHO.

As life is, the common use has changed from 10mmF to 10p.

I don't mind AOE showing variety in usage, and maybe, a change in
time. If there's a tradeoff between working on updating schematics, or
on updating text and/or presenting new schematics, I'd stay with the
old schematics. If they are worth to be understood - and the old ones
are - 4.7k will be understood as well as 4k7. For new ones, nowadays,
4k7 is the way to go IMHO.

If I got Win right, the old schematics have to be computerized. I
would try to avoid spending time and concentration on changing the
notation. Win has too much contents to tell, and has a way too good
style of writing, to be dogmatic on notation, IMHO.

aha
--
The sooner you get behind, the more time you'll have to catch up.


JeffM

unread,
Feb 16, 2004, 7:44:24 PM2/16/04
to
>there's no need to stick to one particular style
>Stefan Heinzmann

My mantra: Consistency is your friend. Consistency is your friend.
(I'm only working with 4.77MHz wetware.
I'll take anything that makes life easier.)


>non-native English speakers [should] learn a complete foreign language
>in pursuit of their electronics education

Native English speakers too. 8-)
It's too bad we're not all immersed in a 2nd language early on
before that part of the brain atrophies.

John Larkin

unread,
Feb 16, 2004, 8:00:01 PM2/16/04
to
On Mon, 16 Feb 2004 15:02:08 -0800, Guy Macon
<http://www.guymacon.com> wrote:


>
>Even in schematics, the use of the number+prefix+unit
>standard is used for units such as Hertz or Amperes.
>Nobody writes 4M77Hz next to a crystal or 3m3A next to
>a constant current diode. They write 4.77Mhz and
>3.3mA.

Hmmm, if you adapt 3v3 for 3.3 volts (which people seem to do) how do
you express 3.3 mv or 3.3 uv? Or 3.3 milliohms?

And why, of all the branches of science and engineering, do we want to
be the only goofballs who use a weird units notation?


John


Ben Bradley

unread,
Feb 16, 2004, 8:02:26 PM2/16/04
to
In sci.electronics.design, Andreas Hadler <Andreas...@t-online.de>
wrote:

>Guy Macon <http://www.guymacon.com> wrote:

>>They use "4.7K." _The Art of Electronics_ should strive
>>to move engineers and technicians in the direction of
>>using the same terminology in schematics, speech, and
>>written communication.
>
>AOE should strive for technical clearness and common use, IMHO.
>
>As life is, the common use has changed from 10mmF to 10p.

I think there should be a page or table or something explaining
different notations, including historical notations. I think there's a
significant chance that new engineers reading old books would read
10mmF as "ten millimilliFarads" or ten microfarads if they've never
seen it before, and it would be good for a book to explain "Here's
what it really means."
For another point, I really don't think it's neccesary to have a
CPS-to-Hz conversion table (especially since someone thoughtfully put
one on the web), but it would be good to describe cycles-per-second,
cps and c/s, and tell when the "official" change to Hertz (abbreviated
Hz) was done (I forget exactly, I know it was '70's or maybe '80s).
Have a whole section on "International differences in and historical
changes in electronic nomenclature."

>I don't mind AOE showing variety in usage, and maybe, a change in
>time. If there's a tradeoff between working on updating schematics, or
>on updating text and/or presenting new schematics, I'd stay with the
>old schematics. If they are worth to be understood - and the old ones
>are - 4.7k will be understood as well as 4k7. For new ones, nowadays,
>4k7 is the way to go IMHO.

This thread shows significant disagreement over that. I don't have
a strong opinion either way, but I do have a strong opinion that
somewhere these differences should be listed and explained (with at
least one 'alternative' schematic).

>If I got Win right, the old schematics have to be computerized. I
>would try to avoid spending time and concentration on changing the
>notation. Win has too much contents to tell, and has a way too good
>style of writing, to be dogmatic on notation, IMHO.
>
>aha

-----
http://mindspring.com/~benbradley

Tim Auton

unread,
Feb 16, 2004, 9:08:34 PM2/16/04
to
John Larkin <jjla...@highlandSNIPtechTHISnologyPLEASE.com> wrote:
[snip]

>And why, of all the branches of science and engineering, do we want to
>be the only goofballs who use a weird units notation?

Perhaps because nobody else has to draw and work from schematics, with
all the problems that brings? If mathematicians sometimes had to work
from a single faded copy of a proof that had been reduced to 1/2 size
and had a tea stain on it they too would want more robust notation.


Tim
--
Love is a travelator.

Guy Macon

unread,
Feb 16, 2004, 10:02:29 PM2/16/04
to

Andreas Hadler <Andreas...@t-online.de> says...

>
>Guy Macon <http://www.guymacon.com> wrote:
>
>>In addition, engineers and technicians do not, as a
>>rule, use terms such as "4k7" in conversation, when they
>>write technical documents, or when they post to USENET.
>
>AS A RULE ?
>
>Am I missing a smiley?
>
>All engineers and technician I've worked with, nearly all european I
>admit, prefer 4k7, even in technical documentation, except when
>talking to or writing for the non-initiated, e.g. sales or users ;-).
>Even the unit is included in these official cases.

My error. I have worked with various European engineers here
in California and they behaved as I described I must assume
that they were adapting to local custom.

Guy Macon

unread,
Feb 16, 2004, 10:03:48 PM2/16/04
to

JeffM <jef...@email.com> says...

>
>>there's no need to stick to one particular style
>
>My mantra: Consistency is your friend. Consistency is your friend.
>(I'm only working with 4.77MHz wetware.
>I'll take anything that makes life easier.)

Don't you mean 4M77 Hz? :)

Tom Del Rosso

unread,
Feb 16, 2004, 10:04:30 PM2/16/04
to
In news:l2q1309hc83suc8cu...@4ax.com,
John Larkin typed:
>>
>>> What's wrong with decimal
>>> points?
>>
>> They don't reproduce well. Have you EVER seen a baby decimal point?
>
> Out West, we have huge herds of robust, hardy decimal points. Maybe
> those effite inbred European dp's are harder to raise.

I never saw a baby decimal point, but I've heard a Baby Elephant Walk.


--
-Reply in group, but if emailing add 2 more zeros-
-and remove the obvious-


Tom Del Rosso

unread,
Feb 16, 2004, 10:04:31 PM2/16/04
to
In news:10769776...@news-01.evisp.enertel.nl,
Jan Panteltje typed:

>
> Thomas
> Why don't we just draw the colored bands in the diagrams ;-)?
> These days with inkjet color printers no problem.

Not a problem, but it's even more cryptic. If there was room on a
resistor (and there is less so all the time :)) then you'd just print
the number on it.

I could get used to having the unit multiplier where the decimal point
belongs, but why is it necessary if they don't do that with printed
numbers in other fields of engineering?

Guy Macon

unread,
Feb 16, 2004, 10:04:38 PM2/16/04
to
John Larkin <jjla...@highlandSNIPtechTHISnologyPLEASE.com> says...

>And why, of all the branches of science and engineering, do we want to
>be the only goofballs who use a weird units notation?

Why indeed?

Guy Macon

unread,
Feb 16, 2004, 10:06:09 PM2/16/04
to
Ben Bradley <ben_nospa...@mindspring.example.com> says...

> I think there should be a page or table or something explaining
>different notations, including historical notations. I think there's a
>significant chance that new engineers reading old books would read
>10mmF as "ten millimilliFarads" or ten microfarads if they've never
>seen it before, and it would be good for a book to explain "Here's
>what it really means."
> For another point, I really don't think it's neccesary to have a
>CPS-to-Hz conversion table (especially since someone thoughtfully put
>one on the web), but it would be good to describe cycles-per-second,
>cps and c/s, and tell when the "official" change to Hertz (abbreviated
>Hz) was done (I forget exactly, I know it was '70's or maybe '80s).
>Have a whole section on "International differences in and historical
>changes in electronic nomenclature."

I still miss the Mho...

Guy Macon

unread,
Feb 16, 2004, 10:18:20 PM2/16/04
to
Tim Auton <tim.auton@uton.[groupSexWithoutTheY]> says...
>
>John Larkin <jjla...@highlandSNIPtechTHISnologyPLEASE.com> wrote:

>>And why, of all the branches of science and engineering, do we want to
>>be the only goofballs who use a weird units notation?
>
>Perhaps because nobody else has to draw and work from schematics, with
>all the problems that brings? If mathematicians sometimes had to work
>from a single faded copy of a proof that had been reduced to 1/2 size
>and had a tea stain on it they too would want more robust notation.

I don't see how the "goofball weird units" notation is any better
when reduced than "the way all of the other the branches of science
and engineering do it" is. Every demonstration of the supposed
inferiority of the number+prefix+unit (NPU) system depends on the
use of a proportional font to hide the decimal points. In a
monospace font, the difference between 47uF and 4.7uF is obvious
even if the latter looks like 4 7uF (4 7uF). Mechanical engineers
have to work from faded copies reduced to 1/2 size with tea stains,
and they haven't felt a need to re-invent their notation.

John Larkin

unread,
Feb 16, 2004, 10:21:20 PM2/16/04
to

People are doing optical design, mechanical design, civil engineering,
architecture, and spacecraft design successfully using SI units in
standard form; no scientist would dare publish any other way. Only
some schematic scribblers seem to be incapable of managing this. It
looks stupid.

It's ironic that, now that everything is CAD and schematics are easily
transmitted as PDFs or whatever, people are citing the inaccuracies of
blueprint machines (which many of them have never seen!) as a reason
we can't use decimal points.

All our schematics are CAD files. When we want a print, we shoot one
out on a 1200 dpi B-size laser printer. All the parts selections are
stored in a separate BOM file, as company part numbers... *not* as
mere descriptions of value. We don't fax schematics, we e-mail them.
And we don't regularly spill tea on our documentation.

John

qrk

unread,
Feb 16, 2004, 11:27:12 PM2/16/04
to
On 15 Feb 2004 14:14:10 -0800, wh...@picovolt.com (Winfield Hill)
wrote:

AoE is used for physics instrumentation classes. The non-electronics
type in these courses may have a problem with the 4k7 notation.
Guaranteed that you will have a gaggle of confused physics grad
students. If you use it, be sure to have a usage topic before the
first chapter. I personally don't care as I use both systems. Book
printing is good enough where the decimal is visible. It's only the
scanned copy floating on the warez groups where you might get dropped
decimals.

Mark

Harry Dellamano

unread,
Feb 16, 2004, 11:33:05 PM2/16/04
to
Upon further review 4-way junction are bad and should be avoided.
Thanks all
Harry

Guy Macon

unread,
Feb 16, 2004, 11:57:40 PM2/16/04
to

Harry Dellamano <har...@tdsystems.org> says...

>Upon further review 4-way junction are bad and should be avoided.

Durn Right! 6-way junctions are the way to go!

And not those wussy 2 dimensional 6-way junctions, either.
USE your Z-axis! Make every angle 90 degrees!

Guy Macon

unread,
Feb 17, 2004, 12:26:18 AM2/17/04
to

John Larkin <jjla...@highlandSNIPtechTHISnologyPLEASE.com> says...

>
>On Tue, 17 Feb 2004 02:08:34 +0000, Tim Auton
><tim.auton@uton.[groupSexWithoutTheY]> wrote:
>
>>John Larkin <jjla...@highlandSNIPtechTHISnologyPLEASE.com> wrote:
>>[snip]
>>>And why, of all the branches of science and engineering, do we want to
>>>be the only goofballs who use a weird units notation?
>>
>>Perhaps because nobody else has to draw and work from schematics, with
>>all the problems that brings? If mathematicians sometimes had to work
>>from a single faded copy of a proof that had been reduced to 1/2 size
>>and had a tea stain on it they too would want more robust notation.
>
>People are doing optical design, mechanical design, civil engineering,
>architecture, and spacecraft design successfully using SI units in
>standard form; no scientist would dare publish any other way. Only
>some schematic scribblers seem to be incapable of managing this. It
>looks stupid.
>
>It's ironic that, now that everything is CAD and schematics are easily
>transmitted as PDFs or whatever, people are citing the inaccuracies of
>blueprint machines (which many of them have never seen!) as a reason
>we can't use decimal points.
>
>All our schematics are CAD files. When we want a print, we shoot one
>out on a 1200 dpi B-size laser printer. All the parts selections are
>stored in a separate BOM file, as company part numbers... *not* as
>mere descriptions of value. We don't fax schematics, we e-mail them.
>And we don't regularly spill tea on our documentation.

Right! The same people who spill tea on schematics gave us those
british "every logic gate is a box" schematics. REAL engineers
spill Jolt Cola and Red Bull on schematics...

John Woodgate

unread,
Feb 17, 2004, 1:26:13 AM2/17/04
to
I read in sci.electronics.design that Guy Macon <http@?.guymacon.com>
wrote (in <94SdnckryLm...@speakeasy.net>) about 'H&H AoE --
advice re: component value notation', on Mon, 16 Feb 2004:

>I still miss the Mho...

Yes, we can't now claim that 2 ohm resistors introduce a time delay of
half a mho.
--
Regards, John Woodgate, OOO - Own Opinions Only.
The good news is that nothing is compulsory.
The bad news is that everything is prohibited.
http://www.jmwa.demon.co.uk Also see http://www.isce.org.uk

Rick

unread,
Feb 17, 2004, 3:14:50 AM2/17/04
to
John Larkin <jjla...@highsniplandthistechpleasenology.com> wrote:
>
> While England is a cute little island

England is no more an island than is the USA - it's glued to
neighboUring countries called Scotland and Wales.


--
Rick

Frank Bemelman

unread,
Feb 17, 2004, 3:30:55 AM2/17/04
to
"Tom Del Rosso" <tdne...@att.net.invalid> schreef in bericht
news:3xfYb.9573$aH3.3...@bgtnsc04-news.ops.worldnet.att.net...

> In news:10769776...@news-01.evisp.enertel.nl,
> Jan Panteltje typed:
> >
> > Thomas
> > Why don't we just draw the colored bands in the diagrams ;-)?
> > These days with inkjet color printers no problem.
>
> Not a problem, but it's even more cryptic. If there was room on a
> resistor (and there is less so all the time :)) then you'd just print
> the number on it.
>
> I could get used to having the unit multiplier where the decimal point
> belongs, but why is it necessary if they don't do that with printed
> numbers in other fields of engineering?

Because the unit symbol is bigger, *and* also tells about the
value range, R, K or M, for resistors and p, u or n.

4,7K is 4 characters, 4K7 is 3. One character saved.

We can also adapt the numerical notation used on SMD resistors,
for instance 3300 for a 1% 330 ohm resistor ;-)

Personally, I don't really care, I can read all variations just
as easily. It's not that it upsets me.

--
Thanks, Frank.
(remove 'x' and 'invalid' when replying by email)


Frank Bemelman

unread,
Feb 17, 2004, 3:33:52 AM2/17/04
to
"John Woodgate" <j...@jmwa.demon.contraspam.yuk> schreef in bericht
news:eZYCwhAF...@jmwa.demon.co.uk...

> I read in sci.electronics.design that Guy Macon <http@?.guymacon.com>
> wrote (in <94SdnckryLm...@speakeasy.net>) about 'H&H AoE --
> advice re: component value notation', on Mon, 16 Feb 2004:
>
> >I still miss the Mho...
>
> Yes, we can't now claim that 2 ohm resistors introduce a time delay of
> half a mho.

I once selected a resistor that introduced a time delay of
three weeks. ;-)

It is loading more messages.
0 new messages