Offpage connector symbols

Hi,

As a side-comment to the schematic preferences thread
(hopefully not another lengthy thread  :> ), I'm curious
as to what folks use as an offpage connector symbol.

Given that I prefer these to be *at* the edges of the schematic
*and* given that the signal name will be "outboard" of it,
I try to use very narrow symbols.  And, since it is common for
other such "offpage" symbols to be located immediately above
and/or below, I try to keep the height of this symbol to
"one intersignal spacing unit" (IsSU?  :> ).

I also like to show direction of signal flow in the symbol.

This has led me to a set of six (<frown>) symbols:
Output Right
Input Right (blech!)
BiDir Right
and the corollaries for "Left". (I.e., left and right refer
to the edges of the page at which it is most appropriate to
place these symbols).

For unidirectional signal flow, I use a pair of "concentric"
(wrong word) arrow heads.  E.g., >> or <<.  These can be spaced
close enough (horizontally) together that they occupy very little
space on the page (i.e., 1 IsSU square).

For BiDir signals, I use one of each arrow head (< + >).
Since BiDir symbols should occupy the same amount of space
(an arbitrary but desirable condition I impose), I overlap these
together.

If they don't overlap much (or, at all), you end up with a
diamond (<>) or an X (><).  I compromise and end up with
an assymetrical "stacked pair of X's" -- sort of like a
slice out of a DNA helix.

This is intentionally assymetric -- you could shift one or
the other arrow head to obtain better symmetry throughout the
X
X
but then placing two or more of these BiDir symbols above each
other ends up looking like *needlepoint* (can't see where one
signal begins and the previous one ends!)

Other techniques?

On 1/19/2010 4:17 PM, D Yuniskis wrote:
> Hi,
>
> As a side-comment to the schematic preferences thread
> (hopefully not another lengthy thread :> ), I'm curious
> as to what folks use as an offpage connector symbol.
>
> Given that I prefer these to be *at* the edges of the schematic
> *and* given that the signal name will be "outboard" of it,
> I try to use very narrow symbols. And, since it is common for
> other such "offpage" symbols to be located immediately above
> and/or below, I try to keep the height of this symbol to
> "one intersignal spacing unit" (IsSU? :> ).
>
> I also like to show direction of signal flow in the symbol.
>
> This has led me to a set of six (<frown>) symbols:
> Output Right
> Input Right (blech!)
> BiDir Right
> and the corollaries for "Left". (I.e., left and right refer
> to the edges of the page at which it is most appropriate to
> place these symbols).
>
> For unidirectional signal flow, I use a pair of "concentric"
> (wrong word) arrow heads. E.g., >> or <<. These can be spaced
> close enough (horizontally) together that they occupy very little
> space on the page (i.e., 1 IsSU square).
>
> For BiDir signals, I use one of each arrow head (< + >).
> Since BiDir symbols should occupy the same amount of space
> (an arbitrary but desirable condition I impose), I overlap these
> together.
>
> If they don't overlap much (or, at all), you end up with a
> diamond (<>) or an X (><). I compromise and end up with
> an assymetrical "stacked pair of X's" -- sort of like a
> slice out of a DNA helix.
>
> This is intentionally assymetric -- you could shift one or
> the other arrow head to obtain better symmetry throughout the
> X
> X
> but then placing two or more of these BiDir symbols above each
> other ends up looking like *needlepoint* (can't see where one
> signal begins and the previous one ends!)
>
> Other techniques?

I use bigger things with the signal name inside--arrow directed outwards 
for outputs, inwards for inputs, like this:


  /------------
<  CONV_CLOCK |-----    and
  \------------

    *-------\
---|  DONE  >
    *-------/

Cheers

Phil Hobbs
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs at electrooptical dot net
http://electrooptical.net

Hi Phil,

Phil Hobbs wrote:
> On 1/19/2010 4:17 PM, D Yuniskis wrote:
>>
>> As a side-comment to the schematic preferences thread
>> (hopefully not another lengthy thread :> ), I'm curious
>> as to what folks use as an offpage connector symbol.

>> Other techniques?
> 
> I use bigger things with the signal name inside--arrow directed outwards 
> for outputs, inwards for inputs, like this:
> 
>  /------------
> <  CONV_CLOCK |-----    and
>  \------------
> 
>    *-------\
> ---|  DONE  >
>    *-------/

Yeah, but that implicitly limits the length of a signal name.
I opted for small symbols (so they don't consume much lateral space)
which lets me dynamically rearrange the space adjacent to the
symbol for a lengthier name, etc.  :-/

On 1/19/2010 5:13 PM, D Yuniskis wrote:
> Hi Phil,
>
> Phil Hobbs wrote:
>> On 1/19/2010 4:17 PM, D Yuniskis wrote:
>>>
>>> As a side-comment to the schematic preferences thread
>>> (hopefully not another lengthy thread :> ), I'm curious
>>> as to what folks use as an offpage connector symbol.
>
>>> Other techniques?
>>
>> I use bigger things with the signal name inside--arrow directed
>> outwards for outputs, inwards for inputs, like this:
>>
>> /------------
>> < CONV_CLOCK |----- and
>> \------------
>>
>> *-------\
>> ---| DONE >
>> *-------/
>
> Yeah, but that implicitly limits the length of a signal name.
> I opted for small symbols (so they don't consume much lateral space)
> which lets me dynamically rearrange the space adjacent to the
> symbol for a lengthier name, etc. :-/

Nah, I do it in pencil--the name can be as long as my ruler. ;)

Cheers

Phil Hobbs

-- 
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058

email: hobbs at electrooptical dot net
http://electrooptical.net

On Tue, 19 Jan 2010 14:17:24 -0700, D Yuniskis wrote:

> Hi,
> 
> As a side-comment to the schematic preferences thread (hopefully not
> another lengthy thread  :> ), I'm curious as to what folks use as an
> offpage connector symbol.
> 
> Given that I prefer these to be *at* the edges of the schematic *and*
> given that the signal name will be "outboard" of it, I try to use very
> narrow symbols.  And, since it is common for other such "offpage"
> symbols to be located immediately above and/or below, I try to keep the
> height of this symbol to "one intersignal spacing unit" (IsSU?  :> ).
> 
> I also like to show direction of signal flow in the symbol.
> 
> This has led me to a set of six (<frown>) symbols: Output Right
> Input Right (blech!)
> BiDir Right
> and the corollaries for "Left". (I.e., left and right refer to the edges
> of the page at which it is most appropriate to place these symbols).
> 
> For unidirectional signal flow, I use a pair of "concentric" (wrong
> word) arrow heads.  E.g., >> or <<.  These can be spaced close enough
> (horizontally) together that they occupy very little space on the page
> (i.e., 1 IsSU square).
> 
> For BiDir signals, I use one of each arrow head (< + >). Since BiDir
> symbols should occupy the same amount of space (an arbitrary but
> desirable condition I impose), I overlap these together.
> 
> If they don't overlap much (or, at all), you end up with a diamond (<>)
> or an X (><).  I compromise and end up with an assymetrical "stacked
> pair of X's" -- sort of like a slice out of a DNA helix.
> 
> This is intentionally assymetric -- you could shift one or the other
> arrow head to obtain better symmetry throughout the X
> X
> but then placing two or more of these BiDir symbols above each other
> ends up looking like *needlepoint* (can't see where one signal begins
> and the previous one ends!)
> 
> Other techniques?

Most of the schematic editors I've used distinguish an innie from an 
outie, at least graphically (if not as part of a hierarchical schematic 
editor that even does some rules checking, like no nets with all inputs, 
no nets with two alway-on outputs, etc).

-- 
www.wescottdesign.com

Hi Tim,

Tim Wescott wrote:
> On Tue, 19 Jan 2010 14:17:24 -0700, D Yuniskis wrote:
> 
>> As a side-comment to the schematic preferences thread (hopefully not
>> another lengthy thread  :> ), I'm curious as to what folks use as an
>> offpage connector symbol.
>>
>> Other techniques?
> 
> Most of the schematic editors I've used distinguish an innie from an 
> outie, at least graphically (if not as part of a hierarchical schematic 
> editor that even does some rules checking, like no nets with all inputs, 
> no nets with two alway-on outputs, etc).

That's why it's important to make sure the pin attribute on
such connectors is correct.

On Jan 19, 4:43=A0pm, Phil Hobbs
<pcdhSpamMeSensel...@electrooptical.net> wrote:
> On 1/19/2010 4:17 PM, D Yuniskis wrote:
>
>
>
>
>
> > Hi,
>
> > As a side-comment to the schematic preferences thread
> > (hopefully not another lengthy thread :> ), I'm curious
> > as to what folks use as an offpage connector symbol.
>
> > Given that I prefer these to be *at* the edges of the schematic
> > *and* given that the signal name will be "outboard" of it,
> > I try to use very narrow symbols. And, since it is common for
> > other such "offpage" symbols to be located immediately above
> > and/or below, I try to keep the height of this symbol to
> > "one intersignal spacing unit" (IsSU? :> ).
>
> > I also like to show direction of signal flow in the symbol.
>
> > This has led me to a set of six (<frown>) symbols:
> > Output Right
> > Input Right (blech!)
> > BiDir Right
> > and the corollaries for "Left". (I.e., left and right refer
> > to the edges of the page at which it is most appropriate to
> > place these symbols).
>
> > For unidirectional signal flow, I use a pair of "concentric"
> > (wrong word) arrow heads. E.g., >> or <<. These can be spaced
> > close enough (horizontally) together that they occupy very little
> > space on the page (i.e., 1 IsSU square).
>
> > For BiDir signals, I use one of each arrow head (< + >).
> > Since BiDir symbols should occupy the same amount of space
> > (an arbitrary but desirable condition I impose), I overlap these
> > together.
>
> > If they don't overlap much (or, at all), you end up with a
> > diamond (<>) or an X (><). I compromise and end up with
> > an assymetrical "stacked pair of X's" -- sort of like a
> > slice out of a DNA helix.
>
> > This is intentionally assymetric -- you could shift one or
> > the other arrow head to obtain better symmetry throughout the
> > X
> > X
> > but then placing two or more of these BiDir symbols above each
> > other ends up looking like *needlepoint* (can't see where one
> > signal begins and the previous one ends!)
>
> > Other techniques?
>
> I use bigger things with the signal name inside--arrow directed outwards
> for outputs, inwards for inputs, like this:
>
> =A0 /------------
> < =A0CONV_CLOCK |----- =A0 =A0and
> =A0 \------------
>
> =A0 =A0 *-------\
> ---| =A0DONE =A0>
> =A0 =A0 *-------/
>
> Cheers
>
> Phil Hobbs
> Dr Philip C D Hobbs
> Principal
> ElectroOptical Innovations
> 55 Orchard Rd
> Briarcliff Manor NY 10510
> 845-480-2058
>
> email: hobbs at electrooptical dot nethttp://electrooptical.net- Hide quo=
ted text -
>
> - Show quoted text -

Yes!!
Working in production is one thing, fixing it in the field is
something else!

For me, I definitely prefer signal names inside the offpage connector
(arrow).
NOT adjacent to the connector, or squished between the lines. Very
cumbersome outside the factory.

Also, try to avoid all nomenclatures that attempt to describe the
function (such as "in" or "out", etc..) when the interpretation
depends on some outside intervention (such as the prespective of the
tech).

For example, I wouldn't recommend using "Audio Out Right" because that
same term won't make any sense when it gets to where it's going.
Try "Unamplified Audio, R+", or something like that.

You get the idea.

On Jan 19, 7:14=A0pm, mpm <mpmill...@aol.com> wrote:

> Also, try to avoid all nomenclatures that attempt to describe the
> function (such as "in" or "out", etc..) when the interpretation
> depends on some outside intervention (such as the prespective of the

Jesus H. Christ on a stick YES. Every, EVERY time we have to connect
two boards that have "RXD, TXD, GND", there is confusion over whether
that means "data from me to you" or "data from you to me". ARGHHH!! We
have literally boards that go through six "flips" of RX/TX lines
because of this. Each time something in the middle got revved, someone
got it wrong, and someone else added a buffer and a flip in a
different subsystem.

On Tue, 19 Jan 2010 14:17:24 -0700, D Yuniskis
<not.going.to.be@seen.com> wrote:

>Hi,
>
>As a side-comment to the schematic preferences thread
>(hopefully not another lengthy thread  :> ), I'm curious
>as to what folks use as an offpage connector symbol.
>
>Given that I prefer these to be *at* the edges of the schematic
>*and* given that the signal name will be "outboard" of it,
>I try to use very narrow symbols.  And, since it is common for
>other such "offpage" symbols to be located immediately above
>and/or below, I try to keep the height of this symbol to
>"one intersignal spacing unit" (IsSU?  :> ).
>
>I also like to show direction of signal flow in the symbol.
>
>This has led me to a set of six (<frown>) symbols:
>Output Right
>Input Right (blech!)
>BiDir Right
>and the corollaries for "Left". (I.e., left and right refer
>to the edges of the page at which it is most appropriate to
>place these symbols).

Pretty much.  Output right (input left) makes sense at times, though I
try not to use them either.

>For unidirectional signal flow, I use a pair of "concentric"
>(wrong word) arrow heads.  E.g., >> or <<.  These can be spaced
>close enough (horizontally) together that they occupy very little
>space on the page (i.e., 1 IsSU square).
>
>For BiDir signals, I use one of each arrow head (< + >).
>Since BiDir symbols should occupy the same amount of space
>(an arbitrary but desirable condition I impose), I overlap these
>together.

Yeah, I'd like to fit the sheet number cross references in there, but
that's asking for too much.

>If they don't overlap much (or, at all), you end up with a
>diamond (<>) or an X (><).  I compromise and end up with
>an assymetrical "stacked pair of X's" -- sort of like a
>slice out of a DNA helix.

Diamond works for me.

>This is intentionally assymetric -- you could shift one or
>the other arrow head to obtain better symmetry throughout the
>X
>X
>but then placing two or more of these BiDir symbols above each
>other ends up looking like *needlepoint* (can't see where one
>signal begins and the previous one ends!)
>
>Other techniques?

larwe wrote:
> 
> On Jan 19, 7:14 pm, mpm <mpmill...@aol.com> wrote:
> 
> > Also, try to avoid all nomenclatures that attempt to describe the
> > function (such as "in" or "out", etc..) when the interpretation
> > depends on some outside intervention (such as the prespective of the
> 
> Jesus H. Christ on a stick YES. Every, EVERY time we have to connect
> two boards that have "RXD, TXD, GND", there is confusion over whether
> that means "data from me to you" or "data from you to me". ARGHHH!! We
> have literally boards that go through six "flips" of RX/TX lines
> because of this. Each time something in the middle got revved, someone
> got it wrong, and someone else added a buffer and a flip in a
> different subsystem.


   Wouldn't it be simpler to have pads for zero ohm resistors to
configure the port?  Or two positions to put the connector to chose the
right configuration?


-- 
Greed is the root of all eBay.