Re: gEDA-user: Metric, Imperial, Rounding, DRC, and board houses

2011-02-24 Thread jpka
> Use metric grid/unit

+1
Pcb's grid is much more precise (it's float point variable) than even 
internal pcb units (which is small enough but integer), and errors 
introduced by grid itself (if any) is not cumulative across board items 
(pins, etc). So simply set grid to 1 mm should give excellent results.



___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user


Re: gEDA-user: Metric, Imperial, Rounding, DRC, and board houses

2011-02-22 Thread Peter Clifton
On Tue, 2011-02-22 at 02:37 +0100, Kai-Martin Knaak wrote:
> Russell Dill wrote:
> 
> > In addition, my board house rounds everything to
> > 2.4 format (0.1 mil). I can envision several scenarios where my design
> > meets DRC in PCB, but fails when I send it to the board house.
> > 
> > What is my best option?
> 
> Time to change the board house? 

Steven is American... and I'd imagine board houses in the US are likely
to be the last to abandon imperial units.

-- 
Peter Clifton

Electrical Engineering Division,
Engineering Department,
University of Cambridge,
9, JJ Thomson Avenue,
Cambridge
CB3 0FA

Tel: +44 (0)7729 980173 - (No signal in the lab!)
Tel: +44 (0)1223 748328 - (Shared lab phone, ask for me)


signature.asc
Description: This is a digitally signed message part


___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user


Re: gEDA-user: Metric, Imperial, Rounding, DRC, and board houses

2011-02-21 Thread Kai-Martin Knaak
Russell Dill wrote:

> In addition, my board house rounds everything to
> 2.4 format (0.1 mil). I can envision several scenarios where my design
> meets DRC in PCB, but fails when I send it to the board house.
> 
> What is my best option?

Time to change the board house? 


> Just use imperial units and cope with weird grids?
> 
> Use metric spacing, but recalculate DRC based on worst case rounding?

I'd choose the second option. PCB internal units are 0.01 mil. This adds 
maximum 10% to the rounding error, the board house introduces anyway. If 
you do grids that fit the board house metric the rounding error will still 
happen.  The 1mm pitch will make some balls off by 0.05 mill.  
The internal units are 254 nm. I sincerely doubt that a design better to 
be etched in 35 µm copper can be that critical. 

---<)kaimartin(>---
-- 
Kai-Martin Knaak
Email: k...@familieknaak.de
Öffentlicher PGP-Schlüssel:
http://pool.sks-keyservers.net:11371/pks/lookup?search=0x6C0B9F53



___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user


Re: gEDA-user: Metric, Imperial, Rounding, DRC, and board houses

2011-02-21 Thread Steven Michalske
I use mil grids, as many boards houses expect mil units, like yours.

PCB layout is engineering, engineering involves tolerances...

I start my traces from off grid components so that the stub going from
the pin/pad is off grid byt quickly becomes on grid after the first 45
degree bend.

Steve


On Mon, Feb 21, 2011 at 2:44 PM, Stefan Salewski  wrote:
> On Mon, 2011-02-21 at 15:23 -0700, Russell Dill wrote:
>> I'm starting a new design and all my components are metric based,
>> including a few 1mm pitch BGA components. I'd really like to do the
>> layout in metric, but I'm worried about two factors. The first of
>> which is that PCB does not yet have the option to store things
>> internally in metric (at least from what I understand) so rounding may
>> occur on that end. In addition, my board house rounds everything to
>> 2.4 format (0.1 mil). I can envision several scenarios where my design
>> meets DRC in PCB, but fails when I send it to the board house.
>>
>> What is my best option?
>>
>> Just use imperial units and cope with weird grids?
>>
>> Use metric spacing, but recalculate DRC based on worst case rounding?
>>
>> Some other option?
>>
>
> Use metric grid/unit, I guess most of us use that.
> Internal resolution is 0.01 mil, which is good. Of course nm would be
> better. One problem for me was, that a few 0.01 mil garbage lines were
> generated for the layout. Not a big problem. Some not really smart
> people have tried to use a very very  fine grid, 0.01 mm or so. That is
> like using no grid at all, I call that silly. Try to use a useful basic
> grid like 0.25mm grid if most of your parts have 0.5mm pitch -- I think
> that was what I did, I am not sure. Important: Enable snap to pads/pins,
> so you can make good connections to imperial parts which are not on the
> metric grid.
>
>>my board house rounds everything to
>> 2.4 format (0.1 mil).
>
> Strange -- so pitch of metric parts may vary from pad to pad a bit. That
> was a problem in very old days of PCB, when 1mil internal units was
> used.
>
>
>
>
> ___
> geda-user mailing list
> geda-user@moria.seul.org
> http://www.seul.org/cgi-bin/mailman/listinfo/geda-user
>


___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user


Re: gEDA-user: Metric, Imperial, Rounding, DRC, and board houses

2011-02-21 Thread Stefan Salewski
On Mon, 2011-02-21 at 15:23 -0700, Russell Dill wrote:
> I'm starting a new design and all my components are metric based,
> including a few 1mm pitch BGA components. I'd really like to do the
> layout in metric, but I'm worried about two factors. The first of
> which is that PCB does not yet have the option to store things
> internally in metric (at least from what I understand) so rounding may
> occur on that end. In addition, my board house rounds everything to
> 2.4 format (0.1 mil). I can envision several scenarios where my design
> meets DRC in PCB, but fails when I send it to the board house.
> 
> What is my best option?
> 
> Just use imperial units and cope with weird grids?
> 
> Use metric spacing, but recalculate DRC based on worst case rounding?
> 
> Some other option?
> 

Use metric grid/unit, I guess most of us use that.
Internal resolution is 0.01 mil, which is good. Of course nm would be
better. One problem for me was, that a few 0.01 mil garbage lines were
generated for the layout. Not a big problem. Some not really smart
people have tried to use a very very  fine grid, 0.01 mm or so. That is
like using no grid at all, I call that silly. Try to use a useful basic
grid like 0.25mm grid if most of your parts have 0.5mm pitch -- I think
that was what I did, I am not sure. Important: Enable snap to pads/pins,
so you can make good connections to imperial parts which are not on the
metric grid.
  
>my board house rounds everything to
> 2.4 format (0.1 mil).

Strange -- so pitch of metric parts may vary from pad to pad a bit. That
was a problem in very old days of PCB, when 1mil internal units was
used.




___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user


gEDA-user: Metric, Imperial, Rounding, DRC, and board houses

2011-02-21 Thread Russell Dill
I'm starting a new design and all my components are metric based,
including a few 1mm pitch BGA components. I'd really like to do the
layout in metric, but I'm worried about two factors. The first of
which is that PCB does not yet have the option to store things
internally in metric (at least from what I understand) so rounding may
occur on that end. In addition, my board house rounds everything to
2.4 format (0.1 mil). I can envision several scenarios where my design
meets DRC in PCB, but fails when I send it to the board house.

What is my best option?

Just use imperial units and cope with weird grids?

Use metric spacing, but recalculate DRC based on worst case rounding?

Some other option?


___
geda-user mailing list
geda-user@moria.seul.org
http://www.seul.org/cgi-bin/mailman/listinfo/geda-user