[kicad-users] PTH, pad one side only

[[e-mail] [kicad-users]]

I'm vaguely wondering if it is feasible to design a board where pin-headers have a pad at the back, a PTH, then no pad at the front.

The reason is to allow a rather tight spacing between the PTH and an SMD i.c. footprint on the top.

A reasonable pad just touches the i.c. pad, wheras a plain PTH leaves adequate room.

I think this would be mechanically still fairly robust...

I guess however I'd need to use a 'via' along with a zone-defined 'pad' on the back of the board?

['John Woodgate' [e-mail] [kicad-users]]

I doubt that you can plate through without at least a tiny pad on both ends of the hole.

With best wishes DESIGN IT IN! OOO – Own Opinions Only
<http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England

Sylvae in aeternum manent.

From: kicad-***@yahoogroups.com [mailto:kicad-***@yahoogroups.com]
Sent: 14 July 2017 16:59
To: kicad-***@yahoogroups.com
Subject: [kicad-users] PTH, pad one side only


I'm vaguely wondering if it is feasible to design a board where pin-headers have a pad at the back, a PTH, then no pad at the front.

The reason is to allow a rather tight spacing between the PTH and an SMD i.c. footprint on the top.

A reasonable pad just touches the i.c. pad, wheras a plain PTH leaves adequate room.

I think this would be mechanically still fairly robust...

I guess however I'd need to use a 'via' along with a zone-defined 'pad' on the back of the board?

[Clemens Koller [e-mail] [kicad-users]]

Hi!

You can do so and good manufacturers will know what to do. You might really tell them you want to do that for a good reason because the hole plating may become less stable if there are no pads available.
This depends on the (plasma- or chemical-) de-smearing treatment of the PTH to clean the hole walls and improve the stability of the plating in the hole itself - without support from the pads on the layerrs. Some manufacturers might want to skip that treatment and ask for the pads on top+bot and also on the inner layers.

YMMV

Regards,

Clemens

Post by ***@yahoo.co.uk [kicad-users]
I'm vaguely wondering if it is feasible to design a board where pin-headers have a pad at the back, a PTH, then no pad at the front.
The reason is to allow a rather tight spacing between the PTH and an SMD i.c. footprint on the top.
A reasonable pad just touches the i.c. pad, wheras a plain PTH leaves adequate room.
I think this would be mechanically still fairly robust...
I guess however I'd need to use a 'via' along with a zone-defined 'pad' on the back of the board?

[Clemens Koller [e-mail] [kicad-users]]

Some more thoughts on this:

If you decide you leave only a tiny little ring (i.e. 0.1mm) as a pad, some manufacturers will complain that the ring might get destroyed during the drilling process if they need to drill after etching (which depends on your layer stackup).

Talk to your manufacturer. :-)

Regards,

Clemens

Post by Clemens Koller ***@embeon.de [kicad-users]
Hi!
You can do so and good manufacturers will know what to do. You might really tell them you want to do that for a good reason because the hole plating may become less stable if there are no pads available.
This depends on the (plasma- or chemical-) de-smearing treatment of the PTH to clean the hole walls and improve the stability of the plating in the hole itself - without support from the pads on the layerrs. Some manufacturers might want to skip that treatment and ask for the pads on top+bot and also on the inner layers.
YMMV
Regards,
Clemens

Post by ***@yahoo.co.uk [kicad-users]
I'm vaguely wondering if it is feasible to design a board where pin-headers have a pad at the back, a PTH, then no pad at the front.
The reason is to allow a rather tight spacing between the PTH and an SMD i.c. footprint on the top.
A reasonable pad just touches the i.c. pad, wheras a plain PTH leaves adequate room.
I think this would be mechanically still fairly robust...
I guess however I'd need to use a 'via' along with a zone-defined 'pad' on the back of the board?

[[e-mail] [kicad-users]]

Thanks for the feedback.

I do not yet understand why opinion seems to be I need a pad larger than just the 'via itself' at the front.

Surely... the via is fine as it is, it does not seem to need any copper land beyond it's own boundaries?

Sure, a via has some inbuilt/intrinsic annular ring, which is intended to allow connection from the hole plating to the board surface - but that annular ring is small. The default on KiCad seems to be Via Dia, Via Drill = (0.6, 0.4).

For instance, sometimes when I define a mechanical hole, it gets plated, and it looks okay, and there is no excess copper surrounding the hole.

['John Woodgate' [e-mail] [kicad-users]]

Using a via is OK. I thought you were thinking of a plain hole with a pad only on one side of the board. That might work, or might not.

With best wishes DESIGN IT IN! OOO – Own Opinions Only
<http://www.jmwa.demon.co.uk/> www.jmwa.demon.co.uk J M Woodgate and Associates Rayleigh England

Sylvae in aeternum manent.

From: kicad-***@yahoogroups.com [mailto:kicad-***@yahoogroups.com]
Sent: 16 July 2017 21:39
To: kicad-***@yahoogroups.com
Subject: [kicad-users] Re: PTH, pad one side only


Thanks for the feedback.

I do not yet understand why opinion seems to be I need a pad larger than just the 'via itself' at the front.

Surely... the via is fine as it is, it does not seem to need any copper land beyond it's own boundaries?

Sure, a via has some inbuilt/intrinsic annular ring, which is intended to allow connection from the hole plating to the board surface - but that annular ring is small. The default on KiCad seems to be Via Dia, Via Drill = (0.6, 0.4).

For instance, sometimes when I define a mechanical hole, it gets plated, and it looks okay, and there is no excess copper surrounding the hole.