Copper pour or traces?

[Matt]

Hello,

I am laying out a little breakout board that spreads synchro data
(60/400Hz) throughout a system. This is the first board I have ever
layed out that I have contimplated using a 2-layer board instead of a
4-layer board. Because of this, I will not have independent GND/PWR
planes. Looking around at other 2-layer PCBs, I see that some folks
like to use a copper pour on the TOP and BOT layers. I am assuming one
layer is traces+GND and the other is traces+PWR. My questions are
these:

1. Is this a sound design (no high frequency and the signals are very
strong).
2. From reading, I was convinced to try to use hatching instead of a
solid copper pour to prevent warping.
3. If I do use cross hatching what should my grid be if my global
spacing for all traces, pads, layers, etc. is 8 mils.

Thank you.

 

[John Larkin]

Hello,

I am laying out a little breakout board that spreads synchro data
(60/400Hz) throughout a system. This is the first board I have ever
layed out that I have contimplated using a 2-layer board instead of a
4-layer board. Because of this, I will not have independent GND/PWR
planes. Looking around at other 2-layer PCBs, I see that some folks
like to use a copper pour on the TOP and BOT layers. I am assuming one
layer is traces+GND and the other is traces+PWR. My questions are
these:

1. Is this a sound design (no high frequency and the signals are very
strong).

At 400 Hz, practically anything will work.

2. From reading, I was convinced to try to use hatching instead of a
solid copper pour to prevent warping.

Copper pours don't seem to warp boards nowadays, if they ever did.

John

 

[Ecnerwal]

Copper pours don't seem to warp boards nowadays, if they ever did.

However, providing a little relief around the pads can make them easier
to solder and de-solder (ie, put a hole around the pad, and then tie the
pad to the pour with 3 or 4 traces, rather than having no pad.) I'll
often just run a pattern of round holes in the pour rather than a grid
(resulting in square holes), which in thoery is better for etching
behavior (less sharp corners) but I'm not sure this is really needed or
makes any practical difference. However, it keeps me happy, and does not
seem to detract any.

 

[John B]

Hello,

I am laying out a little breakout board that spreads synchro data
(60/400Hz) throughout a system. This is the first board I have ever
layed out that I have contimplated using a 2-layer board instead of a
4-layer board. Because of this, I will not have independent GND/PWR
planes. Looking around at other 2-layer PCBs, I see that some folks
like to use a copper pour on the TOP and BOT layers. I am assuming one
layer is traces+GND and the other is traces+PWR. My questions are
these:

Rule 1: Don't assume anything.

It is unlikely that you will ever find a board with an external power
plane. Double sided boards will normally have a copper pour on each
side both of which are connected to GND. Having said that, go back and
invoke rule 1 above.

1. Is this a sound design (no high frequency and the signals are very
strong).
2. From reading, I was convinced to try to use hatching instead of a
solid copper pour to prevent warping.
3. If I do use cross hatching what should my grid be if my global
spacing for all traces, pads, layers, etc. is 8 mils.

Thank you.

Just use solid copper pours. FR4 will not warp if you have copper pours
on both sides.

 

[Joerg]

At 400 Hz, practically anything will work.




Copper pours don't seem to warp boards nowadays, if they ever did.

They used to if you didn't balance the thermal mass between both sides
from the core layer. In fact, this was the main reason for one of the
blitz consulting trips. 35 hours was travel time, the remainder was
finding that warpage was the cause of end test failures, donning the old
army boots to walk over to a lumber yard (roads too iced up for
driving), making a frame, running a few boards.

 

[Joerg]

Rule 1: Don't assume anything.

It is unlikely that you will ever find a board with an external power
plane. Double sided boards will normally have a copper pour on each
side both of which are connected to GND. Having said that, go back and
invoke rule 1 above.




Just use solid copper pours. FR4 will not warp if you have copper pours
on both sides.

Provided they are somewhat summetrical to the core. Else go back to rule
1

 

[John Larkin]

They used to if you didn't balance the thermal mass between both sides
from the core layer. In fact, this was the main reason for one of the
blitz consulting trips. 35 hours was travel time, the remainder was
finding that warpage was the cause of end test failures, donning the old
army boots to walk over to a lumber yard (roads too iced up for
driving), making a frame, running a few boards.

We did one board with microwave laminate as the first dielectric, and
FR-4 for the rest. It looked like a potato chip. It was cool: you
could lay it on your desk and give it a twirl, and it would spin for
almost a full minute. It took vise grips to get it out of the card
cage.

John