In industry holes are plated with copper ("plated through holes").? This is not something you can easily do yourself at home, instead you can just solder both sides where possible, or use a via (small hole moving a trace from one side to the other) with a bit of wire or micro rivet.

On Wed, Aug 1, 2018 at 1:02 PM, Vagulus <rick@...> wrote:

My project involved several proprietary breakouts attached to a main PCB.? They are physically mounted on the top of the board and the pins are? soldered on the B.Cu layer (as I believe is normal).? That's where I get lost!?

One of the proprietary breakouts is the Arduino Pro Mini - the brains of the operation - which has a footprint like this

Assume I have a two-layer PCB:? most of the pins of the Pro Mini are connected to traces on B.Cu but some must connect to traces on F.Cu.? As illustrated, this footprint is just a collection of standard pads.? How do they connect through the fibreglass of the board?

I guess this is a silly question to those of you in the know, but it has me stumped.? I cannot find anything on the Net or in any literature which explains this mystery.? I hope one or more of you can help.

Thanks

I would use the wire. I have seen rivets come loose and break the circuit. A single strand is usually enough or 22 -24 gauge tinned wire

Craig, I spent over 45 years of my life maintaining electronic equipment, many items in the tens and even hundreds of thousands of dollars range and most by major, well known companies like RCA and Sony. I have seen circuit boards that depended on staked on connections between a top and bottom foil layer. In virtually ALL of these circuit boards these connection devices were soldered on both the top and bottom foil layers. These connections were either eyelets, staked on pins that component leads could be attached to or simple "Z" wires (a piece of wire through the hole and bent down on both sides.. Many times I had to trace problems to a bad connection through these connections. Among the problems I have seen were the use of brass pins that sat around too long and developed a layer of oxide that did not "take" the solder, pads that just did not get soldered in the first place, and pins where the solder took but subsequently broke. I can not remember a single case where a plated through hole was bad. I can not remember a single case where a "Z" wire where the solder "took" that failed. All the failures that I had to repair were either on eyelets or on staked on pins.

In the vast majority of the problems that I can remember, the problem was the use of brass pins or eyelets that were NOT PLATED and which had developed a layer of oxide before they were inserted in the board. I guess the use of non plated components was an economy measure. In many of these cases it was plainly visible that the solder did not "take". There was an insulating layer between the eyelet or pin and the solder or the copper trace on the board. But in a significant number of them, the wave solder machine had successfully covered the staked on bottom of the eyelet or pin with an unbroken layer of solder and all looked OK. But when I dragged a soldering iron with a bit of melted solder on it over that staked on end, the solder was drawn completely off the eyelet of pin, exposing the oxidized surface. I had to first clean all that oxide off before re-soldering it or the same, bad connection would repeat.

I do not know the brand name of these eyelets or pins. But the common element that I saw was that they were NOT PLATED. It was not that they "came loose". They were still firmly attached to the board and ON TOP of the copper trace. But the oxide layer prevented conduction.

The really bad thing about this is that these bad joints were not discovered when the boards were made nor in the first few years of use. Many of them only became evident after 5 or 10 or more years of use. The were a time bomb, waiting to go off. Thus testing after manufacture would not have caught the problem.

My one piece of advise if you are going to use some kind of inserted connection between the top and bottom foil on a PCB is to USE PLATED wire or eyelets or pins. And by "plated" I mean SOLDER plated. do not use bare copper wire. Do not use bare brass pins. Pay the extra penny and get plated ones.

Thank you very much for taking the time to give an informative and useful reply. It has given me some things to consider.

?

The process I am considering using involves cutting copper wire to a few thousands longer than the thickness of the PCB board and pressing it in place from both sides resulting in a connection that is flush or nearly flush with the surface of the board. The advantages of this are low cost, use of easily available material and connections that could possibly be situated under components if absolutely necessary. Unfortunately, therefore, the cut and pressed ends of the wire which are exposed to be soldered would not be plated. Also, it seems the best workflow would be to place them after etching and drilling the board but before adding a solder mask rather than immediately before soldering. I guess the best option is to make sure they are well cleaned immediately before soldering. Another option might be to use silver solder wire or sterling silver wire which would not oxidize so quickly. This would also be an advantage for thermal vias given the greater heat transfer of silver. This is not so outlandish as it would seem at first glance - the cost with 20 gauge wire is less than two cents per "via" and for 24 gage wire less than a penny.

?

Craig

On Wed, Aug 1, 2018 at 08:20 PM, Paul Alciatore wrote:

Craig, I spent over 45 years of my life maintaining electronic equipment, many items in the tens and even hundreds of thousands of dollars range and most by major, well known companies like RCA and Sony. I have seen circuit boards that depended on staked on connections between a top and bottom foil layer. In virtually ALL of these circuit boards these connection devices were soldered on both the top and bottom foil layers. These connections were either eyelets, staked on pins that component leads could be attached to or simple "Z" wires (a piece of wire through the hole and bent down on both sides.. Many times I had to trace problems to a bad connection through these connections. Among the problems I have seen were the use of brass pins that sat around too long and developed a layer of oxide that did not "take" the solder, pads that just did not get soldered in the first place, and pins where the solder took but subsequently broke. I can not remember a single case where a plated through hole was bad. I can not remember a single case where a "Z" wire where the solder "took" that failed. All the failures that I had to repair were either on eyelets or on staked on pins.

In the vast majority of the problems that I can remember, the problem was the use of brass pins or eyelets that were NOT PLATED and which had developed a layer of oxide before they were inserted in the board. I guess the use of non plated components was an economy measure. In many of these cases it was plainly visible that the solder did not "take". There was an insulating layer between the eyelet or pin and the solder or the copper trace on the board. But in a significant number of them, the wave solder machine had successfully covered the staked on bottom of the eyelet or pin with an unbroken layer of solder and all looked OK. But when I dragged a soldering iron with a bit of melted solder on it over that staked on end, the solder was drawn completely off the eyelet of pin, exposing the oxidized surface. I had to first clean all that oxide off before re-soldering it or the same, bad connection would repeat.

I do not know the brand name of these eyelets or pins. But the common element that I saw was that they were NOT PLATED. It was not that they "came loose". They were still firmly attached to the board and ON TOP of the copper trace. But the oxide layer prevented conduction.

The really bad thing about this is that these bad joints were not discovered when the boards were made nor in the first few years of use. Many of them only became evident after 5 or 10 or more years of use. The were a time bomb, waiting to go off. Thus testing after manufacture would not have caught the problem.

My one piece of advise if you are going to use some kind of inserted connection between the top and bottom foil on a PCB is to USE PLATED wire or eyelets or pins. And by "plated" I mean SOLDER plated. do not use bare copper wire. Do not use bare brass pins. Pay the extra penny and get plated ones.

From my experience, I wish you good luck. You will need it if you do
this on any volume of vias.

--
E. Paul Alciatore III
Beaumont, TX
PAlciatore@...