Is increasing PCB copper' thickness through galvanization a good idea?

[malisz]

Is increasing PCB copper' thickness through galvanization a good idea?

I am about to prepare PCBs for Aleph 2 and I am tempted to increase the thickness of copper layer to let the currents flow ... The option is to go from standard 105 micro meters up to 170 micro. Is it a good idea ? My concern is that galvanized copper will have probably a different crystal structure from the original copper layer what potentially may result in some weird electrical phenomenons that can impair the sound.

In sum, am I looking at potential improvement/de-provement of sound or maybe I should not bother at all ?

malisz

 

[Kilowatt]

I wouldn't worry about weird electrical phenomena, a conductor in the signal path will not make something sound different. But, there has been discussion before on increasing PCB trace thickness, but with solder, and it was concluded that it doesn't really help, since the copper is more conductive, most of the current will just flow throught it. The same applies to zinc, I'm sure.

 

[SY]

As a practical matter, that's a tough thing to do on an already-etched board. Making contact to all the traces and getting current densities evened out (especially from edge to center of each trace) will be more of a challenge than just starting with thicker copper in the first place.

 

[cocolino]

Hi malisz,

You might take a look at this thread:

Strengthening PCB tracks with solder?

 

[Circlotron]

I just lay 1.5mm diameter solid copper wire along the traces that matter, and solder them along the length. The resistance of solder is woeful compared to the same cross section of copper IIRC. Don't know how bad zinc is but wouldn't be as good as copper.

 

[malisz]

To clarify, my question was about adding extra layer of COPPER and not tin or zinc. PCB shops can do it in galwanization process. First they produce the PCB and then place it in a galwanization "bath" to thicken the traces.

Nota bene shops can use also raw boards with thicker copper foil but these are actually hard to find and therefore most common approach is to use galwanization process.

From your responces I assume that the move is right and I'll go for 170 micro.

THX
malisz

 

[grataku]

As a practical matter, that's a tough thing to do on an already-etched board. Making contact to all the traces and getting current densities evened out (especially from edge to center of each trace) will be more of a challenge than just starting with thicker copper in the first place.

SY
I believe you brought this up once and I was rather intrigued with the notion they would plate before etching. Do you know that for a fact?
I always thought that they had a way to plate after etching. Intuitively it should give a much cleaner trace and produce less Cu poisoned iron chloride solution. I don't see why it would be too difficult to find a way around connecting the interrupted traces.

 

[SY]

Not that they would plate before etching, but rather that one starts with thicker copper laminate. Using up the FeCl3 is not usually an issue, it's pretty cheap stuff. I haven't done much professional work with etching rigid boards, but we did a lot of this with copper-laminated flex circuits.

 

[AuroraB]

I production of through-plated boards, one usually starts with either a very thin layer, like 10-15 um, or blank fiberglass. Then all the holes are drilled, and copper is then catalythically, or immersion plated, to the vias and other component holes. Then the rest of the board is electroplated to the desired thickness, and finally etched. Doing the plating after etching is not value for money, as it requires a LOT of manual labour to connect all the traces....

 

[haldor]

SY
I believe you brought this up once and I was rather intrigued with the notion they would plate before etching. Do you know that for a fact?
I always thought that they had a way to plate after etching. Intuitively it should give a much cleaner trace and produce less Cu poisoned iron chloride solution. I don't see why it would be too difficult to find a way around connecting the interrupted traces.

I'm sure you can plate after etching, but the process for creating double sided PCB's with plated through holes that I am familiar with is as follows

- Start with a double sided board.
- Drill all hole that are to be plated through (using a CNC drill).
- Apply the negative photo resist to both sides of the PCB. Traces and pads are exposed, copper you want to remove is covered by resist.
- Apply a conductive liquid ink to all holes.
- Blow out excess ink from the holes with compressed air (a thin layer of conductive ink remains to line the holes. Let ink dry.
- Plate the PCB (plating all tracks and pads, plus plating the through holes in the process).
- Strip the resist from the PCB.
- Etch the unplated copper from the PCB. The etching solution works on copper, but not tin. The tin plating acts as the resist while etching the copper.
- Apply solder mask.
- Stencil on text.

This process can also be used to make multi-layer PCB's except the inside layers are pre-etched using a positive photo resist before laminating them with the outer layers which are not etched. Once the sandwhich of layers is formed you use the same process as for double sided PCB's.

Blind and half blind vias are a different issue and I have no idea how they make PCB's with those.

Phil

 

[carlmart]

I production of through-plated boards, one usually starts with either a very thin layer, like 10-15 um, or blank fiberglass. Then all the holes are drilled, and copper is then catalythically, or immersion plated, to the vias and other component holes. Then the rest of the board is electroplated to the desired thickness, and finally etched. Doing the plating after etching is not value for money, as it requires a LOT of manual labour to connect all the traces....

This process sounds very interesting but may not be as common to find. Likely expensive too.

Something which I do when I design my pcbs is widening every copper track, particularly those that may carry heavier current, until my pcb looks like a virgin copper board with 1mm "river like" separations between large copper "islands".

Apparently that is something that is bad for industrial boards but we can use for DIY. But I saw it twice: in a Yamaha amp and on a Rotel amp.

Using solder to increase the thickness has been commented as having some problems sonically, but I never used so I can't confirm on that.


Carlos

 

[haldor]

Something which I do when I design my pcbs is widening every copper track, particularly those that may carry heavier current, until my pcb looks like a virgin copper board with 1mm "river like" separations between large copper
Carlos

Hi Carlos,

The reason manufacture's avoid fat traces has to do with solderability, specifically flow soldering since all the extra copper conducts heat away from the solder joint and increases the chance of getting a bad solder joint.

The way this is handled with ground planes is by including a thermal relief feature at the through hole pad. You just make an island out of the pad and connect it to the heavy trace (or ground plane) using a couple of thinner traces. For DIY this is probably not important since you can just turn the heat up on your iron (or use a Metcal iron that seems to never run lack for heating capability).

Phil

 

[carlmart]

The reason manufacture's avoid fat traces has to do with solderability, specifically flow soldering since all the extra copper conducts heat away from the solder joint and increases the chance of getting a bad solder joint.

The way this is handled with ground planes is by including a thermal relief feature at the through hole pad. You just make an island out of the pad and connect it to the heavy trace (or ground plane) using a couple of thinner traces. For DIY this is probably not important since you can just turn the heat up on your iron (or use a Metcal iron that seems to never run lack for heating capability).

Hi, Phil!


What I had been told about why manufacturers avoid large traces was that boards tend to twist physically in continuous soldering baths.

I never had any trouble (bad solder joint) after I started designing my pcbs like that. On top of that I never had any pcb track degluing from the fiber either.


Carlos