开云体育

ctrl + shift + ? for shortcuts
© 2025 开云体育

PCB through hole activation for copper plating


 

The problem was not lack of solder inside the eyelet, the component leads soldered fine with solder filling up the eyelet. It also soldered well to the bottom layer.
Once the eyelet channel filled up with solder there was no motivation for it to flow further across and down over the edge to finally join the top layer of the PCB.
The slit provided a path to reach the top layer without having to flow over the top.
Bertho

-----Original Message-----
From: [email protected] <[email protected]> On Behalf Of Donald H Locker
Sent: Saturday, June 13, 2020 20:42
To: [email protected]
Subject: Re: [homebrewpcbs] PCB through hole activation for copper plating

More likely provided the capillary wick necessary for the solder to follow. With the plain barrel eyelets, the solder would have to wick up the center hole or between the eyelet and the substrate - one gap is too large and the other is too small (as well as having a non-wettable surface facing the eyelet.) The slit provides a wettable, narrow channel for the solder.

Isn't surface tension wonderful!

Donald.
--
*Plain Text* email -- it's an accessibility issue
() no proprietary attachments; no html mail /\ <>

On 2020-06-13 4:40 p.m., Lee Studley wrote:
@ Bertho: That's an very interesting note. Thanks for sharing that. I
wonder if the slit also broke some sort of surface tension at play to
help the solder get where it needed to go.


 

Hi all,
I tried it, unsuccessfully.?

In my experiments, it works, but not reliably enough to be really useful. Even getting 95% or even 99% of the vias activated is awful, because you cannot trust what you built.

Maybe it is possible to get better results, but I doubt it would be practical.

You should try it though, these were fun experiments!


Le mar. 11 ao?t 2020 à 11:07, MVS Sarma <mvssarma@...> a écrit?:
I sawc ideos using graphite powder or its based ink as substitute for pth conductive formation later copper electroplating.?

Hope it helps.? May be toonthin holes may not be coverable.?


 

I sawc ideos using graphite powder or its based ink as substitute for pth conductive formation later copper electroplating.?

Hope it helps.? May be toonthin holes may not be coverable.?


 

开云体育

Ah, now there, I have some better tools perhaps than average.

Stereo binocular microscope, Metcal soldering and desoldering equipment with a reasonable amount of tips, and enough experience (hard gained) to manage to do a decent job.

I normally deal with parts 0608 and larger, TQFP with 0.5 mm pin spacing, and double sided boards (cause they're cheaper).? There's a modicum of through hole parts as needed, but mostly all SMD.? The through hole devices are mostly pin headers at this point.

I'd likely use *their* assembly services *if* I had to use a BGA.? Managed to avoid that so far.? Don't have the desire or ability to do them properly.

For parts, I manage to use what I find at hamfests, which is likely obsolete, but still good. (as in: works).? Not sure that they'd have them, but if I have a few hundred, I'll see what I can design with them.

Harvey


On 8/1/2020 6:29 PM, gnuarm wrote:

On Sat, Jun 13, 2020 at 07:51 PM, Harvey White wrote:

all of this VERY much depends on the level of complexity of your board.? Me?? I've moved to Chinese made boards with 6 mil traces, TQFP-144 chips and plated through holes.? I don't know how to make them myself.

Harvey

I also find assembly to be a pain.? There is a house that makes the PCB and will assemble your board.? The only catch is you have to use parts in their list of 15,000 or so.? That may be too restrictive for professional use, but for many home projects it's ok.? It's also hard to search the parts for electrical parameters like you can at Digikey.? Still, if you have the time to search the part categories one at a time it is an inexpensive way to get a board built.


 

On Sat, Jun 13, 2020 at 07:51 PM, Harvey White wrote:

all of this VERY much depends on the level of complexity of your board.? Me?? I've moved to Chinese made boards with 6 mil traces, TQFP-144 chips and plated through holes.? I don't know how to make them myself.

Harvey

I also find assembly to be a pain.? There is a house that makes the PCB and will assemble your board.? The only catch is you have to use parts in their list of 15,000 or so.? That may be too restrictive for professional use, but for many home projects it's ok.? It's also hard to search the parts for electrical parameters like you can at Digikey.? Still, if you have the time to search the part categories one at a time it is an inexpensive way to get a board built.


 

More likely provided the capillary wick necessary for the solder to
follow. With the plain barrel eyelets, the solder would have to wick up
the center hole or between the eyelet and the substrate - one gap is too
large and the other is too small (as well as having a non-wettable
surface facing the eyelet.) The slit provides a wettable, narrow channel
for the solder.

Isn't surface tension wonderful!

Donald.
--
*Plain Text* email -- it's an accessibility issue
() no proprietary attachments; no html mail
/\ <>

On 2020-06-13 4:40 p.m., Lee Studley wrote:
@ Bertho: That's an very interesting note. Thanks for sharing that. I
wonder if the slit also broke some sort of surface tension at play to
help the solder get where it needed to go.


 

开云体育

I'll agree.? Other than what's mentioned, I've done two layer boards at home.?

Problems are:

1) you can't do vias under chips unless the chips are DIL.

2) the vias have to be fairly large for you to be able to solder to them.

3) lacing wire through 200 holes is a pain.

4) No way to do 4 layer boards, or even 3 layer boards.? Yes, you can, but you're going to have to drill a pit in the top layer and run a wire jumper to a pad.

5) most automatic routers assume you're going to a commercial PCB manufacturer, and route the board accordingly.? You're likely to need to manually route the boards.? (Vias underneath chips (fixable), close pad spacing, etc.? You can do it, likely enough, but will the design autoroute?)

We won't talk about the problem in getting the top and bottom layers registered, however you decide to do it.

My immediate response is that past a certain level of complexity, making your own boards needs stuff you don't have, and is a significant investment (chemical baths to plate through holes, upside down drills to accurately drill holes (did that one....), and the like.? Even the tin-lead plating involves chemicals that seem to be nasty.

all of this VERY much depends on the level of complexity of your board.? Me?? I've moved to Chinese made boards with 6 mil traces, TQFP-144 chips and plated through holes.? I don't know how to make them myself.

Harvey


On 6/13/2020 4:48 PM, Lee Studley wrote:

>From a pro,
If its only 2 layers and you insist on real through hole, it's most time and cost effective to just have them made at an online boardhouse( dirt cheap).
Otherwise the threading a long length of fine wire and soldering if wave soldered, then snipping, or if smt, as a post-task is the way to go.

If more than 2 layers, get them made.

?


 

From a pro,
If its only 2 layers and you insist on real through hole, it's most time and cost effective to just have them made at an online boardhouse( dirt cheap).
Otherwise the threading a long length of fine wire and soldering if wave soldered, then snipping, or if smt, as a post-task is the way to go.

If more than 2 layers, get them made.

?


 

@ Bertho: That's an very interesting note. Thanks for sharing that. I wonder if the slit also broke some sort of surface tension at play to help the solder get where it needed to go.


 

开云体育

Just a comment about eyelets, not at all suggesting to use them:

Many, many decades ago I worked at Motorola when eyelets were used.

The original style, and common today, is a tube with the top edge rolled over.? There was no problem wave-soldering the bottom or the component wire if present.? The problem was that the solder inside the eyelet would not consistently flow over the top edge? of the eyelet onto the top PCB pad.? In effect only an unreliable mechanical contact was formed on the top side.

?

The problem was resolved by using eyelets with a slit in them that allowed the solder to flow out onto the top pad.

Actually there was no rolled over edge, the eyelets were straight slit tube sections that were expanded on top and bottom as part of the eyelet insertion process.

Bertho

?

?

From: [email protected] <[email protected]> On Behalf Of Terry Gray
Sent: Friday, June 12, 2020 19:46
To: [email protected]
Subject: Re: [homebrewpcbs] PCB through hole activation for copper plating

?

And again, not looking for a mechanical solution.? Already tried eyelets...they suck for so many reasons, the foremost being inconsistency of connection. Already tried soldered wires...this also sucks...mainly being a PITA to solder and they only address vias.? Any through hole connectors can't be done this way.

Also have tried conductive inks, paints, resins.? Never got more than about 75% of the holes being conductive.


Jim Higgins
 

Received from Terry Gray at 6/13/2020 00:17 UTC:

Jim, Sodium Hypophosphite Monohydrate is apparently a controlled substance but for whatever reason it seems to be much easier and far cheaper to obtain than its cousin Calcium Hypophosphite, and I haven't talked to anyone that knows why! I can obtain it from several places in the US as well as from several Chinese vendors on Alibaba for a much lower price than CH.

Perhaps I'm wrong, but I bet you found places that list a price for it, and seem to indicate it's available, but the real hassle will come when you actually try to order it. Established labs/companies with a good reason can get it in limited quantities, but as a home hobbiest (assuming here) it's another story. I truly wish you luck.

Jim H


 

Here's another method for making conductive thru holes:

?<??>

A Google search turns up many other methods, some of them may actually be usable ;-)

Roy


 

I think that are difficult to understand all the content of the site, but I want put focus in this process that doesn't use calcium hypophosphite:





On Friday, June 12, 2020, 5:53:15 PM GMT-3, Terry Gray <twgray2007@...> wrote:


Thanks everyone for your replies and sugestions.

Thanks, Matheus!? While not exactly what I was hoping for, this is a reasonable solution.? I will give a shot at trying to "make" calcium hypophosphite.? Been a looong time since I dabbled in chemistry but maybe it is like riding a bicycle and you never forget how!

And, for anyone who cares, calcium hypophosphite is available from several vendors on alibaba.? They all advertise the stuff for pennies per gram but when you request a quote all of them come back in the neighborhood of $100 per 500G.? Then when you tell them you are in the US, the price jumps up to about $500 per 500G.? This is well beyond the range of most hobbyists...at least well beyond mine.?

Now I am off to price diy calcium hypophosphite!

Thanks again!


 

Jim, Sodium Hypophosphite Monohydrate is apparently a controlled substance but for whatever reason it seems to be much easier and far cheaper to obtain than its cousin Calcium Hypophosphite, and I haven't talked to anyone that knows why!? I can obtain it from several places in the US as well as from several Chinese vendors on Alibaba for a much lower price than CH.


 

开云体育

Eyelets + solder paste + reflow oven?



Sent from my T-Mobile 5G Device


-------- Original message --------
From: Terry Gray <twgray2007@...>
Date: 6/12/20 6:45 PM (GMT-06:00)
Subject: Re: [homebrewpcbs] PCB through hole activation for copper plating

And again, not looking for a mechanical solution.? Already tried eyelets...they suck for so many reasons, the foremost being inconsistency of connection. Already tried soldered wires...this also sucks...mainly being a PITA to solder and they only address vias.? Any through hole connectors can't be done this way.

Also have tried conductive inks, paints, resins.? Never got more than about 75% of the holes being conductive.


 

And again, not looking for a mechanical solution.? Already tried eyelets...they suck for so many reasons, the foremost being inconsistency of connection. Already tried soldered wires...this also sucks...mainly being a PITA to solder and they only address vias.? Any through hole connectors can't be done this way.

Also have tried conductive inks, paints, resins.? Never got more than about 75% of the holes being conductive.


 

I did have an idea, which may or may not work:

It's not a replacement for plated through holes, but may address the same problem.

I was working on a strictly mechanical basis, which meant eyelets or jumper wires

Eyelets have a history of being badly soldered and not being as permanent as people would want.

What I thought of was an automated system that would push a piece of soft copper wire into a hole.

From the bottom it would be center punched to swage it in the hole (and it wouldn't be flush)

Then at the top, cut off a bit above flush, and swaged to cover the top pad as much as possible.

What I'm thinking of is effectively a solid rivet.

While I was thinking of this reply, I thought that a wire protuded through the board, then bent at right angles.

Then at the top, cut off the wire above the surface (distance matters!), then bend the wire over to make a C shaped jumper between top and bottom layers.

The idea is that this could be automated with a special setup that had the appropriate wire benders as needed.

Solder that, and you've got a better solution, I think.

Harvey

On 6/12/2020 6:53 PM, Jim Higgins wrote:
Received from Matheus Costa via groups.io at 6/12/2020 15:16 UTC:

There are another method that don't use calcium hypophosphite. It's in this site (in portuguese, but google translate can be effective to get the essential):



Anyway, in that site there are a homebrew solution to manufacture calcium hypophosphite too.

Best regards
Matheus

The method at this site starts with Sodium Hypophosphite, which is as much a controlled substance as Calcium Hypophosphite.

Jim H




Jim Higgins
 

Received from Matheus Costa via groups.io at 6/12/2020 15:16 UTC:

There are another method that don't use calcium hypophosphite. It's in this site (in portuguese, but google translate can be effective to get the essential):



Anyway, in that site there are a homebrew solution to manufacture calcium hypophosphite too.

Best regards
Matheus

The method at this site starts with Sodium Hypophosphite, which is as much a controlled substance as Calcium Hypophosphite.

Jim H


 

Thanks everyone for your replies and sugestions.

Thanks, Matheus!? While not exactly what I was hoping for, this is a reasonable solution.? I will give a shot at trying to "make" calcium hypophosphite.? Been a looong time since I dabbled in chemistry but maybe it is like riding a bicycle and you never forget how!

And, for anyone who cares, calcium hypophosphite is available from several vendors on alibaba.? They all advertise the stuff for pennies per gram but when you request a quote all of them come back in the neighborhood of $100 per 500G.? Then when you tell them you are in the US, the price jumps up to about $500 per 500G.? This is well beyond the range of most hobbyists...at least well beyond mine.?

Now I am off to price diy calcium hypophosphite!

Thanks again!


 

There are another method that don't use calcium hypophosphite. It's in this site (in portuguese, but google translate can be effective to get the essential):


Anyway, in that site there are a homebrew solution to manufacture calcium hypophosphite too.

Best regards
Matheus

On Thursday, June 11, 2020, 9:19:23 PM GMT-3, Terry Gray <twgray2007@...> wrote:


There is an Instructable on this very subject.

I have had no luck with the conductive ink or paint method, never getting close to 95%, which is why I am inquiring about the chemical method.

If I can't obtain Calcium Hypophosphite then I would like to find an alternative chemical method that doesn't require it.