Author Topic: Filling vias with solder during reflow  (Read 4710 times)

0 Members and 1 Guest are viewing this topic.

Online HwAoRrDkTopic starter

  • Super Contributor
  • ***
  • Posts: 1476
  • Country: gb
Filling vias with solder during reflow
« on: December 10, 2017, 03:37:31 am »
I made a thread in another forum section a while back asking about quantity and size of vias I would need for a trace to survive a 20A surge, and I came away with a conclusion of 9 x 0.6 mm diameter vias. It was suggested to me that filling the vias with solder would enhance the current-carrying capacity. I wondered whether that would be possible to do during reflow soldering by leaving these vias unmasked and adding apertures for these vias to the solder paste stencil.

I had been looking around for any kind of advice, article, app note, etc. to see if this was possible and wasn't having much luck until I stumbled upon the magic term: "pin-in-paste". That is, the practice of soldering THT components by reflow. So I found and was following this application note from Littlefuse where they recommended a paste volume twice that of the volume of the hole to be filled (I'm ignoring subtracting the component lead volume and adding volume for fillet).

But my resulting numbers seem quite... extreme, and I'm not sure whether I've made a mistake or not. :-// By my calculations, for a 0.6 mm diameter via in a 1.6 mm PCB, I need a paste stencil aperture of 2.76 mm diameter! :o Via volume works out at 0.45 mm^3, making 0.9 mm^3 paste; using a typical stencil thickness of 0.15 mm then gives radius of paste 1.38 mm.

There's no way I could use such a large area of paste on these vias. For a start, there would be overlap because I have multiple in close proximity. I'm also sceptical of the ability of such a relatively huge area of paste to reflow properly down in to the via's hole.

Am I on a hiding to nothing with this? Should I forget it? It's really just a 'nice to have' to get these vias solder-filled, and I thought it'd be convenient to get it done with the solder paste; perhaps not.
 

Offline janekm

  • Supporter
  • ****
  • Posts: 515
  • Country: gb
Re: Filling vias with solder during reflow
« Reply #1 on: December 10, 2017, 07:29:58 am »
Might it be helpful to use a larger number of smaller vias instead?
In pin-in-paste applications most of the hole is filled by the component pin, so the volume of solder need should be relatively smaller.
Solder does have very good ability to flow into holes though, so it probably won't be an issue so long as you can find the physical area.
 

Offline ocset

  • Super Contributor
  • ***
  • Posts: 1516
  • Country: 00
Re: Filling vias with solder during reflow
« Reply #2 on: December 10, 2017, 12:30:31 pm »
for a 20A surge you just need a couple of 0.6mm or so vias...unfilled....20A for 50us say isnt going to heat up a via like that significantly.
Vias filled with solder are a pipe dream.....people say it can be done, but it cant be done reliably, and needs xraying to check each via is properly filled.
 

Online HwAoRrDkTopic starter

  • Super Contributor
  • ***
  • Posts: 1476
  • Country: gb
Re: Filling vias with solder during reflow
« Reply #3 on: December 10, 2017, 08:13:37 pm »
Might it be helpful to use a larger number of smaller vias instead?

Not really. If I went with 0.3 mm diameter vias, that still leaves me with a paste aperture of 1.36 mm diameter per via. I'd still have overlap problems.

Solder does have very good ability to flow into holes though, so it probably won't be an issue so long as you can find the physical area.

That's really the main issue - I simply don't have the physical area. Not for such large amounts of paste anyway.

for a 20A surge you just need a couple of 0.6mm or so vias...unfilled....20A for 50us say isnt going to heat up a via like that significantly.
Vias filled with solder are a pipe dream.....people say it can be done, but it cant be done reliably, and needs xraying to check each via is properly filled.

It'd probably be way longer than 50us, as were such a surge to take place, the means of it stopping would be a fuse blowing (20A rated fuse on power feed to circuit), which I'm assuming could take 10s of ms. I'm pretty sure the number of vias I have is appropriate, and I have the space for them.

I think I'll forget about the matter and just keep the vias unfilled (but still unmasked). Thanks both for contributions.
 

Offline HalFET

  • Frequent Contributor
  • **
  • Posts: 512
  • Country: 00
Re: Filling vias with solder during reflow
« Reply #4 on: December 10, 2017, 09:16:31 pm »
If you are really worried about this, either use more vias or pay the price to get them filled during plating.

A cheapo solution that I did use once was to install extra test points along the high current path, turns out that a brass through hole pin makes for a decent high current conductor. But that only works if you're going to wave solder.
 

Offline DerekG

  • Frequent Contributor
  • **
  • Posts: 882
  • Country: nf
Re: Filling vias with solder during reflow
« Reply #5 on: December 10, 2017, 11:16:19 pm »
It was suggested to me that filling the vias with solder would enhance the current-carrying capacity. I wondered whether that would be possible to do during reflow soldering by leaving these vias unmasked and adding apertures for these vias to the solder paste stencil.

The biggest problem is that an ordinary PCB is 1.6mm thick (so the via will be 1.6mm long) & a "thick" solder paste stencil is 0.2mm. The numbers just don't equate.

On the other hand, using wave soldering from the bottom layer to fill the vias equates very well :)

Alternatively, if you only have a few vias to fill, you could fill these by hand soldering.
I also sat between Elvis & Bigfoot on the UFO.
 

Offline Mattylad

  • Regular Contributor
  • *
  • Posts: 143
  • Country: gb
Re: Filling vias with solder during reflow
« Reply #6 on: December 11, 2017, 06:55:20 am »
Put a pth component leg in the hole.
Matty
CID+
 


Share me

Digg  Facebook  SlashDot  Delicious  Technorati  Twitter  Google  Yahoo
Smf