Cleaning up solder from a stubborn thruhole

[lpc32]

Drilling is a last ditch effort at best, as you run a real risk of destroying the hole you're trying to clear.

Sounds like it's used regularly by some of the people here. Maybe less risk when using narrow bits?

It's possible to remove a part if the solder reaches it's plastic range and sufficient mechanical force is applied

That's an interesting point. The capacitors weren't terribly difficult to remove. Partially pull one lead, the other, back to the first... and after a few rounds it's out. Even assuming partial melting, if it was possible to pull the cap out won't it be similarly possible to later mechanically push the solder out?

But this bring me to what AG6QR said...

the remaining ones has come clean using a little piece of resistor lead, and heating the hole and lead with an iron while pushing the lead through the hole.

Do you specifically mean that component leads are better heat conductors than needles? I guess the end is wider, at the very least.

FWIW, even the best irons need a preheater if the PCB has enough layers sucking off the heat.

If it works well enough for some of the people here, on similar boards, using similar basic gear, I do wonder what's different.

I don't dismiss preheating entirely, but I would rather go for the minimalistic approach if possible. I don't think I have anything suitable handy, and it looks awkward for larger things, but I'll read up more and see how it works.

Heat the pad and whack the board against the table.

Even in cases when that can work, I wouldn't want to be brutal to the poor boards.

[AG6QR]

the remaining ones has come clean using a little piece of resistor lead, and heating the hole and lead with an iron while pushing the lead through the hole.

Do you specifically mean that component leads are better heat conductors than needles? I guess the end is wider, at the very least.

I haven't used needles, so my experience is limited.  All I know for certain is that component leads have been good enough in my experience.  Also, they're cheaper for me, since I'm always clipping them off and I always have a bunch around that I'm about to throw in the trash anyway.  But I do know that, if you look up the material properties, copper's heat conductivity is much better than stainless steel.  Steel needles would probably be more rigid and better suited to applying force, though.

Also, for getting good heat transfer, it's useful to wet the tip of the soldering iron with plenty of molten solder to provide a good contact area.  I am certain that molten solder conducts heat a LOT better than air.  When you have a dry iron contacting a dry circuit board, with a dry needle or dry component lead, you'll only have a couple of points of solid contact, almost zero area.  You can improve the heat conduction hugely if you fill all of the air gaps up with molten solder.  This factor is much more important than steel needle versus copper component lead. 

[nanofrog]

Sounds like it's used regularly by some of the people here. Maybe less risk when using narrow bits?

Unlike blunt needles and wire, drill bits are designed to cut. If sized properly (same size as the hole, so will barely clear it) any misalignment, and you'll end up removing some or all of the copper from the hole itself. Possibly tearing a pad off as it exits. Not the best odds, so not good practice generally speaking.

Even assuming partial melting, if it was possible to pull the cap out won't it be similarly possible to later mechanically push the solder out?

Possible? Yes.

It depends on the temp and force applied after inserting some tool (tool can conduct heat away). So you could insert the tool, say a piece of wire or component leg, and it solidify rather than clear the hole out (you're boned without more heat if this happens). Even as it enters the plastic range, use too much force and you'll damage the board (lifted or torn traces). It's a balancing act to pull off, as you're flirting with the absolute minimal heat requirement for the given conditions, and there's considerable risk involved.

Simple solution is more heat.

Hair dryer, electric skillet, heat gun, or oven can help you with this, and you don't need to get anywhere near melting point. As mentioned before, 100C will do wonders. Just work quickly so it doesn't cool off too much if you can't apply preheating while soldering or desoldering. 

But this bring me to what AG6QR said...

the remaining ones has come clean using a little piece of resistor lead, and heating the hole and lead with an iron while pushing the lead through the hole.

Do you specifically mean that component leads are better heat conductors than needles? I guess the end is wider, at the very least.

FWIW, even the best irons need a preheater if the PCB has enough layers sucking off the heat.

If it works well enough for some of the people here, on similar boards, using similar basic gear, I do wonder what's different.

I think you're missing a critical point, and that's that the solder is fully molten when using wire, component leads, needles, probe tips as needles, and so on that's pushing a stick through a hole.

The one exception as I read it, are those that have used drills to clear holes they couldn't get the solder to melt. Unmolten solder = insufficient heat. Its really that simple. The causality can be more complicated (tip size, molten solder on the tip, and so on, that are all related to thermal transfer <get the iron's heat into the joint>).

I don't dismiss preheating entirely, but I would rather go for the minimalistic approach if possible. I don't think I have anything suitable handy, and it looks awkward for larger things, but I'll read up more and see how it works.

Assuming you've an appropriate tip size, have enough molten solder on the tip, sufficient flux, and possibly adding a lower melting point solder as previously discussed for best thermal transfer for the particular joint & equipment used, the proper approach is more heat.

This ^ is where I'm assuming you're at.

So adding heat is the minimalistic approach with this particular board with your equipment and techniques. Any other means without adding heat under these conditions, is risking damage to the board.

[neuraxon77]

I've seen suck mentioned but I haven't seen blow (didn't read every post though), so try blowing it our with compressed air or perhaps pressurized electrical cleaner:

Lay the board down on news paper and put on safety glasses. Tape around the area if you're concerned about solder blowout, preferably kapton and not too close to the hole.
Add flux to the hole (if you have it).
Add solder.
Add heat.
Melt the solder through.
While molten, blow the hole out with compressed air or electrical cleaner.
Repeat on the other side if necessary.
Clean any mess with solder wick or the tip of an iron.
If the thru-hole plating is still intact it should come out clean, if not, you'll need to drill or as a very last resort poke it out while being careful not to lift tracks.

[Alex Eisenhut]

Sounds like it's used regularly by some of the people here. Maybe less risk when using narrow bits?

Unlike blunt needles and wire, drill bits are designed to cut. If sized properly (same size as the hole, so will barely clear it) any misalignment, and you'll end up removing some or all of the copper from the hole itself. Possibly tearing a pad off as it exits. Not the best odds, so not good practice generally speaking.

Then it's not properly sized. I don't know what you think people are doing, but no one is suggesting using a drill press. The solder is far softer than the copper and the drill bit will simply bite into the solder.

I don't understand the resistance to a simple and effective solution. I think trying all these different heat-based approaches are the risky ones.

No one is tearing off pads or removing copper with a #61 wire drill twisted by hand. And even if they are so clumsy, you can see the color of the tailings change.

This is in the repair section, not the IPC-approved NASA man-rated PCB assembly section.

I don't get it.

[pickle9000]

Sounds like it's used regularly by some of the people here. Maybe less risk when using narrow bits?

Unlike blunt needles and wire, drill bits are designed to cut. If sized properly (same size as the hole, so will barely clear it) any misalignment, and you'll end up removing some or all of the copper from the hole itself. Possibly tearing a pad off as it exits. Not the best odds, so not good practice generally speaking.

Then it's not properly sized. I don't know what you think people are doing, but no one is suggesting using a drill press. The solder is far softer than the copper and the drill bit will simply bite into the solder.

I don't understand the resistance to a simple and effective solution. I think trying all these different heat-based approaches are the risky ones.

No one is tearing off pads or removing copper with a #61 wire drill twisted by hand. And even if they are so clumsy, you can see the color of the tailings change.

This is in the repair section, not the IPC-approved NASA man-rated PCB assembly section.

I don't get it.

When drilling (by hand) the correct procedure is to have the bit matched to the lead size (of the component to be installed) and NOT the hole size (which is much larger). The plating may be nicked true but not to an extent that it will be removed.

If excessive heat has been used then any method used has the potential to destroy the through hole plating as the support structure may be weakened. 

Choosing a bit that matches the hole size will remove the plating and is not a correct in any circumstance.

If after drilling it is decided that more solder must be removed (hole too small). Great care must be taken if using another slightly larger drill bit. In this case the bit can grab an detach the through hole. I normally apply fresh solder try to suck or wick it out from the ground plane side. If that fails I use a slightly larger, still smaller than the hole but big enough to fit the new component. Drill only a hole filled completely with solder and be gentle, solder is soft.

The main issue with this type of repair is this. While it can be done well, there is a potential that if done several times the through hole will be at least degraded, at most destroyed. Keeping good records is important. With double sided boards even if the through plating is destroyed it can be fixed, multi-layer boards are another story.

[lpc32]

Ultimately, I made progress on the 2nd board (not the one photoed) after replacing the needle with a cut lead. While heating the lead, I push it into the hole and pulled it out. I repeated it a few times, most of which resulted in the hole closing after the lead was removed, but once or twice it did leave some hole, though too narrow. I followed with repeated wicking and reapplying solder to the hole. In the end it was wide enough, though still not as nicely cleaned as the easier holes.

The board seems functional, but it didn't undergo as much work as the other one.

On the 1st board, pushing a lead into the molten hole also wasn't too difficult, but once the lead was removed it never left a hole, not even a narrow one. This board's holes are wider, I'm not sure if it's a factor.

What's the expected result with this lead poking technique? It didn't push enough solder out, and the extra solder I had to add (to heat the lead) tended to counter the some that was removed from the hole. Do you push it in one direction, or push then pull back? What do you use to handle the lead with?

Regarding just leaving the old solder there and sliding in the new component; is the only reason to clean holes to make it easy to insert new components, or are there advantages to actual removal of the old solder?

[pickle9000]

- As a general rule solder should be removed. This is the best practice. You may then use your own mix.
- If not as much as possible should be removed. And the joint should be finished with solder of the same type and flux.

As for the stainless pin, hollow pin / needle. As far as I have seen and done myself. Cut the lead to be removed in such a way that the inside of the needle can be slipped over the lead. Apply fresh solder and heat until you have pushed the stainless out through the other side of the pcb. At this point the lead will be inside the tube. Remove the heat while rotating the tube left to right while the solder solidifies. Once hardened try to pull the lead from the needle. If that fails, snip the lead from the component side and remove the needle. The solder won't stick to the outside needle, the solder will often prevent the lead from being removed from the needle.

In my experience needles are removed once the solder has cooled.

 

[lpc32]

What's the reason for removal being preferred? By "solder of the same type", do you mean just the distinction between leaded and leadfree?

I might try these hollow pins later on; maybe it's a combination of the best properties of needles and component leads. I do wonder if the lead worked for me (partially anyway), unlike the needle, because of the wider end or because of the material. Or both.

[pickle9000]

There are many variations / formulations when it comes to solder. Pre-leadfree the rules where the same. There is no way of knowing either the quality, environmental conditions or how the solder will react over time. Remove as much of the original as reasonable and use the best you have and trust.

The solder issue is important but for consumer gear a quality brand name solder will probably do everything you need.

[kaindub]

I only saw one mention of this method.
I had similar problems of clearing holes until I tried this method.

I get a solder sucker, but you need the build type because you want to blow the solder clear through the hole.
Apply a little solder to the pad so the solder is slightly above the surface.
Then heat the solder and blow through the hole with the solder sucker.
Works most of the time. There may be a little tail left on the reverse side of the board

Robert

[digital]

Use a pin vise and a twist drill the same diameter as the component lead and just be careful and it will work,be gentle.