Unable to melt solder in holes?

[zerorisers]

Hello, I have been having a small issue with a repair. I cannot seem to melt the solder inside of the holes. I have tried flux to see if that would help with the thermal contact but there has to be an air gap between the two sides, I do not have a solder sucker. Currently everything is bodged and works just fine. I was just wondering what may cause the phenomenon stated.

[McBryce]

It won't come out by itself, you'll have to use gauss or a solder sucker. Alternatively, you could (over) fill the hole with fresh solder. Then you can melt the entire solder and poke the pin through (or poke a toothpick through if you want the hole left empty).

McBryce.

[zerorisers]

It won't come out by itself, you'll have to use gauss or a solder sucker. Alternatively, you could (over) fill the hole with fresh solder. Then you can melt the entire solder and poke the pin through (or poke a toothpick through if you want the hole left empty).

McBryce.

I have tried adding more solder, I will heat one side of the hold it well melt perfectfully find but the other side of it is still hard. It is a dual layer board. http://schematicsforum.com/schematics/Audio/JBL/PSWD110_DPS-10%20sm.pdf schematic, and board layout near the bottom of this pdf.

Tried heating from either side, even trying to bring the part through the hole with the one side melted and still nothing.

[Richard Crowley]

Does your board not have plated-thru holes?  If you can melt one side and the other side is still hard, it sounds like you do NOT have plated-thru holes. Otherwise, maybe your iron is just too wimpy?

[flynwill]

Either that or it's actually a 4-layer board and the hole in question is bonded to one of the inner ground or power layers without any heat-relief.

[mzzj]

Quite common problem with some boards. 4- or more layers can be a total pain in the a** but even ordinary 2-layer is sometimes tricky.

GOOD desoldering station like Hakko 472 assisted with normal iron is really good with these. Use desoldering iron from one side and heat the same via from another side of the board with ordinary soldering iron.

4++ layer power supply pcb with extra thick coppers and huge groundplanes (from cellular base station hardware) was just total nightmare. Of course it had to be also ROHS-tin... I had to  pre-heat it close to 150C and use combination of Hakko 471 and my Metcal to get anything off from that bugger!

[Shock]

Photo would be nice.

If you are soldering a lead in then you either need more heat, more preheat, more flux, better solder or clean the area and component lead before you start.

If you are removing solder that is just covering a hole get a solder sucker or wick. For cases where the solder is stubborn and you don't have a desoldering station and all else fails you can tap or flick the circuit board as soon as you receive melt. Make sure you put something under to catch the solder or protect the rest of the PCB from splashes. Obviously it's not ideal but better than nothing.

[mzzj]

So something to try:
1. fill vias with fresh (non-rohs) solder
2. "pre-heat" another side of the board with the soldering iron (if there is lots of copper you can heat like 1 minute per via)
3. use desoldering station and soldering iron simultaneously on both sides of the board. Desoldering wick and 2 irons is probably too tricky unless you happen to have 3 arms 
4. pre-heat the board until its crispy... something like 100-120C degrees should be good. Oven, hot air gun, BBQ grill or  whatever you have.


Not so good ideas (been there...)
-drill the holes open (and easily drill away the hole plating.)
-melt the solder with soldering iron and use compressed air to blow the molten solder away. Solder splashes fly everywhere including between ic pins on your board and your eyes -

[zerorisers]

So something to try:
1. fill vias with fresh (non-rohs) solder
2. "pre-heat" another side of the board with the soldering iron (if there is lots of copper you can heat like 1 minute per via)
3. use desoldering station and soldering iron simultaneously on both sides of the board. Desoldering wick and 2 irons is probably too tricky unless you happen to have 3 arms 
4. pre-heat the board until its crispy... something like 100-120C degrees should be good. Oven, hot air gun, BBQ grill or  whatever you have.


Not so good ideas (been there...)
-drill the holes open (and easily drill away the hole plating.)
-melt the solder with soldering iron and use compressed air to blow the molten solder away. Solder splashes fly everywhere including between ic pins on your board and your eyes -

I guess i can bake the board like someone would a tray of cookies. problem is my oven iv very uneven in terms of thermal. the back gets way hotter than the front by 20 F

Photo would be nice.

If you are soldering a lead in then you either need more heat, more preheat, more flux, better solder or clean the area and component lead before you start.

If you are removing solder that is just covering a hole get a solder sucker or wick. For cases where the solder is stubborn and you don't have a desoldering station and all else fails you can tap or flick the circuit board as soon as you receive melt. Make sure you put something under to catch the solder or protect the rest of the PCB from splashes. Obviously it's not ideal but better than nothing.

Another thing to try when I get home. Hopefully it will work. there are 3 components being stubborn. C1, C7A, and C7B. All are currently bodged. and working as of the moment.

Quite common problem with some boards. 4- or more layers can be a total pain in the a** but even ordinary 2-layer is sometimes tricky.

GOOD desoldering station like Hakko 472 assisted with normal iron is really good with these. Use desoldering iron from one side and heat the same via from another side of the board with ordinary soldering iron.

4++ layer power supply pcb with extra thick coppers and huge groundplanes (from cellular base station hardware) was just total nightmare. Of course it had to be also ROHS-tin... I had to  pre-heat it close to 150C and use combination of Hakko 471 and my Metcal to get anything off from that bugger!

I do not currently have the money for a desoldering iron. been trying to save up for a graphics card for my computer so that I can do 3D modeling. I am probably going to try and use a lead from an old LED hence it is thinner and cover it all in flux, adding more flux as I go through. If that doesn't work I will have to wait til I can get a desoldering station. I can try and preheat using my sisters hair dryer! 


Thank you all for the helpful posts!

Either that or it's actually a 4-layer board and the hole in question is bonded to one of the inner ground or power layers without any heat-relief.

I have provided a shecmatic that shows PCP layout in Original post

Does your board not have plated-thru holes?  If you can melt one side and the other side is still hard, it sounds like you do NOT have plated-thru holes. Otherwise, maybe your iron is just too wimpy?

The iron is decent and has worked on items with much more thermal capacity. originally tried on 250 C then 300 C. not willing to go higher than that unless there is a decently sized heatsink on the board, or I cannot melt the solder in 5 seconds max, ( usually instant melt for me)

[woodchips]

Another possibility is that the hole uses leaded solder and your iron uses unleaded. The resultant low lead solder doesn't melt at all like electonics solder, think plumbing and wiping soldered joints. As you would expect, swapping the two solders results in the same unpleasant alloy.

[grumpydoc]

you'll have to use gauss

I'm not sure a magnet will help......... 

ITYM solder wick and were confusing spellings for gauze (open woven fabric) and gauss (unit of magnetic field).

I sympathize - I'm still trying to learn French beyond a few stuttered words and have been doing so for 30 years without very much progress.

[suicidaleggroll]

The iron is decent and has worked on items with much more thermal capacity. originally tried on 250 C then 300 C. not willing to go higher than that unless there is a decently sized heatsink on the board, or I cannot melt the solder in 5 seconds max, ( usually instant melt for me)

You can go MUCH hotter than that.  That's definitely where the problem is.  Use a nice big tip on your iron and set the temp at closer to 400 C.

I don't know why some people are so hesitant to turn up the heat on their irons.  I always run mine at ~750 F (400 C), and I'll turn it up even hotter when doing big through holes on ground planes.  I only turn it down for items that melt easily (connectors with big cheap plastic bits) or that need a lot of solder and it'll take a while to feed it all in anyway, but I never go below ~600 F (315 C).

[miguelvp]

Two videos that might help:

[McBryce]

you'll have to use gauss

I'm not sure a magnet will help......... 

Yes, I blaim my colleague for that. I called it braid or wick for years and then started working with an Australian  guy who called it gauze (which I had no idea how to spell). Since then I have been saying (and misspelling) it too.

McBryce.

[m100]

You can go MUCH hotter than that.  That's definitely where the problem is.  Use a nice big tip on your iron and set the temp at closer to 400 C.

I don't know why some people are so hesitant to turn up the heat on their irons.  I always run mine at ~750 F (400 C), and I'll turn it up even hotter when doing big through holes on ground planes.  I only turn it down for items that melt easily (connectors with big cheap plastic bits) or that need a lot of solder and it'll take a while to feed it all in anyway, but I never go below ~600 F (315 C).

I agree,  look here for the tip temperatures that Metcal irons run at (no adjustment other than changing tip)  the 500(obsolete)/600/700 series are  302,  357, or 412 deg C respectively with the default for use on FR4 tips being the 412 deg C ones

http://wsbenelux.nl/wp-content/uploads/2014/11/metcal-tip-temperatures.pdf

Or here for the Weller Magnastat tips, 260, 310, 370, 425 deg C  (most common I encountered were iirc 370 deg C but it was probably a couple of decades since I last used them)

http://www.farnell.com/datasheets/44151.pdf

For problem holes use lots of new 60 tin 40 lead solder (or preferably Sn62 which has 62% tin, 36% lead and 2% silver) and a flux core, high temperatures, at least 350 deg C and preferably closer to 400 deg C, a well tinned iron with a suitably sized and shaped bit (I prefer conical for through hole and desoldering, some prefer chisel)  Solder sucker available for less than five quid on ebay or desolder braid, or a syringe needle.  Get enough solder in the interface between the iron and the joint to effectively heat the joint.  If you can get a melt to fill the hole with the new solder then you should be able to get a melt to remove all the excess solder from the hole.  Putting an iron on a joint at 300 deg C is a waste of time.  If the board is genuinely FR4 and manufactured correctly then you won't damage the laminate or the bond between the copper and the laminate using 400 deg C as a tip temperature even if it takes 20 or 30 seconds to fully desolder the joint

Very occasionally you'll get a boards with holes you will never fully desolder as you'd like, but they are typically the ones with huge flood filled power / ground planes with copper three times the usual thickness or with a damaged via. 

Most of my desoldering is with a 30+ year old 45W Weller EC2002D (infinitely adjustable 250 -450 deg C)  or a 40W Metcal with a 700 series tip (412 deg C) with a sucker or desolder braid (servisol or chemtronics).  I've got a Weller desoldering station with a built in vacuum pump that iirc is 60W but it is very rarely used these days.

[G7PSK]

If there is no wire left in the hole I have used very fine drill bits and turned the with my fingers to clear stubborn holes, the solder always flows into the holes easier than it comes out. Never use a drill bit bigger than the hole size.