Desoldering headache - copper film PCB?

[sasa]

I tent to be quite careful and pedantic when desolder/replace components. If replace capacitors I usually melt it by iron (260C, short duration, second or two at most) and pull up one lead at a time on other side for small and use vacuum pump for larger caps. Now I have a bit of headache with my old TV power supply board...

The most of caps I changed without problems as described, however after replacing a small one (10uf, 50v) on corner trace, opposite pin lifted the whole pin out. And on other large cap (68uf, 400V) I have used vacuum pump and it lifted the whole trace area around it...

Examining closely, they looks and behave just as a film or a foil - actually exactly as aluminum foil.

Is it possible that some power supply PCBs are made entirely from tin copper foil?
Is it some non destructive proper way to determinate it is made of foil?

I had to use jumpers to fix and it is not hard to imagine how ugly that may looks. The whole experience make me to be extra caution.

[helius]

The traces on all PCBs are effectively copper foil (I don't know of any other trace material). It is bonded to the substrate, but can be separated with heat and abuse. Phenolic boards might be more vulnerable to trace separation, as well as boards without any soldermask layer over the copper. Your TV PSU board likely is of this type.
Individual copper foils cut into strips are used to repair broken traces using special tools and techniques to press and burnish over the existing traces. They also come with annular rings to be used on through-hole pad repairs.

[DimitriP]

...or you could do all this:   

[bd139]

You’re applying heat for too long for the sake of being able to suck that joint out.

Don’t use vacuum pumps! They’re terrible. We weren’t allowed to use them doing rework in the defence sector. They lift the pads, deposit fine solder particles over everything and lead to you overheating the joints.

To desolder:

1. Clip the component leads about 2-3mm off the top of the board. If you take the component off it will require less time to melt the solder joint. Time is everything.
2. With some needle nose pliers, wedge some paper in them and grab the lead ends sticking out on the top. The paper stops the pliers acting as a heat sink.
3. Touch the iron on the bottom and pull through. Wait for it to completely cool.
4. Clean the hole up with desoldering braid (chemtronics)
5. Clean the hole with IPA.

The big things:

1. Don’t use a vacuum pump. Get some chemtronics soder wick.
2. Apply heat for the minimum amount of time possible.
3. Let the joint cool completely.
4. Use a temperature controlled iron set up properly.
5. Don’t pull a component out. Clip it off first.

[tautech]

Common rework trap sasa.

All modern appliances/equipment use that Pb free muck on their PCB's.
First step....identify the solder type.....if you have to sit for long on a joint to melt it  , then it's most likely that Pb free muck.
Don't want to risk lifting pads and traces....dilute the muck with some leaded solder to lower the melting point to significantly reduce the dwell time spent trying to melt the Pb free muck. Did I mention it's muck ?

Once diluted, solder suck as normal and at lower temps where the risk of lifting pads is sooooo much less.

[sasa]

Thank you all for comments.

I have gave some tips for desoldering on this forum as well and homemade my own PCBs for some decade ago and also repair my own electronic - all from time to time, in order to avoid excessive cost and sloppy work of electronic services to replace some capacitor with a garbage one and forced me to go again after a month or so and get rip off again and again... Thus, I'm not complete beginner, but this board make me to ask these questions and reconsider all.

With my own homemade PCBs, I never had experience as such. I still use 0.5mm 40/60 Pb based solder wire and do always re-flow joint with fresh on when desoldering, keep off extensive heat, checking first the joint is soldered with lead or lead free, etc, etc.

However, with copper boards I use for homemade, I never experience anything similar. The pads on my PCBs are the same size, quite solid, traces as well 2-3mm wide for such PSU purpose, however I never manage to pull it out (not even 1mm wide one), even I abuse it with 260-350C heat and desoldering pump during initial testing every new copper board type I used. Smallest 0.3mm tick pads and traces can be easily pulled up, but they will stay stiff, not act as an aluminum foil...

There is a large difference in quality of pertinax material (I use only for homemade), "lamination" and copper thickness, but I never had a problems as on this "manufacturers" PCB...

BTW, I can hardly recall what exactly is a standard for thickness of copper (I read long time ago somewhere), however it is some figure around 50 microns. In any event, I will compare it on this board and sheets I use...

I will avoid further both methods I used on manufacturers PCB so far and find some fresh desoldering wick instead.

Thanks again to all for valuable comments.

[bd139]

Good stuff. Home made boards aren’t made by the lowest bidder. Some of the substrate in consumer electronics, particularly that brown stuff is awful. You have to be really careful. It’s worse than the SRBP they used in the 1950s.

[SeanB]

Some boards are made by electroplating the traces on ( most common on things that are mass produced, like calculators and such, where the volume is in the millions and the saving in copper is considerable in plating as opposed to removing and recovering copper) and then using conductive ink as a bonding layer. Think of remote control PCB, which is typically done this way, then has carbon loaded ink screened on as contacts and vias and links on the top side. Those are horrid to repair, not only for the ultra thin traces used, but for the epoxy transitioning to glass when you solder it, and peeling off if you are not careful. Then you get the boards made by Sharp for cash registers, which are the same SRBP, just a lot larger, and with 100V on them as well for the VFD display, and separation is almost non existent, like soldermask as well, except on the via's. At least they do screenprint info on the board.