Desoldering trouble - Pad repair questions

[kca]

Re: Desoldering trouble - Pad repair questions

Thanks again, Michael.

I got the Rubycon ZL capacitors (3300uf 10v 105 degree) from a vendor in Austin, TX. They arrived yesterday and look to be the genuine article. Bought 10 of them so I can experiment a bit with one or two, have two for the actual fix, and then have five or more left over for repairs of other boards in the future.

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

That sounds just the job, a quick clean of of the flux with flux remover after you have finished and you wont be able to tell they have been changed, always get good quality caps from a reputable supplier , a lot of stuff for sale on e bay is fake out counterfeit crap .

[kca]

Re: Desoldering trouble - Pad repair questions

Okay.
That's good to know about the top ring pad not being imperative to the repair.

----------

I agree, Michael, with what you said earlier. I am actually trying to change as little about the configuration of the board as possible.


Given all of the good advice that you, and others, have given, here is my plan of attack for the repair. Please let me know if you think it's good in total, or if it is flawed in any way:

I am going to leave both of the through-hole positives alone and just solder in the new cap leads there per usual as they both show very clear through-hole, or via, conductivity.



On the C8059 negative, I will scrape the top (being sure to stay within the light green track as I do), and solder the strand wire to the top of the board as you've suggested. Then, drop that through the hole and just solder the strand and the cap lead together on the bottom of the board without any further scraping on the bottom of the board.

On the C8060 negative, I will scrape the top (staying in its light green track), and solder that strand wire to the top of the board as you've suggested. Then, drop that strand wire through the hole. I will then put that cap lead through and cut it down to about 1/2 inch. I'll slip the circular ring pad I saved over this cap lead (and the strand wire too, if possible) and fit that pad down into the burned area so it fits as tight there as possible. I will carefully scrape the area of dark green just to the left and slightly up from the C8060 hole here on the bottom of the board (about the 10 o'clock position), and then bend both the cap lead and the strand wire onto this scraped area and then solder them there. I'll finish by making sure the total amount of solder here on the bottom of C8060 negative covers everything related to the fix (meaning, the solder will form over the cap lead and strand wire where they come out of the hole at the and through the replaced ring pad, and that the solder also flows all the way over to the newly scraped area in the 10 o'clock position so that it is one, continuous "river" of solder in that area).


Does that make sense as an overall approach?


Note: I am also going to try to flow as much solder down into the through holes during the fix as possible, to help ensure that those via conductivities are as effective as possible.

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

Thank you for the detailed and clear pictures, Michael. Much appreciated. I can see the technique now.

I am assuming that I will need to stay within the light green traces on both the C8059 and C8060 negative sides when choosing where to scrape, and be judicious when applying the solder onto those scraped areas.

Is that true, or does it not matter where one scrapes? Another related question is, if I happen to "spill" some of the fresh solder from one trace to another, does that matter or is it a moot point because it is forming over the green laminate of the PCB board itself (as opposed to any scraped/exposed part)?

Hope those queries are clear. If not, I can reword them.

On these two photos the top ring which is missing does not appear to be connected to anything and in this case you dont need to do anything just solder the cap in .

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

Hi , If what you are saying is that although the track is damaged you still have top to bottom continuity but if this is on the negative of the capacitor while it may work o.k without any repair you may have ground currents flowing where they were not intended to do, so the set may function o.k but you could introduce different faults such as RF interference on you picture, and dont forget someone took a long time to design that board so why try to change it .

[kca]

Re: Desoldering trouble - Pad repair questions

Also, on another thread, I got some advice I'd like to run by you. This other poster asked me to take my Digital Multimeter and measure for Ohm shorts between the top an bottom through-holes on all four holes (2 on C8059 and the 2 on C8060).

Here are the results from that testing I did last night:


I did as you suggested and my DMM read shorted (i.e., it sounded) on all of the following top to bottom connections:


C8059 top of board positive to bottom of board positive. This sounded as shorted as expected.


C8059 top of board negative through to the bottom of board negative.
(So, despite the partially damaged ring pad, that one looks like it can be soldered as you suggested in the last post. Is it still best to scrape some coating off of the bottom trace even though the DMM sounded every time, then bend the cap lead in that same direction? If so, I assume you mean to scrape a bit off in the downward direction where the trace clearly leads down out of the hole.

______________ ______________ _____________

C8060 top of board positive to bottom of board positive. This sounded as shorted as expected.

C8060 top of board negative through to the bottom of board negative.
(Here, somewhat unexpectedly, I also got the sounded short signal when I stuck the DMM probe deeply enough into the through hole to apparently make contact with some conductive material. So, I was glad to see I hadn't damaged that in full. Next, I took the in-tact ring pad that had come off of this burnt hole (I had saved it during the desoldering process), and laid that gently on top of the burned area. When I tested again with the DMM, this time I only laid the probe on top of the loose ring to see if it was making contact with the metal in the through hole and it was. The meter sounded short with this method as well.

So, in regard to the badly damaged C8060 negative on the bottom of the board, can I simply fit that small ring pad over the cap lead once it is in and then solder that whole joint fully, or should I scrape the board and bend the lead as we discussed earlier just to be more sure?

_____________________________________________________

Michael ~ Does any of that change your opinion about the need for the strand of wire through the hole technique you have suggested I use? It seems like it would still be a good idea to get those wires through at least the two negatives that appear questionable, and maybe the two positives as well. All of which seems like a good insurance policy as I try to make this repair as solid and as lasting as possible.

[kca]

Re: Desoldering trouble - Pad repair questions

Thank you for the detailed and clear pictures, Michael. Much appreciated. I can see the technique now.

I am assuming that I will need to stay within the light green traces on both the C8059 and C8060 negative sides when choosing where to scrape, and be judicious when applying the solder onto those scraped areas.

Is that true, or does it not matter where one scrapes? Another related question is, if I happen to "spill" some of the fresh solder from one trace to another, does that matter or is it a moot point because it is forming over the green laminate of the PCB board itself (as opposed to any scraped/exposed part)?

Hope those queries are clear. If not, I can reword them.

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

Yes this Psu is two layer, it will be fine just push the lead of the cap though as normal and where the pad would be just flow alot of solder and it will go down the lead on to the pad on the other side. I done it a few times on Sus boards and Power Supplies and dont that and they all worked fine

Heres a tip though for boards like that if you plan on taking up more repair jobs, you should really get a desoldering iron which will mean that you wont be ripping the pads off any more.

The big problem with just using a desoldering iron and fiberglass boards is you just cant get enough heat to the top of the board, the desoldering station I use can do anything but it would struggle to remove parts from double sided boards without first pre heating.

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

Hi , As promised photos of how to repair this problem , they are in order and pretty self explanatory, on the top scrape away the varnish to expose some clean copper and solder fine wire on top , push it through the hole and then just solder new cap back in.

[kca]

Re: Desoldering trouble - Pad repair questions

Thanks again, RB.

Yes, I plan on being more prepared next time I attack one of these types of boards. The first few boards I did were far simpler re: the desoldering process, and I didn't realize it would be so difficult to remove the leads this time. I definitely appreciate your advice.

[ReeceyBurger123]

Re: Desoldering trouble - Pad repair questions

Yes this Psu is two layer, it will be fine just push the lead of the cap though as normal and where the pad would be just flow alot of solder and it will go down the lead on to the pad on the other side. I done it a few times on Sus boards and Power Supplies and dont that and they all worked fine

Heres a tip though for boards like that if you plan on taking up more repair jobs, you should really get a desoldering iron which will mean that you wont be ripping the pads off any more.

[kca]

Re: Desoldering trouble - Pad repair questions

Thanks for your input, RB123. I was hoping that was the case, but wasn't sure it would be the long-term fix I am seeking to achieve. So, the key is to get enough of that solder to flow into the through-hole to make a connection with the copper ring on top, right?

Can you tell me why the 2-layer design allows for a fix where the pad need not be replaced? And, if I were to attempt to fix it in this manner, will the new solder on the bottom side then just connect to the trace "automatically", assuming I cover the cap lead that will stick through the bottom with enough of said solder?

[kca]

Re: Desoldering trouble - Pad repair questions

Photos would be great, Michael! Thank you for offering to post them.

I am in no hurry. The new capacitors have been ordered but they won't be here for several days yet. Even after they arrive, I am going to be sure the repair technique is thorough before I do any soldering of new caps.


I can certainly wait for your further reply.

Thanks ~

[ReeceyBurger123]

Re: Desoldering trouble - Pad repair questions

Yep its a 2 layer, should be fine with out the pads or wires to sub the pads you peeled off. Stick the cap in and flow a a decent amount of solder to the leads it will make its self to the top pads. I done this before and doing so worked fine

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

Thanks, Michael ~

So, this wire strand then just sits in the hole loosely alongside to the cap lead until all of it is soldered together? I'm not sure I am understanding. How does this arrangement connect to the trace on the bottom of the PCB?

If you look closely at hole #1 on C8059, you'll see a slight ripple where that pad leads into the trace. So, I am thinking I have to replicate that same connection on hole #3. It's my first repair of this kind, so I apologize and am in need of some very specific directions and the rationale behind them would be much appreciated as well so I can learn both the how and the why.

No not loose in the hole, scrape away some copper on the top side and solder it to here first, then feed the free end though the hole and re fit the capacitor, soldering the capacitor lead out to the wire sticking out of the hole, if you want and I can still find this thread in the morning when I am back at work I could post some photos ?

[kca]

Re: Desoldering trouble - Pad repair questions

Why so?

Are there any identifying marks on the board itself so I can let you know? I have included two more images for your review.

[stj]

Re: Desoldering trouble - Pad repair questions

i hope it's only a 2 layer board!

[kca]

Re: Desoldering trouble - Pad repair questions

Thanks, Michael ~

So, this wire strand then just sits in the hole loosely alongside to the cap lead until all of it is soldered together? I'm not sure I am understanding. How does this arrangement connect to the trace on the bottom of the PCB?

If you look closely at hole #1 on C8059, you'll see a slight ripple where that pad leads into the trace. So, I am thinking I have to replicate that same connection on hole #3. It's my first repair of this kind, so I apologize and am in need of some very specific directions and the rationale behind them would be much appreciated as well so I can learn both the how and the why.

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

Hello all ~

I am working on a PSU from a Philips 50# (50PF7220/37) plasma and needed to replace two bad caps at the bottom center of the board. I went at it too aggressively and ended up taking one pad off completely and another one nearly off. Hoping to get this back in order before proceeding any further.


BOTTOM OF BOARD:
If you take a look at the pic of the bottom of the board, you'll see (counting from the left) the first pad that is still there, but hanging to the run somewhat precariously.

The second one looks to be in tact still.

The third is the one I'm mostly concerned about, as this is the pad that came completely off (I still have the pad, and it's fully intact (i.e. circular)).

The fourth one seems to be in tact just like the second.

Looking for some expert advice on what I need to do to repair this before installing the two new caps (3300uf 10v 105 degree). Specifically, here are my questions:

1) Will I have to drill hole #3 and put in some kind of funnelet and then attempt to reattach to the trace run? Or, as is my hope, is there an easier way? As I mentioned, I still have the in-tact ring pad and that can simply be dropped back into the blackened hole. But, just soldering that back into place does not seem like it will make the proper connection to the run.

2) Given the type of board it is, do I need to be concerned about repairing both the bottom pads/traces, or the top, or both? I read a few forum posts that indicated I might be able to ignore one or the other, but could not make complete sense of that advice.

3) Can hole #1 simply be soldered after the new cap lead is dropped through? It is still attached to the run, but the ring itself looks like it might be compromised. Will the solder fill in any gaps and form a proper connection?

4) Do holes #2 and #4 look like they can be left alone?



TOP OF BOARD:
Also, I have included a photo of the top of the board too. Here, one of the small copper rings came off during the desoldering process and was lost. I am wondering how imperative it is to find or replace that before installing the new cap.

Thanks in advance for your replies / advice.

On your C8060 the connection to one end of the cap is made on the top of the pcb under the cap itself so it will be very difficult so solder this on top of the board without linking through with some wire first.

[michael dranfield]

Re: Desoldering trouble - Pad repair questions

On some boards the thru board link is very important as it can link a track from the top of the board to the bottom , however if the copper track on the top plane is not connected to anything theres no need to link it through , you can use a single strand of virtually any multi strand flexible wire such as speaker cable , to prevent this happening in the future pre heat the board first to around 100 degrees c before desoldering the component in question.