It is currently Fri Oct 19, 2018 4:45 pm

All times are UTC - 7 hours





Post new topic Reply to topic  [ 16 posts ]  Go to page 1, 2  Next
Author Message
PostPosted: Tue Oct 09, 2018 7:12 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3639
Location: Mountain View, CA
Preface: I'm well aware of the fact that just because a capacitor looks good that it isn't necessarily good. I'm just giving this a shot anyway. And for the record: no, I do not have a desoldering gun/tool, but probably will invest in a nice Hakko unit after this.

History (all this is on my Twitter if you care): my 13-year-and-counting Dell 2407WFP monitor recently died. Review showed bad caps ("Elite" brand), particularly on the video board, but I found general engineering idiocy on the power board as well. I purchased replacement boards (both video and power), so either way I'll have a working monitor at the end of this.

In the meantime, I decided it would be worthwhile to replace the capacitors on the video board with some quality Nichicon ones. Of the 29 total electrolytic caps on the board:

* 17 smaller caps looked OK -- 47uF 25V
* 7 larger caps were bulging -- (5) 220uF 25V and (2) 220uF 35V
* 5 larger caps looked OK but were physically close to the above 7 and of the same spec, so I decided to replace them -- (3) 220uF 25V, (2) 220uF 35V

This board involved what I believe to be lead-free solder, but of a very weird composition (or maybe just age?), where neither my lead-free nor 60/40 tin/lead solders would "mesh" well with it, which made cap removal difficult. I did the first 7, and despite it taking a few hours, I got them out. Continuity tests were fine (I know this isn't a flawless test, but it's better than nothing), so I was happy.

Then I moved on to the remaining 5 today, which proved to be, for some reason, a bigger pain in the butt than the previous batch. The biggest problem was leftover solder in the hole; no matter what I did solder-wise, the stuff just would not come out. I tried higher temperatures too (785F for example -- yikes), no go. I eventually managed to get them clear by replacing my tip with a substantially "finer" tip and working the hole gently while using a pump on the other side of the board. Picture time, followed by details leading to my question:

https://pbs.twimg.com/media/DpHBZiqU0AAiRMD.jpg:large
https://pbs.twimg.com/media/DpHBYnEVAAEcF9h.jpg:large

What I found a bit peculiar -- and maybe this is important -- was that it was always the negative (-) hole/leg (which are all marked in black on that board) that gave me the most difficulty.

I partially damaged at least one of the pads on the component side -- C34 for sure, and possibly C35:

C34: Part of the (-) pad has been torn off entirely on one side of the board (the side visible in the photos). A continuity test between front and back "kind of" works -- depends on the angle I hold the probe. However, as depicted in the photos, I can't tell where the heck that trace goes. Testing the (-) pad to any other (-) pad on the board works the same as front-and-back; depends on the angle I hold the probe.

C35: Visual inspection of the (-) pad is what makes me wary, but continuity tests (from front to back, and from C35 (-) to C36 (-)) pass fine, so I think that one's probably okay, despite looking ugly.

There are no traces on the back of the board for C34 or C35.

So given the above, and assuming the (-) pad of C34 is shot, how exactly do I work around this properly? I literally cannot tell where that trace goes. Do I lift the (-) leg of the replacement capacitor, and solder a jumper wire from that to... what exactly? The same pad but on the back of the board? Just solder the capacitor (-) leg directly to the (+) pin of FD7 (that's connected to C34 (-), I checked)? I'm unsure given that the trace looks like it "splits".

And yes, I know: if this is a multi-layer PCB, then it's very possible I've damaged things beyond repair and I should just give up now.


Top
 Profile  
 
PostPosted: Tue Oct 09, 2018 8:34 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 7671
Location: Seattle
koitsu wrote:
This board involved what I believe to be lead-free solder, but of a very weird composition (or maybe just age?), where neither my lead-free nor 60/40 tin/lead solders would "mesh" well with it, which made cap removal difficult.
I've often found that flooding a joint with fresh hot solder will make removing the older solder easier. Whether it's just better at conducting heat, or if there's additionally something metallurgical, I don't know.

Quote:
What I found a bit peculiar -- and maybe this is important -- was that it was always the negative (-) hole/leg (which are all marked in black on that board) that gave me the most difficulty.
You're interacting with a ground plane, which in addition to being better at reducing electrical resistance, unfortunately is also a huge heat sink. In order to get enough heat to liquefy the solder, you have to get this huge pair of sheets of copper that's attached to this huge chunk of FR4 hot enough at the solder joint to let the existing solder melt.

A higher power iron will have an easier time overcoming the rate at which the copper can conduct heat away.

Quote:
So given the above, and assuming the (-) pad of C34 is shot, how exactly do I work around this properly? I literally cannot tell where that trace goes. Do I lift the (-) leg of the replacement capacitor, and solder a jumper wire from that to... what exactly? The same pad but on the back of the board? Just solder the capacitor (-) leg directly to the (+) pin of FD7 (that's connected to C34 (-), I checked)? I'm unsure given that the trace looks like it "splits".
The black silkscreen makes it harder to see the traces, but all (C25, 26, 34, 35, 36) of those capacitors' (-) sides are connected together.

Quote:
if this is a multi-layer PCB, then it's very possible I've damaged things beyond repair and I should just give up now.
Not multi-layer. Power PCBs almost never are, for heat dissipation reasons.


Top
 Profile  
 
PostPosted: Tue Oct 09, 2018 8:46 pm 
Online

Joined: Sun Sep 19, 2004 11:12 pm
Posts: 20674
Location: NE Indiana, USA (NTSC)
So I guess the ground pin is more likely to be what Celius called the "bitch pin", correct?


Top
 Profile  
 
PostPosted: Tue Oct 09, 2018 8:52 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 7671
Location: Seattle
Yeah, ground "pours" are extremely common.

There are often so-called "thermal [cutout]"s around a through-hole to help with this, but it's still a compromise.


Top
 Profile  
 
PostPosted: Tue Oct 09, 2018 10:35 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3639
Location: Mountain View, CA
lidnariq wrote:
I've often found that flooding a joint with fresh hot solder will make removing the older solder easier. Whether it's just better at conducting heat, or if there's additionally something metallurgical, I don't know.

This is always the method I use. "Going in dry" (har har), i.e. desoldering pencil directly on existing solder, rarely works for me. In this case it worked fine, *except* for old solder stuck between the front and back side of the board -- no matter what I did (incl. using lead-free solder, but 60/40 tin/lead didn't work either), I couldn't get the old stuff to come out. This forum thread sorta documents what I too went through. I think it's a situation that justifies investing in a good desoldering station.

lidnariq wrote:
You're interacting with a ground plane, which in addition to being better at reducing electrical resistance, unfortunately is also a huge heat sink. In order to get enough heat to liquefy the solder, you have to get this huge pair of sheets of copper that's attached to this huge chunk of FR4 hot enough at the solder joint to let the existing solder melt.

A higher power iron will have an easier time overcoming the rate at which the copper can conduct heat away.

Ah ha! I had a feeling it might be that. I should have done a continuity test across all the capacitor holes to see if there was a common one amongst them all. More on that in a moment.

My WESD51 can go up to 850F, but I've been told repeatedly by multiple sources not to desolder at temperature, and instead to stick to ~750F tops. I believe extensive prolonged heat (while trying to deal with this problem) is what caused the pad to essentially melt/break off. I guess this is just something I need to get more experience with (desoldering, I mean). It's something I'm OK at, but I don't do it often.

lidnariq wrote:
The black silkscreen makes it harder to see the traces, but all (C25, 26, 34, 35, 36) of those capacitors' (-) sides are connected together.

I just checked -- yeah, you're right. Here's a visual: https://imgur.com/a/HiNv0jh

Okay, so these are all ground. So let's assume C34 (-) has a missing pad entirely -- what do I need to do with the new capacitor's (-) leg? Should I use a jumper wire to solder it to another GND pin/hole, ex. C35 or C36? (Remember, the rear of the board has no traces to any of these capacitors)

I'm just not sure what's the "right thing" to do here for functionality. Maybe I'm being paranoid because I just watched some videos on capacitors in series vs. in parallel, haha.

lidnariq wrote:
Not multi-layer. Power PCBs almost never are, for heat dissipation reasons.

This is the video board, not the power board. The thickness likely implies it's 2-layer only, but I'm never sure.

The power board actually connects to this one in a completely separate area of the board -- far, far away from where I'm working; I'm not quite sure what these caps are used for. This video board has -- no joke -- 4 interconnects on it that all go to different other boards: an 18-pin FFC that goes somewhere (very close to the power connector), a 30-pin connector that goes to the power board, a 10-pin connector that connects to something (I think the CCFL/lighting control board? Unsure), and finally a 5-pin connector to an SD/USB board. Here's another picture with those interconnects labelled, just for fun: https://imgur.com/a/Ylo64n0

Also, thanks for the help and information! I'm always trying to absorb hardware-oriented knowledge so I feel like less of an idiot...


Top
 Profile  
 
PostPosted: Tue Oct 09, 2018 11:46 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 7671
Location: Seattle
koitsu wrote:
My WESD51 can go up to 850F, but I've been told repeatedly by multiple sources not to desolder at temperature, and instead to stick to ~750F tops.
Not hotter! 650-700F should be fine. But higher power isn't the same thing—it's the ability of the iron to keep up with something sinking heat. Turning up the temperature is a hack that uses the thermal mass of your iron to balance out the lack of heating power.

Quote:
I believe extensive prolonged heat (while trying to deal with this problem) is what caused the pad to essentially melt/break off. I guess this is just something I need to get more experience with (desoldering, I mean). It's something I'm OK at, but I don't do it often.
Absolutely correct. The copper delaminates from the FR4 because of some combination of temperature and mechanical pressure—I've never really tried to figure out why, just the sinking feeling when I botch it again.

koitsu wrote:
Okay, so these are all ground. So let's assume C34 (-) has a missing pad entirely -- what do I need to do with the new capacitor's (-) leg? Should I use a jumper wire to solder it to another GND pin/hole, ex. C35 or C36?
I mean, it is on a big fat wide trace to minimize series resistance, but if you just connect to C35 on the back it's probably short enough to not pose a problem.

Usually I leave the component's leg untrimmed, put a piece of insulation on it, and fold it over to solder somewhere else, using the insulation to prevent shorts.

Quote:
This is the video board, not the power board. The thickness likely implies it's 2-layer only, but I'm never sure.
Ok. This is still clearly another power section—large capacitors, inductors, wide traces, and SOICs where lots of legs are shorted together. And so it's still almost always two layers, for the same reason.

(Either way, it's easy to see if something is 2 layer or if there's more hidden: if you backlight it, you'll see random bits of light shine though in random places if it's 2 layer. Multilayer boards basically always have an internal ground pour, blocking light transmission)

I think the only exception I've ever seen is planar transformers where some versions do a turn per layer. A few years ago I'd seen a video of someone doing a destructive disassembly of some very high end and outdated Sun server. They set up their milling machine to take off a couple thou, then took a picture, and made a timelapse of it. Can't find it now, though.

koitsu wrote:
I'm always trying to absorb hardware-oriented knowledge so I feel like less of an idiot...
And asking questions is how you do that :)


Top
 Profile  
 
PostPosted: Wed Oct 10, 2018 10:33 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3639
Location: Mountain View, CA
What I ended up doing: running both legs of C34 through their solder holes, soldering them in place (as best I could for the (-) leg**), then very carefully bent the excess portion of the (-) leg over towards the (-) leg of C35. Did the same thing with C35 as well, then soldered the two (-) legs together, wrapped that in insulation, then covered the ~1x1" patchwork area with a square of electrical tape (which doesn't look like it's going to hold).

I won't know if it works until I get the replacement power board, but that just arrived in the US today, so hopefully it'll reach me by end of week.

If one if the (+) legs doesn't have good connection, or isn't connected at all (on the component side), any ideas of what could happen? I imagine anything is possible, but I'm mainly worried about things exploding.

Also: the back of these boards -- both my repaired board and the replacement video board -- are absolutely filthy. I can't describe it with words exactly; it's like there's a layer of red/brown liquid scattered all over it. I did notice that when I applied flux that the flux seemed to clean it off; on a paper towel the stuff came out brown/red-ish. Any idea what that substance is? Probably toxic as hell...

Soldering feels so much fun when you're doing it on a new/fresh board: everything flows nicely, you can tell the condition of everything, and the results look great. Desoldering stuff that used a weird lead-free mixture? Not fun at all. Maybe I'd get more enjoyment out of it -- feelings of accomplishment rather than anxiety and worry -- if I took the plunge and got a nice Hakko desoldering station. I don't do this type of work often enough to justify the cost, but at the same time, when I DO have to do it these days, it's almost always on something "old" (10+ years, sometimes 30 years, like when I replaced ~20 caps on a Dragon Breed arcade PCB and the end result worked and felt great. Lead/tin solder FTW!)

** -- In fact, many of these solder holes did not have good "through" flow. I had to re-do the soldering several times on several caps, but 4 or 5 did not want to behave. So, I think there may be cold joints, though continuity tests from the rear of the board (solder side) to the component side seem to pass; I tested as many (+) legs as I could too. I think I damaged more pads with the higher temperatures than I originally thought. Ah well, if it blows up, at least I have a spare video board.


Top
 Profile  
 
PostPosted: Wed Oct 10, 2018 11:11 pm 
Offline

Joined: Fri Feb 24, 2012 12:09 pm
Posts: 654
I am normally using a fine 1mm tip with 310'C (600'F) for normal stuff. And a bolder 2-3mm tip with 370'C (700'F) for desoldering, and for power/ground sections, or for lead-free solder. Some extra heat is helping there, and I think the bolder tip is also useful to get the heat transported.

For flux, I am just adding new solder (which has fresh flux in it's core). I've never used 'raw' flux, but doing so could be useful for desoldering and for SMD soldering. Just make sure that the flux is supposed for use with electronics, I think the more aggressive flux variants could damage the board/electronics.

I've had troubles getting capacitors pulled from tiny holes, too. For installing new capacitors, I would just heat up the solder pads, and push the capacitors in (there's no need to remove the old solder for components with only two pins).


Top
 Profile  
 
PostPosted: Wed Oct 10, 2018 11:14 pm 
Offline

Joined: Sun Apr 13, 2008 11:12 am
Posts: 7671
Location: Seattle
koitsu wrote:
If one if the (+) legs doesn't have good connection, or isn't connected at all (on the component side), any ideas of what could happen? I imagine anything is possible, but I'm mainly worried about things exploding.
While it's conceivable that too little bulk capacitance will cause something to oscillate and blow up, in my limited experience it mostly just causes some self-test somewhere (in the power supply or somewhere else) to shut things down.

(I recently recapped an old LCD monitor for my parents; the symptom was that the screen would only display content for a second or two; afterwards both the LCD drivers and backlight would turn off.)

Bad solder joints on the capacitors ought to be comparable.

Quote:
Also: the back of these boards -- both my repaired board and the replacement video board -- are absolutely filthy. I can't describe it with words exactly; it's like there's a layer of red/brown liquid scattered all over it. I did notice that when I applied flux that the flux seemed to clean it off; on a paper towel the stuff came out brown/red-ish. Any idea what that substance is? Probably toxic as hell...
... if it was only the original board, I'd've guessed it was the electrolytic capacitor electrolyte. But on the replacement board too? Eh....


Top
 Profile  
 
PostPosted: Thu Oct 11, 2018 5:10 pm 
Offline
User avatar

Joined: Mon Apr 04, 2011 11:49 am
Posts: 2092
Location: WhereverIparkIt, USA
Just curious if you tried the method where you fill the hole with fresh solder, stick a plunger solder sucker to the back side, heat with iron from the front side (just heat the pad, don't block the hole), then click the plunger. If the joint is fully heated, and sucker is well seated it's quite a suction force to be reckoned with.

I do use a cheapo bulb desolder iron as a starting point, but these cheap tools have kept me from feeling like a expensive vacuum desolder iron.

When it comes to surface mount, desolder wick (always apply kester186 flux to wick & joint before using) and an old used weller heat gun I picked up from ebay for $60 covers all my needs.

Anyway, just something to think about. The sum of all these desolder tools cost me well under $100 and has covered my through hole and surface mount needs for years. Expensive tools aren't always the only/best option, but one's budget has a lot to do with that choice.. If one does a large amount of through hole desoldering regularly I can see ~$300 gun being worthwhile, I'm sure it's much faster than my methods. But I only desolder through hole components a few times a year at most.

That said, I slaughter weller soldering irons like yours every year or two it seems. I assume it's something to do with me frequently powering on/off multiple times a week for very short periods. The platinum temperature sensor always goes out in the wand/pencil tip and costs ~$50 to replace. So when my ~5th one in the past decade I finally gave up on weller decided to try my luck with a hakko FX-888D on ebay for $85. I'm especially looking forward to the bent tip T18-BR02. I used the bent tip on a friend's hakko awhile back and it was an absolute dream for soldering 0.5mm pitch SMD, couldn't find anything comparable that was compatible with weller.

_________________
If you're gonna play the Game Boy, you gotta learn to play it right. -Kenny Rogers


Top
 Profile  
 
PostPosted: Thu Oct 11, 2018 8:10 pm 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3639
Location: Mountain View, CA
infiniteneslives wrote:
Just curious if you tried the method where you fill the hole with fresh solder, stick a plunger solder sucker to the back side, heat with iron from the front side (just heat the pad, don't block the hole), then click the plunger. If the joint is fully heated, and sucker is well seated it's quite a suction force to be reckoned with.

This is close to the method I used at the end, but the quest getting there might shed some light on things. It's hard to explain, but I'll try.

The front (component) side of the board only contains small/thin "rings" for what would be called solder pads. There was literally no clearance between the bottom of the caps and the board. The back side of the board had actual solder pads (larger silver areas surrounding the hole).

The old solder had a very "dull" look -- light grey in colour, but a strong matte/rough finish. Not a single joint was shiny. (Research told me that this was likely lead-free solder, but keep reading)

Removing the old caps involved:

- Applying a very large amount of heat (~800F) to the legs of the (bad) capacitors on the back of the board. This got me literally no where; the old solder would not liquefy. All that happened was that the caps themselves got quite hot (makes sense)

- Applying a reasonable (600F) amount of heat to the solder pad on the back of the board. Sometimes the solder would liquefy on the back of the board (easy to tell visually), but not enough for the cap legs to wiggle -- i.e. the solder in the hole and near the front of the board would not liquefy. I did not dare push/pull/wiggle the caps harder due to worry that I might break pads/traces/conduits

- Applying new solder (lead-free) to the existing solder pad/leg on the back of the board, at around 650F. The new solder liquefied easily, but would not "mesh" or "mix" well with the old/existing solder

- Applying new solder (60/40 lead/tin type) to the existing solder pad + leg on the back of the board, at around 650F. This had the same problem as the above

- Same method as just described, except with very large amount of heat (800F). This worked -- the cap legs started to wiggle, and the caps usually came out.

In almost all cases, this left residual solder in the holes.

I also want to add that the old caps had their pin pitch wrong -- too wide by about 0.2-0.3mm (the replacement caps I used were a proper 0.5mm so they dropped right in no problem). So whoever did the assembly chose to bend the legs kinda like this:

Code:
|            |
|            |
| Cap bottom |
--------------
    /    \
  _/      \_
  \        /
   \      /
    \    /
     |  |

Sometimes only one leg was bent like this. Regardless of why, it made removing ones like this stupidly difficult when compounded with the solder situation.

How I went about cleaning the holes out -- all working from the back of the board, as the front had components and I didn't want to risk working there unless I had to (I eventually did):

- Applied new solder (60/40 lead/tin type) to the existing solder pad on the back of the board, at around 650F, while simultaneously using a spring-loaded desolder pump to pull the solder out from the hole (also from the back). This method would remove solder, but only the new stuff I just put down, not the old matte stuff stuck in the holes

- Same as above, but using high temps (800F). This worked some of the time, but 5 or 6 holes did not want to give up their solder

- Switched to using desoldering braid to try and empty the holes out -- got some, but solder remained in the holes anyway. I also started using flux at this point (my lead/tin one supposedly already has flux on it, but I figured applying some wouldn't hurt)

- Examined front of the board, where I would find "blobs" of solder sticking through the holes, but not connected to the "rings" I described further up. In other words: solder was visibly in the hole via the front, *and* back, of the board -- yet didn't look connected to anything

- Switched back to the pump, changed soldering tip to an extremely fine one (tip is almost like a needle), ran at 650F, using the tip almost like a drill, very gently working it into the hole on the back of the board, while using the pump from the front of the board. This worked for I think 1 hole, but nothing else

- Same as above, but at 800F -- this got the remaining holes cleared for sure, but by then the damage had been done.

So, given the story, I think how the pad(s) got damaged becomes obvious. Looking back on it all, it sounds like I put the board through hell (feels more like it put me through hell).

In the middle of this craziness, that's when I found the aforementioned-linked thread of another fellow running into the same problem. There's good info in the thread, but it's all rehashing existing methods. He ended up using an actual needle or something like that and very slowly "drilled" the holes out. Another guy actually recommended using drill bit. Yeah, sorry, not gonna do that.

One person said something that resonated with me: that basically the problem stems from lead-free solder. I did some further digging around online, and it seems that with the start of the RoHS movement -- which started around 2003 in the United States -- some companies started experimenting with different mixtures. There's supposedly large variance between what's used today and what may have been used 12-15 years ago. So If I had the exact type of solder they had used during the assembly process, the stuff probably would have come out without a problem.

If the board doesn't work, I'd love to send it to someone just as a general "here, YOU try it" experiment. Once you see what I was dealing with, I think it might make more sense. But then again, doing this type of work is not something I do day in day out. Working with new components/boards/etc. is so much nicer.

Switching subjects since it was brought up:

How would I know if the temperature sensor in the wand/pencil tip "goes out"? I've had the same pencil that came with the WESD51 since I got it long ago: a PES51. I actually bought brand new tips for this project (and they're quite nice, actually, despite being Chinese) -- the ETS tip (one on the far left) is what ended up helping me clear the holes. The reason I ask about the sensor is, maybe that sensor or some part of my PES51 pencil isn't working right any more, or maybe the temperature read-out is way off?

I'm not a big fan of dropping money on something unless I know it'll improve the situation and act as an investment. I've just found that every time when dealing with "old boards" and "old solder" I run into situations akin to this in some way, which makes me think a Hakko FR-301 with some extra tips would minimise me causing pad damage.


Top
 Profile  
 
PostPosted: Fri Oct 12, 2018 12:40 pm 
Offline
User avatar

Joined: Mon Apr 04, 2011 11:49 am
Posts: 2092
Location: WhereverIparkIt, USA
Quote:
One person said something that resonated with me: that basically the problem stems from lead-free solder.

Yeah it does sound like that especially with the problems you had with new solder not mixing.. That combined with the bend leads is what sounds like created the ultimate hell. Based on your story, I'm not convinced an expensive desolder tool would have actually made your task 'easy'..

Quote:
How would I know if the temperature sensor in the wand/pencil tip "goes out"?

It simply doesn't turn on. The 'currently heating' LED doesn't turn on. I measure the resistance of the pins on the wand and find the RTD pins are open to confirm that's what's broke. The latest one that failed me it seemed to be an intermittent open, a bit of mechanical agitation got it working again. But didn't take long till a good whack didn't fix it.

I have seen issues with these weller irons not regulating temp properly. Was using my brother's awhile back and several mins into the job the temp shot through the roof. IDK what the temp actually was but it was obviously over temp as noted by how quickly the iron tip got crazy gross from over heating. Turned it off, and back on and problem would go away. But later the next day it did the same thing all over again.

So I guess things can fail somehow without fully failing like I'm used to with my 5 previous weller irons. It's possible that if you've got a poorly sized tip (might happen if it's not legit weller), where the RTD doesn't make good thermal contact with the tip. So the RTD is sensing a lower temp than it really is, thus causing the tip to overheat. Perhaps if you're seeing distinctly different temp/thermal behavior between tips of comparable size something like this may be going on. I don't think that would really explain the issues I had with my brother's iron. Sounds like that issue was some problem with the station that got resolved with power cycling.

_________________
If you're gonna play the Game Boy, you gotta learn to play it right. -Kenny Rogers


Top
 Profile  
 
PostPosted: Sun Oct 14, 2018 3:01 am 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3639
Location: Mountain View, CA
infiniteneslives wrote:
Yeah it does sound like that especially with the problems you had with new solder not mixing.. That combined with the bend leads is what sounds like created the ultimate hell. Based on your story, I'm not convinced an expensive desolder tool would have actually made your task 'easy'..

I'm not convinced either -- tools can't replace experience, and I certainly need more of that. It's just that every time I deal with old equipment and replacing simple parts, the desoldering effort turns into a nightmare due to some annoying aspect. I feel like if I had a tool that was a combination of a heated tip and vacuum, I could avoid risking damaging pads (like in this case). I'm sure if I had one of you guys doing it, you'd have had a lot less trouble, aside from the caps with their bent legs; the few times I've seen actual EE guys do desoldering work, they do it so quickly and easily that it almost seems magical.

infiniteneslives wrote:
It simply doesn't turn on. The 'currently heating' LED doesn't turn on. I measure the resistance of the pins on the wand and find the RTD pins are open to confirm that's what's broke. The latest one that failed me it seemed to be an intermittent open, a bit of mechanical agitation got it working again. But didn't take long till a good whack didn't fix it.

Ah! I haven't run into that with my WESD51 / PES51, but I have possibly run into something you're about to describe in the next 2 paragraphs:

infiniteneslives wrote:
I have seen issues with these weller irons not regulating temp properly. Was using my brother's awhile back and several mins into the job the temp shot through the roof. IDK what the temp actually was but it was obviously over temp as noted by how quickly the iron tip got crazy gross from over heating. Turned it off, and back on and problem would go away. But later the next day it did the same thing all over again.

So I guess things can fail somehow without fully failing like I'm used to with my 5 previous weller irons. It's possible that if you've got a poorly sized tip (might happen if it's not legit weller), where the RTD doesn't make good thermal contact with the tip. So the RTD is sensing a lower temp than it really is, thus causing the tip to overheat. Perhaps if you're seeing distinctly different temp/thermal behavior between tips of comparable size something like this may be going on. I don't think that would really explain the issues I had with my brother's iron. Sounds like that issue was some problem with the station that got resolved with power cycling.

This may be something I've seen, but then again maybe this is by design: when increasing the temperature, the warmer will often go far past the temperature I set it at (by sometimes as much as 40F), sit there for a good 15 seconds, then slowly drop down to where I set it. It does this with both the official Weller tips I have (just two) and the 10 Chinese ACE brand tips I got. A key point: I definitely don't remember my WESD51 doing this when I got it years ago (I remember getting it and being absolutely giddy having a decent soldering station).

The inside of my PES51 -- where the tips connect, that is -- looks like the aftermath of a badly-managed camp fire. I could take a photo of it if you're curious; it doesn't look tarnished, it looks downright charred. It makes me wonder if internally something's amiss. Maybe I'm having to set the warmer at really high temperatures to make up for some kind of pencil problem that's developed? I don't know because I haven't used other soldering pencils or stations, just classic soldering irons from the 80s.

If my PES51 is misbehaving, I'm reluctant to replace it simply because of what you've described. It might make more economical sense for me to buy a Hakko soldering station and know I'm getting something reliable (every hobbyist and professional I know that owns Hakko stuff has raved about how well-made it is and how it holds up for many years).


Top
 Profile  
 
PostPosted: Sun Oct 14, 2018 3:28 am 
Offline
User avatar

Joined: Sun Sep 19, 2004 9:28 pm
Posts: 3639
Location: Mountain View, CA
Back on topic a bit:

My video board repair did in fact seem to hold up -- that is to say, the board did work. However, after an hour or so of the monitor being powered on (using the new power board too) + heating up, upon powering it off, it exhibited the same problem as what I originally ran into. That means there's something else wrong with the board. It's not worth me trying to fix at this point, because: I ended up swapping in a new replacement video board, and so far the monitor hasn't had a single hitch when being shut off -- even after being on for several hours. I now once again have a working primary monitor. :-)

So in short: despite the damage I did to the pads, the soldering held up and replacement caps were fine, but the board must have sustained some other damage. I'm lucky I managed to find an identical replacement video board on eBay.

Several forum threads indicate that the Genesis FLI5962H-LF chip on the video board is a common piece that gets internally damaged when a faulty power board rears its head. A common analysis tactic involves using a hair dryer on the video board to see when things begin to crap out, and it's usually that chip. More on heat in a moment because it's relevant.

Another common problem is with the T-con board (which drives the panel, re: timing), which isn't something you can easily get to without disassembling the rest of the monitor, including some panel bits. The one thread I read about that involved a guy who attempted to replace it and ended up damaging all sorts of stuff in the process. The board is easily accessible ("oh it's just 4 screws and 2 flat cables!"), but it won't come out once you undo all of that. It's attached to something else in the monitor (i.e. very close, or up against, the panel) that you can't get to without going in via the front; almost like there's a screw or some other kind of interconnect holding the centre of it in place.

I have a general rule of thumb: I don't mess with panels (I did a Dell laptop panel swap many years ago and somehow managed to completely kill the laptop's video circuitry or GPU), and I absolutely DO NOT mess with high voltage: which is exactly what the inverter board for the CCFL is for. Nope nope nope. No way.

But back to heat: this monitor runs *stupidly* hot. The maximum operating temp is cited at 95F, yet my thermal gun pointed at the back of the ventilation slits (directly above the inverter board and power board) read almost 110F after being powered on for a couple hours -- and that's with the new boards. When I got it new back in 2006 or 2007, it definitely wasn't this hot, so something over time has caused this problem. Others have run into this problem in the past too, nobody is sure why it happens, but it's abnormal. So I suspect hardware has been damaged due to excess heat.

For now I have a working monitor (though [url]with a fan blowing on the back of it as a precaution[/url]), and am in the process of buying + trying other substantially newer monitors (particularly 2560x1440 ones, since I really need the pixels for some NES-related stuff I'm doing)... but there's also a local guy selling a Dell 2407WFP Rev A03 for an incredibly reasonable price, so I might just pick that up anyway. That's how much I love these things.

All in all, it's been a stressful project, but educational. I do enjoy replacing bad caps in things; done so on old motherboards and some arcade boards. It's such a great feeling repairing something and seeing it work once again.


Top
 Profile  
 
PostPosted: Sun Oct 14, 2018 7:07 pm 
Offline
User avatar

Joined: Wed Feb 13, 2008 9:10 am
Posts: 660
Location: Estonia, Rapla city (50 and 60Hz compatible :P)
CFL based panels have the tubes suck more power as they age and the inverter transformers and drivers will get much hotter than they used to over time. I have even seen the PCB looking completely charred and bubbled in some monitors. Only fix is swapping the tubes in the panel but that's often a lot of work and with a very high chance of failure due to the tubes being very PITA to get out and even worse to get them back. They're not made to be servicable.

_________________
http://www.tmeeco.eu


Top
 Profile  
 
Display posts from previous:  Sort by  
Post new topic Reply to topic  [ 16 posts ]  Go to page 1, 2  Next

All times are UTC - 7 hours


Who is online

Users browsing this forum: No registered users and 3 guests


You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot post attachments in this forum

Search for:
Jump to:  
cron
Powered by phpBB® Forum Software © phpBB Group