Post your worthless and/or pointless repairs

[eccerr0r]

Have not forgotten:

Why did I bother... I'm currently using secondhand units from a thrift shop that will never have this problem.

[прямо]

Card runs at X16 PCIe 2.0 on an old AM3 motherboard

[прямо]

Good save!

I don't think you can get a decent 5V 3 Amp power adapter like that for $1 new from the store.

I've seen better design from a $2 adapter believe it or not, and yeah, I bought this awhile back for $1. Price don't change for small power adapters like this.

The core appears to be running nice and cool, though - after all, 52C for GPU @ 100% load is not bad at all. I have the EVGA version on these cards (just regular GTX 560's) with the single fan in the middle, and those heatsinks designs suck even more. Sure my VRM MOSFETs might be getting cooled... but what good is that when the GPU core is running miserably hot?
No thanks, I'll take a refference blower -style cooler any day. It also removes heat from the system (PC case) rather than just re-circulate hot air inside the case.

Yup, the core is cool and dandy but the mosfets aren't, they get very hot. I have two of these cards, with a reference blower and with a big fan in the middle. Both has the same core temp, but mosfets are not running hot on the one with the big fan.

BTW, you mentioned you fixed your card, but I see it's stuck in 1x PCI-E mode in GPU-Z. Is that because you're testing with an extension PCI-E cable (like those meant for mining rigs)

Correct. I took the screenshots with the card running on a 1X PCIe slot mining riser.

And btw, a decent DX 9.0 card which is ATI HD 4870 1GB GDDR5 imho, cost about $3 but with a no display symptom, lol.
If I do a retro gaming, I'd snatch that up and repair it.

[momaka]

My pointless and worthless repair is a 5V 3A switched mode power supply adaptor.

What happened to it was a leaky main cap, and when I plugged, unplugged then quickly plugged back the adaptor, somehow that made the PWM chip blew itself up.

The chip is OB2358AP. It cost about the same as a brand new adaptor, which is around $1.

I replaced the chip and the main leaky cap.
It's been running fine since then.

Good save!

I don't think you can get a decent 5V 3 Amp power adapter like that for $1 new from the store.
The one shown above is built very well - it has an input choke, proper fuse and thermistor, an and proper X2 and Y2 caps.
All the cheapo adapters I can get new here (for more than $1, mind you) are gutless wonders with 1kV ceramic caps between primary and secondary, and a bunch of other cost-cutting measures.

I also have another pointless and worthless repair.
A GTX 560 SE with missing components and ripped pads. Being an old DX11 card with only 1GB GDDR5, it has zero value, but I repaired it anyway.

For someone as far behind as I am on the hardware curve, I'd say a GTX 560 (be it SE or not) is still a very decent video card.
What's funny is I find GTX 460 and 560 cards can now be found on Ebay for very cheap... yet try looking for a high-end DX 9.0 card, and the prices jump back up, sometimes to absurd levels. Not sure what the retro PC crowd is smoking these days, but a GTX560 is perfectly capable of handling any XP games at high res without the cost of an arm and a leg.
Also, many modern game requirements are stupid high nowadays for no reason. (Well, it's not for no reason... the reason is that coders / game devs are not willing to optimized anything for the PC, because it's expected that people that play games will always have good hardware. But Xbox and PS4/5, it's a different story.)

It's also running fine in one of my computers but I don't like the blower fan design, because it fails to cool down the VCORE mosfets.

The core appears to be running nice and cool, though - after all, 52C for GPU @ 100% load is not bad at all. I have the EVGA version on these cards (just regular GTX 560's) with the single fan in the middle, and those heatsinks designs suck even more. Sure my VRM MOSFETs might be getting cooled... but what good is that when the GPU core is running miserably hot?
No thanks, I'll take a refference blower -style cooler any day. It also removes heat from the system (PC case) rather than just re-circulate hot air inside the case.

BTW, you mentioned you fixed your card, but I see it's stuck in 1x PCI-E mode in GPU-Z. Is that because you're testing with an extension PCI-E cable (like those meant for mining rigs) or an error in GPU-Z? I have some dead/dying cards that also run only in 1x... or sometimes in 4x... but it's not due to bad ceramic caps near the PCI-E connector, I already checked. Most likely, it's a dying GPU in those cards.
That said, if yours is connected directly to the motherboard's PCI-E 16x slot and still running in 1x only, check the other "non-damaged" ceramic caps to make sure they are not shorted.
Whenever I see broken ceramic caps on a video card, I always assume it was not handled properly and thus any ceramic cap could be shorted. So for the ones on the PCI-E conn., I always check them all for short-circuit and proper capacitance. They should be either 100 nF or 220 nF.

[прямо]

My pointless and worthless repair is a 5V 3A switched mode power supply adaptor.

What happened to it was a leaky main cap, and when I plugged, unplugged then quickly plugged back the adaptor, somehow that made the PWM chip blew itself up.

The chip is OB2358AP. It cost about the same as a brand new adaptor, which is around $1.

I replaced the chip and the main leaky cap.
It's been running fine since then.

I also have another pointless and worthless repair.
A GTX 560 SE with missing components and ripped pads. Being an old DX11 card with only 1GB GDDR5, it has zero value, but I repaired it anyway.

It's also running fine in one of my computers but I don't like the blower fan design, because it fails to cool down the VCORE mosfets. One of these days I will probably 3D print a shroud for it, that will allow me to mount a standard 90mm graphics card fan which I salvaged from a dead RTX.

[aktinson]

Re: Post your worthless and/or pointless repairs

a lot of intertesting stuff tho here

[aktinson]

Re: Post your worthless and/or pointless repairs

Well, I once attempted a repair on my PS5's controller, even though it was still under warranty and I could have just gotten a replacement. The issue was that one of the trigger buttons (R2) was not registering properly when pressed down. I decided to open up the controller and see if I could fix it myself.

After spending about an hour trying to figure out how to disassemble the controller without breaking anything, I finally got to the trigger mechanism. I found that the issue was caused by a small piece of debris that had gotten lodged underneath the trigger. I removed the debris and reassembled the controller.

However, when I tested the controller, the R2 button still wasn't working properly. After spending another hour trying to troubleshoot the issue, I gave up and just got a replacement controller from Sony.

In the end, I wasted two hours of my time trying to fix something that I could have easily gotten replaced for free. Lesson learned: sometimes it's better to just take advantage of the warranty instead of trying to be a hero and fix things yourself.

[eccerr0r]

Re: Post your worthless and/or pointless repairs

I'd do that without second thought. Sort of reminds me when I modded my 32K HP48g to 256K, had to stack two 128k SMD chips and add wires to connect most of the pins together.

[jayjr1105]

Re: Post your worthless and/or pointless repairs

You can close the thread now, this is the most pointless repair in history, all this work for a $15 game on ebay.

[jiroy]

Re: Post your worthless and/or pointless repairs

Ugh!
Yeah, anything with roaches is where I draw the line.

Case in point: I found a MW oven with a blown HV fuse on the HV output (5000V, 0.5 Amp.) Cap checked out OK. Diode - not sure. Electronics and door latches - all OK. MW was dropped, though (probably when it was thrown away), so the door was on slightly crooked. No gaps, but you could tell the front "face" was not quite right. The plastic button area on the side had cracks and was missing a piece.


Can't go wrong with re-using good Lithium grease.
I used to do that a lot... but about 2 months ago, I just got tired of doing this all these years. So I went to Home Depot (hardware store chain here) and bought a big tube of Lithium grease for $5.

Yep , roaches actually done lot of damage in the mechanism area , so no prob.. I've always have like 50-60 different devices to decide upon , either full tear-down or repairing . Actually , it put lot of stress on me , lol .. Once something is arranged in some box or drawer , find it if you can

I'm working now on many projects including a solar project ; actually an experimental one , deep into solar cells what's and what'snot .
About grease , I won't change my habits ,

[momaka]

Re: Post your worthless and/or pointless repairs

This is or was a Sony x317 VHS system , given to me one year ago for free .
...
further on , three dead roashes was it ..

Ugh!
Yeah, anything with roaches is where I draw the line.

Case in point: I found a MW oven with a blown HV fuse on the HV output (5000V, 0.5 Amp.) Cap checked out OK. Diode - not sure. Electronics and door latches - all OK. MW was dropped, though (probably when it was thrown away), so the door was on slightly crooked. No gaps, but you could tell the front "face" was not quite right. The plastic button area on the side had cracks and was missing a piece. All in all, it was a fairly recent MW oven (2-3 years old, maybe), so seemed decent enough to fix up. However, the shape that it was in made me starting to doubt if I should. But once I got to the electronics / main PCB and found roach poop everywhere, then a few old roach eggs in the chassis - those were the final nails in the coffin. I stripped what useful parts it had and left it all outside in the freezing cold we had over Christmas. It was probably clear and free from roaches to work on afterwards... but at that point, I just didn't feel like putting more energy into it. If the fuse had blown due to a shorted magnetron, it surely wouldn't be worthwhile to fix. Perhaps I could have tried by shorting the fuse with a piece of thin wire... but again, I had drawn the line already and just scrapped it for parts. There are plenty of free MW ovens on CL here that another one was simply not needed. I already have a spare (that I also need to give away) and the following week also found another one that I donated shortly after (it was an old Samsung from the late 90's, and it worked just fine.)

By the way , I didn't forget to extract a good quantity of grease since this is best quality you can ever find and don't tell anyone I told you so

Can't go wrong with re-using good Lithium grease.
I used to do that a lot... but about 2 months ago, I just got tired of doing this all these years. So I went to Home Depot (hardware store chain here) and bought a big tube of Lithium grease for $5.

The ballast board has two IRF630 clones on it, yay more spare mosfets!

Glad to see I'm not the only one collecting dead bulbs for various projects.
CFLs seem to pack the best assortment of parts.
LEDs are kind of boring most of the time.
Incandescent are good for collecting the Tungsten. If you collect enough, it may be worth a good $ for scrap (or if building your own vacuum tubes?? )
I've also collected a few meant for recessed lighting. I've had this idea for a long time now to build a front light for my bike using the body of the dead recessed light bulb and some LEDs inside angled to make use of the reflector part. I suppose that would be more of a ghetto mod / creation than a worthless repair.

lol good job, alas I think I'd draw the line at that because I can't - with a no longer straight or accurate ruler...

Well, it's still accurate... so long as you use it for relative measures only (i.e. if you measure something to be 11" with this ruler, then you must use the same ruler when measuring something else. Otherwise, you'll get a disagreement.)

I also have a (soft) tape measure that suffers from this - the plastic has stretched, making the first 4-8 inches (10-20 cm) a little longer and more "stretched" than what real inches / cm should be. I used it like this for a long time without knowing when cutting various small lumber pieces for a few projects, and still didn't get more than a mm or so error... so I guess it's still somewhat useful. Now that I know, though, I don't use it for stuff where I need really good accuracy. In some cases, I can get up to 2 mm difference, which can be very annoying and not work right in some cases.

I don't like spending money

Nice one foxer!

[eccerr0r]

Re: Post your worthless and/or pointless repairs

lol good job, alas I think I'd draw the line at that because I can't - with a no longer straight or accurate ruler...

[foxer]

Re: Post your worthless and/or pointless repairs

I don't like spending money

[eccerr0r]

Re: Post your worthless and/or pointless repairs

Well, a salvage:
I had a 3 way 50-100-150W-equivalent CFL lamp. The tube had a weak spot on it and cracked

I opened it and took the ballast board out, rest of it probably goes to trash, I think all the mercury is gone after leaving it outside for a week.

The ballast board has two IRF630 clones on it, yay more spare mosfets!

[eccerr0r]

Re: Wooden Pencil Repair

OK, this one is self-explanatory.

LOL Yeah nothing should be wasted, I remember when "pencil fighting" was a "thing" (am I that old or are kids still doing this) and the pencils that lose usually end in the trash.

I always thought it was a waste of a perfectly good tool...

[jiroy]

Re: Post your worthless and/or pointless repairs

This is or was a Sony x317 VHS system , given to me one year ago for free . They told me it's intermittently working , and gradually became frustrating to deal with . When I had a chance finally to take a thorough look into it , I made many mistakes as taking the whole thing lightly . I've already have a good system for converting into DVDs and so , so it was a light repair to begin with.. First mistake was tryin'it with an important tape for me , although it worked well for 15 minutes or so . Ejecting and inserting repeatedly only worsened the situation and at that point , it was put aside for later disassembly .

When I opened it some weeks later , I took my first measurements on the power supply unit under load then some measurements on main-board and things turned out very normal . My thoughts gradually went into the mechanical system being the one causing the real problems .
And here came my Second mistake when I broke the soft 4 plastic wires from the read head to the main board . Although it could be an easy repair , but further disappointments were when I disassembled the mechanical systems and more weary belts and clutches started to put on heavily on my thoughts .
further on , three dead roashes was it ..

So I decided to change my priorities to salvage my precious tape and extract the magnets (I love magnets ) , dynamos and any other things useful for my projects . Next , it was a full tear down to the last screw and ring , testing my old times speed and abilities at it , lol ..
By the way , I didn't forget to extract a good quantity of grease since this is best quality you can ever find and don't tell anyone I told you so

[momaka]

Wooden Pencil Repair

OK, this one is self-explanatory.




Seriously, how much is a simple wooden pencil worth these days? A few cents, maybe? Does anyone even care to use wooden pencils anymore? My guess is, probably not. But even 2-3 decades ago when I was still in school, I remember no one cared about them back then either.

In the case of the above pencil… wind back about 15 years to my high-school days. Some kid was just goofing around and took another kid’s pencil, then broke it in half and threw it around in the classroom. No one bat an eyelash, of course. Inner me: grrr… must not waste! Must fix!. So of course I had to I pick it up… and I guess I’ve always been a master hoarder after all.

Anyways, since I picked up all of the pieces, I was able to glue it back together with wood glue in one of the engineering elective classes I was taking. The repair lasted me through the remaining two years of high school. Then the pencil broke again (I suspect the classroom wood glue was a little “expired”, as many times people forgot to put the caps back on – another grrr to trigger my OCD. ). Regardless, I glued it again (this time with a slightly different/better wood glue of my own.) And it’s been fine since then. Believe it or not, the pencil was actually almost twice as long as seen in the above picture. I’ve just used it for many years afterwards. In fact, it didn’t even dawn on me how worthless of a repair this was until I picked it up again the other day to write something down. It needed sharpening, and that’s when I noticed I’m about to grind away at the broken part soon.

I guess it’s because I consider a pencil to be a tool – one that let’s you write stuff down… and I’ve always had respect for any kind of tools.

So there’s that worthless repair.

On the up side, I NEVER had to ask my parents to buy me pencils or pens for school. I was always loaded with all kinds. In fact, many kids knew me as “that kid” who they could regularly “borrow” a pencil from. And if/when they forgot to return it – no worries! Sooner or later, the same pencil would find its way back to me / my backpack again.

[eccerr0r]

Re: Post your worthless and/or pointless repairs

Worthless repair:

I noticed one of my (working) 40W-equivalent LED A19 bulbs had a loose bottom end (this is one of the older ones where the heatsink and plastic base were distinct pieces. So I twisted and pulled it to see why it was loose .. and broke the bottom contact connection (it's one of those press-in-place "tacks"). I broke it, it was working before now it doesn't!

So I twisted and pulled it more to try to pull it apart. No dice. So I took a screwdriver and pulled off the "ice cream" cap in this ice cream cone lamp. Pried it off, appears to use RTV to glue the cap together. Then I had to desolder the PCB from the wires and removed the metal LED PCB from the heatsink, finally able to separate the bottom half with the PSU and the LED PCB.

Now I could find the wire that should have connected to the bottom contact an reconnected it. Then I put everything back together as much as I could (no RTV, probably will have to use epoxy but didn't glue everything together yet).

Off to testing. So what do I do to test this to ensure it doesn't go up in flames?

Now I need to get a dim bulb tester to test this light bulb!



Sorry no pictures... should have taken some...

[jiroy]

Re: Dell 65-Watt power adapter cable repair

Does anyone bother to fix laptop power adapters anymore? I know replacements are pretty easy to find and often quite cheap too – even the OEM ones. Nevertheless, I hate waste, so I still “fix” (cobble / hack / McGyver) these when possible.


Again, I figured I'd post this here, since it's probably a “worthless” repair in a sense that it takes relatively long time to do compared to just going out and buying a new / another power adapter. Doing the failure & engineering part was the fun for me, though. I've seen many such power adapters fail and also fixed a good deal of them in the past… just never put so much thought into it before as I did now.

For this kind of repairs , I always solder the wires by their ends , so they take their original shape , then use laptops or batteries pvc bodies with a solder tip to reconstruct a like original shape . A fine sand paper and there it goes .

[momaka]

Dell PA-10 90-Watt power adapter cable repair

And here is “worthless” adapter repair #2.(Note: a lot of text but not many pictures.)

It’s the Dell power adapter for my uncle’s E6400 laptop. As mentioned in this post of the ghettomod thread and also here, this power adapter had a broken / open-circuited ID wire. Essentially the laptop would power on and work, but only with the CPU in the lowest power state and the battery would not charge. Running the built-in Dell Diagnostics from BIOS showed the power adapter as “unknown”.

Once again, I forgot to take pictures of the adapter before I started fixing it, but it’s a standard PA-10 adapter (the flat and wide ones with a rubber strap on the side, if that rings a bell for anyone.)

Knowing that the adapter worked but the ID didn’t appear in Dell Self-Diagnostics (“unknown adapter” instead of “90 Watt”), I cut a small 15 mm incision in the output cable, near where it goes into the power adapter. To my surprise, this one did not use a coaxial cable construction design. Rather, the output cable had 3 wires in it: black, white, and green. Using a thin sewing needle, I poked through the insulation of one of these wires and then used my multimeter on resistance mode to check which connection this corresponded to on the barrel plug. I found the following:
Black --> ground
White --> positive (+) DC supply (19.5V)
Green --> laptop ID wire.

But when probing the green wire to the small middle pin inside the barrel plug, I got nothing at all – open-circuit. So I made another small 15 mm incision on the output cable, this one closer to the barrel plug. Checking the resistance on the green wire between there and the middle pin of the barrel plug still showed nothing. I was a little puzzled. Did the green wire not connect to the ID pin on the barrel plug? To resolve my confusion, I did another test. This time using two needles, I checked the resistance on the green wire between the two incision cuts on the cable… and still got nothing (open-circuit). Well, that surely didn’t make sense – why would the green wire not connect to itself?

That’s when I decided to make a 3rd cut / incision in the cable, this time towards the middle. Probing between that spot and the cut close to the power adapter, I did get continuity / low resistance on the green wire. But between the middle incision and the one closer to the barrel plug, I still saw open-circuit. This meant only one thing: the green ID wire had a break in 2 places!

At this point, I knew I had to cut the cable sleeve completely open on the barrel plug side. And so I did… which revealed that the green ID wire was indeed broken off from the barrel plug, hence the 2nd break in the cable as I suspected.

I considered running an external wire wrapped around the output cable to bypass all of the possibly broken spots in the ID wire. I widened the incision that was in the middle of the cable so that I could get better access to the green ID wire and try to solder/attach a bypass wire on the outside. However, this didn’t work out too well. As soon as I tried to stretch / pull out the green wire even a little bit out of the incision, it broke and ripped a little further downstream. great! :\

Not sure why, but the green ID wire used in this cable was quite fragile. Trying to pull it out from an incision further down (closer to the adapter) made it break again. Thus it was clear that I should just replace it entirely… and for that, I also needed to open the power adapter too. For a second, I even thought about replacing the entire output cable… except I didn’t have another one with the same barrel connector. So something else came to my mind instead – try to pull out the green wire out of the existing cable and replace it with another wire inside.

I tried pulling the green wire from the middle incision and it did pull a little, but then became a little too difficult. More thinking, and I made two more incisions on the output cable, with these spaced between the other three. So with incision sections much closer to each other, I was able to pull the green wire section by section out of the output cable – almost like replacing cable/wire in a conduit or wall… but just on a smaller scale.

Once I did that, I was wondering what to replace it with. I had some very thin gauge PVC sleeved wire…. But it wouldn’t slip easily in the output cable in place of the green wire. Then I tried a single solid-core Ethernet strand and that was a little easier… up until the 2nd incision. After this, it too became difficult to pull further. So I removed it and tried a 23 AWG enameled copper wire (magnet wire.) This time, it pulled through nice and easy.

I soldered it inside the adapter, as seen here (lower left corner of the PCB):


Next, time for the barrel plug end. Just like the previous power adapter repair, I wrapped the three output wires around the barrel plug and then soldered them to it.


Before proceeding to pot everything in hot glue, I connected the adapter to the laptop and gave it a thorough test, This time, the ID of the adapter (Dell 90 Watt) was identified correctly in the E6400. With this out of the way, next step was to pot everything in hot glue and add some heat shrink tubing.


Now remember the discussion in my previous post above about how it might matter which way the cables come out of the barrel plug? Well, this one I also did in a slightly more different manner: I angled the wires at 45° relative to the barrel plug. I figured / noticed that with the previous repair (where the wires come out at 90° from the plug), the cable would sometimes get stuck down when placed on a table and exert extra force on both the wires and the laptop’s plug. So with a 45° angle, this seemed like the perfect compromise between the 90° and straight-out design. When laptop was placed on a surface (from standing in the air), the 45° angle on the wires would naturally allow the barrel connector to rotate inside the laptop plug by itself. Also, when the power adapter cable was tugged hard from the laptop, the barrel connector would pop out much easier compared to the 90° design. And if the laptop was tilted on a flat surface, the 45° design still allowed plenty of tilt (well, 45° really) before the flat surface would cause the wires to bend or exert extra force on the barrel plug. So the 45° design seemed like the best compromise of them all.

Finally, like a true live patient, it was time to close up the power adapter and incisions in the cable. The power adapter case was easy to – just add a few spots of rubber cement glue here and there, and it shut back nicely. I probably could have used PVC glue / PVC cement too, since the adapter’s plastic did contain ABS. But the rubber cement held it down pretty well and would be slightly easier to open, should I ever need to go back in there again. As for the output cable… since it is actually made out of PVC, I certainly used PVC cement for that. Basically, I applied a small drop to each incision and then clipped it together with small clips.


After a few days of R&R (for all of the glue to cure), the adapter looked pretty normal (save for the hot glue on the barrel connector.)


I even considered painting the hot glue with black to make it match… but decided to leave it as-is, so perhaps I could see easier if a break in the cable occurred again on the barrel end.

In terms of longevity, this repair didn’t do too bad. Done in December of 2021 and checked back in the summer (June) of 2022, everything was still mostly holding up. A crack in the hot glue on the wire right before the barrel connector had started to form. But my fix was partially to blame for this. Remember how I said not to put too much heat shrink and leave some of the wires exposed for better flexibility? Well, I didn’t do that here. And with a choke / ferrite bead very close to the barrel plug, it’s obvious how this contributed to the early damage. On the positive side, one of my nephews is now old enough, has a hot glue gun, and loves to use it. So we re-melted the glue with him and I gave him the task to keep an eye on that power adapter. After all, he does go to my uncle’s house quite often.

So there it goes… another one saved from the landfill for a little while longer.