Turbo -> wrote on 2022-04-23, 12:10:

I've made a bracket - stand; I don't know how else would I name it, to prevent my graphics card from bending too much, due to quite heavy passive heatsink. It was time-consuming, but I like the result in my Windows XP build.

I kinda like this solution 🙂
I'd be a little bit more worried about it having sharp edges though. If it has any I'd definitely file those down if only to prevent it cutting any wires (and your flesh 😋).

PcBytes wrote on 2022-04-23, 14:28:

I didn't desolder it. Just followed @Robert B's guide from here: Re: Hello, World! :D - Robert B's PC builds - oogle away freely […]
Show full quote

Kahenraz wrote on 2022-04-23, 14:13:

Well done. How hard was it to desolder the socket?

I didn't desolder it. Just followed @Robert B's guide from here:
Re: Hello, World! :D - Robert B's PC builds - oogle away freely :)
In my case though, the socket proved a bit tighter than the original ECS socket after installing.
Currently has a Duron 1100 (I think?) and 256MB of RAM in it as I plan on doing a ceramic AMD build for once (Morgan core btw!), hopefully the ever so slightly chipped core won't cause me horrible issues.

All in all, this could be a positive "warning" 🤣 - things will get crazier by each day for me, now that I'm fully recovered after a rather severe enterocolitis. Anything goes, from replacing Skt462 sockets to doing Tualeron mods on the cheap.

I knew the top part of the CPU socket could be removed (I repaired a couple boards with broken levers this way), but the thought never had occurred to me that the bottom part of the CPU socket wasn't, like, glued to the PCB or something. I'm definitely gonna remember this one 😀

PcBytes wrote on 2022-04-23, 14:28:

I didn't desolder it. Just followed @Robert B's guide from here:
Re: Hello, World! :D - Robert B's PC builds - oogle away freely :)

Wow, that's wild! Never imagined sockets (or at least the plastic housing) was removable in that fashion.

I'd be terrified of damaging the pins in the process. Judging from their own experience, it looks like they had that happen.

Today started working on Korg 05R/W module number 2 of 3. This unit has no audio output. During one of my audio tests, I did get extremely faint audio for a brief moment, but then it ceased.

This one had a different board revision (KLM-924) than the other units I have (KLM-924A AND KLM-924B).

This particular board also has an odd bodge-diode thing going on. On the subsequent revision board, it looks like they added a transistor (ZD2) near the same location. I'm assuming this may be a factory job if it was to correct some sort of issue with the original board design. Unless this was some sort of aftermarket repair attempt?

Things tested / worked on so far:

• Tested volume pot; seems to be reading okay.
• Replaced through-hole capacitors; original caps were a bit under capacity, but still within spec, so probably not an issue.
• Checked and didn't find any broken solder joints; reflowed the 1/4" output jacks just in case.

The board has minor corrosion in places, but haven't found any broken traces (yet).

Other than a moment of brief extremely faint audio, haven't gotten proper audio output. I'm going to keep checking traces and for shorts. I'll also probably start removing the aluminum SMD caps and testing those. I'm also wondering if it might be worth doing a general reflow on the SMD components to make sure everything has proper joints.

One last item, I did have an original Korg adapter for this unit. It's supposed to output -12V, but it's outputting a whopping -17.5V. Good thing I haven't been using it!

The voltage may be high because there is no load being applied. This is very common on linear supplies, and older ones tend to have a pretty generous range due to the poor tolerances of cheap supplies of the time.

Shponglefan wrote on 2022-04-23, 18:38:

PcBytes wrote on 2022-04-23, 14:28:

I didn't desolder it. Just followed @Robert B's guide from here:
Re: Hello, World! :D - Robert B's PC builds - oogle away freely :)

Wow, that's wild! Never imagined sockets (or at least the plastic housing) was removable in that fashion.

I'd be terrified of damaging the pins in the process. Judging from their own experience, it looks like they had that happen.

On mine, I simply had enough patience and sharp eyes to make sure I wasn't going to bend any pins. Once I saw all of them fit snuggly from the first try, I didn't think further. Pushed the socket evenly until the pins were in their place, then placed the lever and socket cover back.

The worst part was getting a socket out from a donor board. I was initially planning on using a green socket from a scrap GA-7VAX. No dice, that thing felt like it was glued to the board, and even snapped a little bit once I got it out (though it's fixable, from what I've seen, but I wouldn't trust it to be put on a mobo after I'd repair that socket.) Same goes for a yellow socket from a GA-7N400 (which I really didn't mind killing - it was the worst version of the board, without onboard NIC) that literally broke not long after I lifted one of the sides. The only one that seemed to come out absolutely easy was the one from the 7VT600, which was similar to the one that was on the K7VZA originally.

Onwards with restoring the K7VZA, I did a proper recap. Off went OSTs, mixed and matched caps on the higher VRM side (you can tell from my original socket photo) and a bunch of green G-Luxons of 1000uF 6.3v values each, and in went a fresh batch of Rubycon MBZ 3300uF 6.3v for the CPU filtering, and Sanyo WG 1000uF 6.3v for replacing the G-Luxon caps.

This is also a great trick for removing the old socket as well. It's a lot easier to remove one pin at a time than to try and remove the whole socket set once, since invariably there will be a number of joints that won't release cleanly without some mechanical force.

Adventures in board repair continues as I removed various SMD aluminum caps.

Unfortunately lost a couple pads in the process, so I've now ordered some parts to fix those.

There was a mess underneath a number of the caps. Wish I'd taken a 'before' pic as this pic was after doing an initial cleaning.

A number of the solder joints are heavily oxidized, so I'm now going to have to scrape off the surfaces and then tin the pads. A couple of the caps came right off with almost no heat, so I'm wondering if those joints weren't making effective contact anymore.

Kahenraz wrote on 2022-04-23, 20:08:

The voltage may be high because there is no load being applied. This is very common on linear supplies, and older ones tend to have a pretty generous range due to the poor tolerances of cheap supplies of the time.

That's good to know. Though in the mean time, I'll probably just keep using my multivolt adapter. It seems to put out a steady -12.2 V. 😀

The juice from capacitors will destroy few painted on vias. Make sure you check for these too. I had this happen on one TV with PiP module of a model I cannot bypass with pair of capacitors so got PiP module (they are stand up small board via plug in connectors) rebuilt and found few painted on vias destroyed so they get drilled open and piece of wire inserted thru and soldered both sides, as via substitute.

Cheers,

pentiumspeed wrote on 2022-04-24, 00:45:

The juice from capacitors will destroy few painted on vias. Make sure you check for these too. I had this happen on one TV with PiP module of a model I cannot bypass with pair of capacitors so got PiP module (they are stand up small board via plug in connectors) rebuilt and found few painted on vias destroyed so they get drilled open and piece of wire inserted thru and soldered both sides, as via substitute.

Cheers,

That's a good point. Some of them were covered/plugged up with gunk, which has been partially cleaned off now. I'll have to check for continuity to see if any are broken.

Not a PC thing, but I'm not currently hanging on any specialist arcade forums, so y'all can hear it.

Ordered a set of arcade controls for a new cabinet I plan to build. I'm excited because the buttons are Crown 202's, which come with a reputation for excellence that I have not yet experienced. Better be!, they're not cheap at just under $5 a hit. And the main event, the joystick, is my first Seimitsu. A classic LS-32. Gonna be great for playing a range of game styles via the MiSTer which will be powering the eventual cab. I've done plenty of pre- research about what to expect, but I look forward to getting the real life feel for how they compare to my existing cab's Sanwa + Suzo Happ parts.

Lots of 8- and 16- (+ the odd 32- ) bit games will be played. LOTS.

I've recapped two retro gaming consoles: My friend's NES and one of my Slim PSones. The NES had the blinking red light, even with a new 72-pin connector (I would've used a Blinking Light Win, but they haven't been in stock for several months due to a manufacturing issue), so I had to disable the NES10 lockout chip. The reason why I had to recap it was that the image had some static over it which was typically a sign of failing capacitors on the power module. Removing the module from the mainboard was a huge pain because of how well it was soldered on, I had to resort to using the hot air gun and a solder sucker to separate it. I accidentally sucked up an SMD resistor in the process, but luckily I was able to find it and solder it back in place. Now it works as good as new.

The PSone was one I bought on eBay a couple years ago, the listing stated that it worked, but had an extremely dim picture. The last two PSone consoles that I bought before this one had optical drive controller issues (they wouldn't read game discs, even with a new optical drive), so I was happy to finally get one that could read discs properly. It's been sitting in a box for the last couple years waiting to be fixed, and I've manage to track the problem down to the two capacitors near the Multi AV output socket. After replacing the two problematic capacitors, the image quality looks perfect now on every output method (Composite, S-Video, and RGB).

I've also RGB modded two Sony Trinitron TVs (a KV-20FV10 from January 2000, and a KV-27V25 from July 1996) for a friend who fixed my KV-27V20, which had serious convergence issues that I couldn't fix myself. For the KV-27V25, I've also done a Component (YPbPr) mod just like I did to my KV-27V20, since he asked for both RGB and Component for this TV.

Not today but a couple weeks ago, I finally got around to testing the Voodoo 5500 card I bought years ago, works great! Currently running it in a period correct 2000 era system with a PIII 1.0 GHz. Gave it a quick run through Need For Speed II SE, one of my favorite 3dfx titles from back in the day. Man it's cool to see it again, and I remember the feeling of seeing Glide for the first time. Great memories!

Kahenraz wrote on 2022-04-23, 14:13:

Well done. How hard was it to desolder the socket?

if you have a decent de-soldering gun, its not difficult, just tedious. (i've done several swaps)

Shponglefan wrote on 2022-04-24, 00:39:

Adventures in board repair continues as I removed various SMD aluminum caps. […]
Show full quote

Adventures in board repair continues as I removed various SMD aluminum caps.

Unfortunately lost a couple pads in the process, so I've now ordered some parts to fix those.

There was a mess underneath a number of the caps. Wish I'd taken a 'before' pic as this pic was after doing an initial cleaning.

A number of the solder joints are heavily oxidized, so I'm now going to have to scrape off the surfaces and then tin the pads. A couple of the caps came right off with almost no heat, so I'm wondering if those joints weren't making effective contact anymore.

How are missing pads repaired? I always thought that is next to impossible?

It's possible to rebuild pads, but it's very difficult to do.

I let my creativity do something for today.

2GB stick that was advertised as 4GB.

Left is what I did to the stick, right is the original shell which now houses a (real) 8GB TDK pendrive.

I basically had two BIOS chips that were broken, so I just went ahead and used them as a shell for the 2GB pendrive. Works just as intended, might use it for retro drivers.

^it's nuts, but and therefore I like it!

The world's first dual-BIOS USB stick??

I solved a few problems on my Mitsubishi 386, and identified a couple more. Two steps forward, one step back.

I am constantly finding new problems to solve.

And it's glorious. 😀

HanSolo wrote on 2022-04-24, 10:44:

How are missing pads repaired? I always thought that is next to impossible?

They can be replaced and bridged to the adjacent solder traces.

While I haven't attempted this type of repair before, I've ordered some materials and will give it a shot.