Okay... First off, I don't have one of these cards on hand. I thought I'd offer some suggestions and maybe someone with more knowledge can chime in.

The specific missing parts seem rather peculiar. Looking at the board view of the CT4390, was there damage, or was this card used for parts at some point? To confirm, does this pic accurately represent what's missing?

If so cool! Finding the actual schematic for this card might be a lost cause. However, it might not be a super hard task to figure out what those parts were...

Let's start with C52.
That's connected to an ST-TL074 quad JFET opamp. From the pic it looked like C52 is a simple filter cap on the negative voltage rail (VCC-) on pin 11. This should be an identical part to C70 on the opposite side connected to the positive voltage rail (VCC+). From the size I'm guessing it's a 0603 ceramic capacitor between 0.1uf and 1uf. This is just a rail voltage filter, but should be equivalent to the C70 cap. A 6.3V voltage rating would probably be fine there. So maybe this 0.1uf 6.3v -->https://www.digikey.com/en/products/detail/ke … RAC7867/2199778

C66 and C67
These are signal filters in line with the outputs of the TDA1387 DAC. The datasheet doesn't specify a common part here, but the schematic examples on a Google search show this DC blocker to be somewhere between 10uf-22uf. So I'd say this looks like a 1206 ceramic capacitor >6.3v. This would probably be a suitable replacement. --> https://www.digikey.com/en/products/detail/sa … KOHNNNE/3886740

C27 and C28
Not having the board with a meter to trace the path, these are most likely a similarly used DC blocker for the nearby ST33079 quad opamp. You could check to be sure and see if either end of the pads is grounded which would make them rail filter caps, but I'm leaning on the same 1206 >6.3v part as C66 and C67.

C89
Guessing, but looks like another rail or ground filter cap. Meaning it might simply be the positive 5v rail supply for the chip. My guess is based on the fact that on the right side of the 0603 cap, two traces are combined on the IC. So it looks like the right side is fed through a via to a power plane and the left side is grounded to a ground plane. This is probably similar to C52 in this regard. 0603 6.3v (0.1uf-1uf) so the same part would be used here as C52.

EDIT: If C89 is a rail filter cap on the CT-8908 IC, a cap larger than 0.1uf might be better here. So maybe this 1uf part. --> https://www.digikey.com/en/products/detail/ya … R7BB105/2833611

C24
This... is coming from what I assume is the analog half of the CT1745A mixer chip. I couldn't find a datasheet for this and that's not surprising! This is probably very similar or identical to C66 and C67. Or, judging from its proximity to the other (line in or out) on that side of the mixer IC, it's most likely an identical part to C120,C121,C122,C125. These are all performing the same filter function and you could possibly measure the capacitance value of these nearby parts and order something similar to the part for C66 and C67 but with a closer capacitance value if it's different. --> https://www.digikey.com/en/products/detail/sa … KOHNNNE/3886740

L3
This is a guess, but most likely the identical part to L1 and L2. I'll guess that these are very low inductance or ferrite beads that are used to traverse power across the segmented power planes. I'm kind of going out on a limb here so someone can correct me if I'm way off! The trace widths compared to the others nearby tell me this is most likely a positive voltage rail for the IC and is acting like an inline filter. You can actually see if you look carefully that the L1, L2, and L3 span a copper plane that's internal to the PCB. It's very clear on L1 as you can see through the PCB itself. This demarcation goes down and turns right to travel under L2 as well. If you look closely, L3 also has a gap in the internal copper layers just under its left side.

These will be a specific low impedance rating for a specific frequency that the engineers wanted to filter out. Swapping in a random value might not cause any real issues. Even a wire bridge might actually work here. I'd ask you to remove L1 possibly and measure its properties, but you'd need something better than a simple meter to figure out its sub-ohm resistance value. That being said, something like this 1812 chip Ferrite Bead would probably work fine. --> https://www.digikey.com/en/products/detail/la … T800R-10/806659

I'm completely guessing on the actual part sizes since I'm doing this all from a picture.

Hopefully, someone can chime in and find fault with my suggestions. Good luck to you!

A really appreciate the effort you put in this. I found the card at a recycling center. Yes, the missing parts are the ones you encircled.
Once again: Thank you very much. Will keep you posted with my progress.

Bakersor wrote on 2024-06-13, 10:03:

A really appreciate the effort you put in this. I found the card at a recycling center. Yes, the missing parts are the ones you encircled.
Once again: Thank you very much. Will keep you posted with my progress.

Awesome! I hope that helps. Can you post a pic of the damaged card?

Mm... I wonder if this card has other issues than just the missing caps. Are you able to power the card outside of a computer to see if it has possible shorts?

Hope the image is clear enough. Haven't tested it yet. Also, the 4 pins in the top right corner are bent but luckily I see no traces for them.
Edit: I will just straighten the pins with a sharp cutter and make sure they don't short

Well, congrats on finding the card. It's definitely worth fixing. 😎

You should go over very carefully with a magnifying glass and inspect all the QFN chips. The pic isn't super clear but there might be some bent pins around the perimeter! Here's a few areas that looks sus.

Sorry, doing this on my phone on my way to work.

This board obviously had some physical impact to pop these parts off. You should also look very closely for damaged traces like this first pic. Surface scratches can easily be deep enough to cut a trace! Also, how's the back side? I'd be inspecting things close for cut traces!

About testing it, that might not be good in it's current state. Especially with L3 missing. This is most likely the 5v rail for the digital processors on the left side of the board. With the power removed, those chips might try to sync current through the IO pins. This will result in very unpredictable behavior and I'm guessing it won't even be recognized by the PC. Might not damage things, but I'd be very cautious about powering up in this state.

So, fix the pins, check and double check for cut or shorted traces, try to provide power outside of computer to check for board shorts, replace to popped parts, power check again outside of computer for shorts, then try to install in a machine.

Post again with updates. Curious if you can make it work. 😊

Will keep you posted. Will take a bit before I focus my attention on this one.

Thank you so much @Tevian

After ordering the components you suggested and months of life keeping me busy, I finally managed to get myself some time to work on the sound card.

I started by fixing the broken traces (first photo you shared in your last post) and I am proud of my work. I moved to mounting the parts.
Sadly, the chip pins were bashed in (second photo) and the first one broke. Luckily, none of them were connected to any trace as far as I can see and so I continued working assuming they were ground. Out of the remaining 3 that were bashed, i re-soldered 2 because the last one had its pad missing/bashed.

Tested it on DOOM and Day of the Tentacle and managed to get sound.

Once again, thank you so much!!!

Awesome! Ya, things looked rough and I'm glad it's working for you now.

👍😋