VOGONS


Reply 361 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++
root42 wrote on 2020-06-30, 12:53:

The PCM sounds terrible. Broken DAC? The clicking on the left channel somehow sounds a bit like an echo of some OPL sounds.

I would say the TEA2025 is fine. The OPL sound is almost clear, apart from the clicking. So I would say that one is working.

No it is not a broken DAC unless both DACs I have are similarly faulty. Messing with the U5 Op-Amp brought PCM audio back albeit distorted so I would think that (or maybe its socket?) is the culprit possibly in the antialiasing circuit? Maybe I am killing them while soldering the bodge resistor? I have one extra I will try it without the bodge resistor..

EDIT: I tried a fresh LM324 without the resistor soldered on, it made no difference.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 362 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++

I recorded OPL2 music output using Monkey Island 2: https://drive.google.com/drive/folders/1AWtnF … vqX?usp=sharing

Notice the weird sounds when starting up and quitting the game, as well as the clipping/clicking/crackling (more noticable on the left).

PCM audio is still distorted, haven't found any faults in traces/joints so far. I'm thinking one of the logic ICs is probably faulty, but none of them gets hot or anything..

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 363 of 495, by Tronix

User metadata
Rank Member
Rank
Member

This reminded me of my trouble with Snark Barker, for the solution of which I needed to make an ISA8 bus emulator -) In my case, the problem was the broken buffer 74LS245 IC.
Is the card normally detected? Does it display the DSP version correctly? I will attach several utilities in the archive that can detect and display the DSP version of SB. Perhaps this will help with something, help to understand whether the computer<-> SB card connection is working normally.

Attachments

  • Filename
    SB_check.zip
    File size
    88.86 KiB
    Downloads
    47 downloads
    File license
    Public domain

https://github.com/Tronix286/

Reply 365 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++
Tronix wrote on 2020-06-30, 19:03:

This reminded me of my trouble with Snark Barker, for the solution of which I needed to make an ISA8 bus emulator -) In my case, the problem was the broken buffer 74LS245 IC.
Is the card normally detected? Does it display the DSP version correctly? I will attach several utilities in the archive that can detect and display the DSP version of SB. Perhaps this will help with something, help to understand whether the computer<-> SB card connection is working normally.

I could not possibly do these things you did, but fortunately I may not need to:

Tubetime-Snark-Barker-SBCHECK.jpg

So I get a DSP version from SBCHECK. However, I am starting to get the suspicion that one of my BUS logic ICs is faulty as well. I just have no idea which, I may as well order a spare of each and go around replacing them all until I find one..

Cyrix200+ wrote on 2020-06-30, 19:37:

CMS is the same quality as OPL2 music? Also with more distortion on the left?

I added a recording of Monkey Island in CMS mode: https://drive.google.com/drive/folders/1AWtnF … vqX?usp=sharing

As far as I can tell, yes the distortion is there.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 366 of 495, by Cyrix200+

User metadata
Rank Oldbie
Rank
Oldbie
appiah4 wrote on 2020-06-30, 20:06:
<snip> […]
Show full quote

<snip>

Cyrix200+ wrote on 2020-06-30, 19:37:

CMS is the same quality as OPL2 music? Also with more distortion on the left?

I added a recording of Monkey Island in CMS mode: https://drive.google.com/drive/folders/1AWtnF … vqX?usp=sharing

As far as I can tell, yes the distortion is there.

Actually, sounds just like mine. It does not have the crackles like the OPL2 or the distortion of the PCM audio.

1982 to 2001

Reply 367 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++
Cyrix200+ wrote on 2020-06-30, 20:23:

Actually, sounds just like mine. It does not have the crackles like the OPL2 or the distortion of the PCM audio.

Well.. I'm not familiar with how CMS sounds, I guess it just sounds metallic/distorted due to being limited to square waves or something? Regardless, I'm looking at the schematics and I can't figure out what is common to the other two circuits that is not involved in CMS..

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 368 of 495, by root42

User metadata
Rank l33t
Rank
l33t
appiah4 wrote on 2020-06-30, 20:40:
Cyrix200+ wrote on 2020-06-30, 20:23:

Actually, sounds just like mine. It does not have the crackles like the OPL2 or the distortion of the PCM audio.

Well.. I'm not familiar with how CMS sounds, I guess it just sounds metallic/distorted due to being limited to square waves or something? Regardless, I'm looking at the schematics and I can't figure out what is common to the other two circuits that is not involved in CMS..

Check out my SAA1099 upgrade video for CMS samples:

https://www.youtube.com/watch?v=hwY-G7Tutrc

YouTube and Bonus
80486DX@33 MHz, 16 MiB RAM, Tseng ET4000 1 MiB, SnarkBarker & GUSar Lite, PC MIDI Card+X2+SC55+MT32, OSSC

Reply 369 of 495, by Cyrix200+

User metadata
Rank Oldbie
Rank
Oldbie
appiah4 wrote on 2020-06-30, 20:40:
Cyrix200+ wrote on 2020-06-30, 20:23:

Actually, sounds just like mine. It does not have the crackles like the OPL2 or the distortion of the PCM audio.

Well.. I'm not familiar with how CMS sounds, I guess it just sounds metallic/distorted due to being limited to square waves or something? Regardless, I'm looking at the schematics and I can't figure out what is common to the other two circuits that is not involved in CMS..

I pulled out my notes from my own assembly and troubleshooting when I had my TEA2025 issue, but I can't see it either. Maybe R25/R22 and R28/R31?

OPL2 sounds better that the PCM, that's another thing I think is interesting. I am out of my league here though 😜

You changed U 11 right? EDIT: yes you did 😀

1982 to 2001

Reply 370 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++
Cyrix200+ wrote on 2020-06-30, 20:59:

You changed U 11 right? EDIT: yes you did 😀

I just tried this one more time, no difference. Granted, the ICs I am using are TI MC3403N instead of NJM3403A but I don't think it should make a difference, they seem similar enough electrically and I've seen photos of original Sound Blaster cards with the former:

http://www.vgmpf.com/Wiki/images/3/38/Sound_B … ut_SAA1099s.jpg

http://www.yjfy.com/images/oldhard/sound/CT-1320C.jpg

Also, I just listened and there IS a constant static noise in the background of the left channel of Monkey Island CMS recording, it's very evident right at the beginning but once you notice you can hear it throughout.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 371 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++

The only stereo source of music on the card is CMS, so I tried switching the SAA1099P chips around, then removing them completely, but it did not fix the left channel noise. There is no other stereo audio going into the mixer so I guess that can only mean the noise is caused by aa fault at the Mixer/Volume Control/Amplifier/EMI_Filter stages. I will refresh the solder joints on all components/sockets on these stages tonight.

Failing that, I am also considering removing the potentiometer and temporarily replacing it with fixed resistances. It might also be the cause of the noise as well, I suppose?

EDIT1: I resoldered R30, R22, R31, R3, C37, C14, R2, C13, C12, C9, C10, C11, L1 and S1 on the Mixer/Amp circuits. This resulted in no change.

EDIT2: Replacing the potentiometer with another fresh potentiometer yielded no change. Another dead end.

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 372 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++

Replacing U2, U8 and U16 with known working parts from an Adlib card resulted in elimination of crackling/noise from the left channel!

Now all that is left to do is fix the PCM distortion.. Could the distortion sound be the result of a broken antialiasing filter stage or extreme clipping due to too high volume at the mixer stage?

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 373 of 495, by Cyrix200+

User metadata
Rank Oldbie
Rank
Oldbie

Ah, good job! Did you take time to figure out if it was U2, U8 or U16?

So there were (at least) two separate problems.

As for you PCM distortion question. I don't know.

1982 to 2001

Reply 374 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++
Cyrix200+ wrote on 2020-07-01, 18:40:

Ah, good job! Did you take time to figure out if it was U2, U8 or U16?

So there were (at least) two separate problems.

As for you PCM distortion question. I don't know.

As a matter of fact yes, I did. It was U2, the 74LS245 Octal Bus Transceiver. How that actually managed to introduce the crackling to the left channel, I do not know, but with a new chip it's gone.

Hear for yourself: https://drive.google.com/drive/folders/1AWtnF … vqX?usp=sharing

Still working on the PCM issue. No good leads so far 🙁

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 375 of 495, by TubeTimeUS

User metadata
Rank Newbie
Rank
Newbie

Could definitely be more bad logic chips. I'd check U20, U22, U21, U27, U28, U30, and also the DSP itself. It sounds like digital noise at this point. Where are you getting these logic chips from anyway? You have to be careful--a lot of grey market sources (eBay, etc) sell factory rejects or re-marked parts. Even though they're often cheap, they're more trouble than it's worth.

Reply 376 of 495, by root42

User metadata
Rank l33t
Rank
l33t
TubeTimeUS wrote on 2020-07-03, 15:03:

Could definitely be more bad logic chips. I'd check U20, U22, U21, U27, U28, U30, and also the DSP itself. It sounds like digital noise at this point. Where are you getting these logic chips from anyway? You have to be careful--a lot of grey market sources (eBay, etc) sell factory rejects or re-marked parts. Even though they're often cheap, they're more trouble than it's worth.

The DSP is from Mouser, via myself. So it should be fine if it wasn’t damaged in transit.

YouTube and Bonus
80486DX@33 MHz, 16 MiB RAM, Tseng ET4000 1 MiB, SnarkBarker & GUSar Lite, PC MIDI Card+X2+SC55+MT32, OSSC

Reply 377 of 495, by appiah4

User metadata
Rank l33t++
Rank
l33t++
root42 wrote on 2020-07-03, 17:24:
TubeTimeUS wrote on 2020-07-03, 15:03:

Could definitely be more bad logic chips. I'd check U20, U22, U21, U27, U28, U30, and also the DSP itself. It sounds like digital noise at this point. Where are you getting these logic chips from anyway? You have to be careful--a lot of grey market sources (eBay, etc) sell factory rejects or re-marked parts. Even though they're often cheap, they're more trouble than it's worth.

The DSP is from Mouser, via myself. So it should be fine if it wasn’t damaged in transit.

What @root42 said. I doubt it is the DSP chip, though I guess the uploaded program may have gone corrupt somehow? I don't have the means to program it at the moment but if you guys tell me what kind of programmer I need I can try to reprogram it with the HEX file.

The other components came from a variety of local suppliers. The one in particular that supplied all those other chips has sent me defective ICs in the past, so.. I made an order for replacements, and I will try them in place. It should take a few days to reach me, then we will see. *Fingers crossed*

Retronautics: A digital gallery of my retro computers, hardware and projects.

Reply 379 of 495, by root42

User metadata
Rank l33t
Rank
l33t

If you leave the LM393s like that you will have a bunch of baby op amps very soon... same goes for the 74LS74 and a bunch of others...

YouTube and Bonus
80486DX@33 MHz, 16 MiB RAM, Tseng ET4000 1 MiB, SnarkBarker & GUSar Lite, PC MIDI Card+X2+SC55+MT32, OSSC