mkarcher wrote on 2025-08-05, 19:57:

What edition of Monkey Island are you using? If you use a version without speech output, Monkey Island is not supposed to use digital sound output, neither 8 bit nor 16 bit. If you get issues in Monkey Island (not the "CD talkie"), the issue is not related to 16-bit output.

I've seen a YouTube video with similar symptoms, and in that case, the 7809 was overheating, because IIRC the TEA2025 was broken and used too much current. When the 7809 overheats, it shuts down for a moment, then it cools down and turns on again. Depending on the external conditions and the load, the turn on/turn off oscillation of the 7809 can get into the audible range. So I recommend you to test carefully whether the voltage regulators get hot. You can burn your fingers! The overtemperature shutdown of the 78xx series kicks in above 125°C (way more than 300°F) inside the chip, so I would expect a case temperature above 100°C if the overtemperature shutdown activates.

Hi,
there is no overheating! I can put my finger on 7809, and i tested it for few minutes. Stuttering starts as soon as i turn the volume up, no matter what game it is. Only chips who are getting a little bit hotter are U18, U20 and U21, but they are not really hot, around 50 degrees of celsius.

butjer1010 wrote on 2025-08-05, 14:07:

No change with High DMA to 7, both jumpers and software settings. 16but sound still crackling. Would internal amplifier change anything, now it is on disable (jumpers OPSL and OPSR are on 1-2)? If i enable this?

Yout internal amplifier is obviously enabled. Also, the position of pin 1 is indicated in the TH99 diagrams by the one jumper pin that is not fully black, but has a white square in the center. On the SB16 drawing, the "pin 1" is at the top, so your card is jumpered 2-3.

The volume wheel only affects that amplifier. If you disable the amplifier, you get line level, no matter what the volume wheel is set to.

This also means: Nothing on the cards depends on the volume setting except for the final amplifier. If the issue depends on the volume selected at the hardware wheel, the issue is clearly in the final amplification circuit. If you use powered speakers, you can disable the internal amplifier and test whether the problem disappears. Currently, I suspect the TEA2025B chip to be broken.

mkarcher wrote on 2025-08-05, 21:07:

Yout internal amplifier is obviously enabled. Also, the position of pin 1 is indicated in the TH99 diagrams by the one jumper pi […]
Show full quote

butjer1010 wrote on 2025-08-05, 14:07:

No change with High DMA to 7, both jumpers and software settings. 16but sound still crackling. Would internal amplifier change anything, now it is on disable (jumpers OPSL and OPSR are on 1-2)? If i enable this?

Yout internal amplifier is obviously enabled. Also, the position of pin 1 is indicated in the TH99 diagrams by the one jumper pin that is not fully black, but has a white square in the center. On the SB16 drawing, the "pin 1" is at the top, so your card is jumpered 2-3.

The volume wheel only affects that amplifier. If you disable the amplifier, you get line level, no matter what the volume wheel is set to.

This also means: Nothing on the cards depends on the volume setting except for the final amplifier. If the issue depends on the volume selected at the hardware wheel, the issue is clearly in the final amplification circuit. If you use powered speakers, you can disable the internal amplifier and test whether the problem disappears. Currently, I suspect the TEA2025B chip to be broken.

I tried to disable the internal amplifier, and it was ok when i turn down volume in mixerset a little bit above half, but as soon as i turn it one step to the right, sound started to stutter again 🙁
I have made an order for TEA2025B and 7809 (only 1.5A was available, but that shouldn't matter, right?), and i'm waiting for them to arrive. After changing, i will let You know.
Thanks for now 😉

Today the amplifier arrived, i have change it, and now the problem with the volume is gone, but still, the 16bit sound is broken. Tried with older drivers, same problem. What are the option for diagnosing 16bit sound now? Could that be problem with new capacitors, if they are not same value like original, or could it be some else?
Thanks in advance

I don't think the caps are related to the 16-bit sound issue. The CT1746 bus interface chip always sends 16-bit audio data to the CT1701 DAC, no matter whether you play back 8-bit sound or 16-bit sound. This means the DAC and all the analog stuff does not care about bitness, and the only caps you changed that are not in the analog section are C13 and C14. It is very unlikely that C13 or C14 cause this issue.

The CT1746 chip has a dedicated section dealing with 16-bit DMA. If you get the same issue with broken 16-bit sound in multiple systems, and you are sure that your DIAGNOSE installation is not damaged, I would suspect that the CT1746 is broken. Hmm, I wonder what happens if you run DIAGNOSE from a very slow storage device. Does DIAGNOSE stream the 16-bit data from hard disk, and if the hard disk can not keep up with the required rate, it will cause interruptions? Think about this possibility in case you are using a slow CF card in an XT-IDE. You might want to copy the DIAGNOSE stuff to a RAM disk to remove the storage from the equation (or you can confirm by looking at the hard drive LED that the hard drive is not accessed during 16-bit test anyway, in which case my idea is pointless).

mkarcher wrote on Yesterday, 20:16:

I don't think the caps are related to the 16-bit sound issue. The CT1746 bus interface chip always sends 16-bit audio data to the CT1701 DAC, no matter whether you play back 8-bit sound or 16-bit sound. This means the DAC and all the analog stuff does not care about bitness, and the only caps you changed that are not in the analog section are C13 and C14. It is very unlikely that C13 or C14 cause this issue.

The CT1746 chip has a dedicated section dealing with 16-bit DMA. If you get the same issue with broken 16-bit sound in multiple systems, and you are sure that your DIAGNOSE installation is not damaged, I would suspect that the CT1746 is broken. Hmm, I wonder what happens if you run DIAGNOSE from a very slow storage device. Does DIAGNOSE stream the 16-bit data from hard disk, and if the hard disk can not keep up with the required rate, it will cause interruptions? Think about this possibility in case you are using a slow CF card in an XT-IDE. You might want to copy the DIAGNOSE stuff to a RAM disk to remove the storage from the equation (or you can confirm by looking at the hard drive LED that the hard drive is not accessed during 16-bit test anyway, in which case my idea is pointless).

I didn't try card in another system. It is a DOS 6.22 on a CF card, but it works pretty fine if You ask me. I will try on another PC, and see if that makes any change, and i will look at the HDD led to see if "blinks" during 16bit audio.
Thanks

mkarcher wrote on Yesterday, 20:16:

I don't think the caps are related to the 16-bit sound issue. The CT1746 bus interface chip always sends 16-bit audio data to the CT1701 DAC, no matter whether you play back 8-bit sound or 16-bit sound. This means the DAC and all the analog stuff does not care about bitness, and the only caps you changed that are not in the analog section are C13 and C14. It is very unlikely that C13 or C14 cause this issue.

The CT1746 chip has a dedicated section dealing with 16-bit DMA. If you get the same issue with broken 16-bit sound in multiple systems, and you are sure that your DIAGNOSE installation is not damaged, I would suspect that the CT1746 is broken. Hmm, I wonder what happens if you run DIAGNOSE from a very slow storage device. Does DIAGNOSE stream the 16-bit data from hard disk, and if the hard disk can not keep up with the required rate, it will cause interruptions? Think about this possibility in case you are using a slow CF card in an XT-IDE. You might want to copy the DIAGNOSE stuff to a RAM disk to remove the storage from the equation (or you can confirm by looking at the hard drive LED that the hard drive is not accessed during 16-bit test anyway, in which case my idea is pointless).

I bringed another PC with my other CT1750 down from attic, and You were right (again) 😀
Same problem on both cards, on both systems, and both systems have CF card instead of HDD. There is no blinking light on hdd led when accessing CF at 16bit sound testing. I assume the card is ok than, i just need to try it on some newer PC, with hdd, right?

How can i load diagnose.exe to ram, as You mentioned? I know how to to that on Amiga 😀, but i don't know how to do that on PC 🙁

butjer1010 wrote on Today, 11:37:

How can i load diagnose.exe to ram, as You mentioned? I know how to to that on Amiga 😀, but i don't know how to do that on PC 🙁

Add "DEVICE=C:\DOS\RAMDRIVE.SYS 2048 512 /E" to CONFIG.SYS after you loaded HIMEM.SYS, this will use 2MB of XMS to create a RAM drive of that size. This will create a drive with a separate drive letter, on which you can copy the SB16 folder. To be extra safe, you can make sure DIAGNOSE is using the copy on the RAM drive by also setting "SET SOUND=D:\SB16" (or whatever drive letter the RAM drive got).

mkarcher wrote on Today, 16:41:

butjer1010 wrote on Today, 11:37:

How can i load diagnose.exe to ram, as You mentioned? I know how to to that on Amiga 😀, but i don't know how to do that on PC 🙁

Add "DEVICE=C:\DOS\RAMDRIVE.SYS 2048 512 /E" to CONFIG.SYS after you loaded HIMEM.SYS, this will use 2MB of XMS to create a RAM drive of that size. This will create a drive with a separate drive letter, on which you can copy the SB16 folder. To be extra safe, you can make sure DIAGNOSE is using the copy on the RAM drive by also setting "SET SOUND=D:\SB16" (or whatever drive letter the RAM drive got).

I did all that, made RAMDRIVE, copied sb16 folder to d:, started diagnose.exe from d, and it stopped at testing high DMA channel (when configuring the card). Didn't even get to testing sounds.
EDIT: I have select Use setting like low DMA (channel 1), and it worked. You were right again, 16bit testing works normally on RAMDRIVE 😀
So the card is good, but the CF card is problem 🙁
Thanks a lot, now i know the Hard disks are better option.....