I wouldn't be plugging it in with that low resistance between VCC and GND.

zuldan wrote on 2024-05-20, 06:25:

mkarcher wrote on 2024-05-19, 05:15:

These connections to GND and Vcc on the card seem intended and make sense.

So just to clarify, 30 ohms between 5v and GND may be normal with this card?

That's not what I meant with "seem intended and make sense". I referred to the high amount of chips being "shorted" to GND and Vcc. 30 ohms between GND and Vcc might be normal, depending on how Vcc is generated.

zuldan wrote on 2024-05-20, 06:51:

mkarcher wrote on 2024-05-20, 00:06:

Please check whether both VA and VD actually reach 5V with the CS4216 still removed. You can do this by injecting 5V into the ISA 5V contacts as you already did when you found that the CS4216 got too hot, no need to test that in system.

VD is 4.96v and VA is 2.2v

The low voltage on VA is bad. There might be two reasons for it: Either VA is generated from +12V, and you need to also supply +12V to get a good VA supply, or something is shorted on VA. Of course, both could be true at the same time. Looking at the whole card, I identify VR2 as 7805 regulator. This chip makes a very clean 5V supply from a minimum input voltage of around 7 to 8V. It is very common on sound cards to generate the +5V for analog stuff (VA) this way from the +12V voltage on the ISA bus. Maybe everything was fine, and you were "just" missing the 12V supply. In that case, the CS4216 might have tried to supply VA for the whole card from VD, and got hot due to the current it needed to supply.

So my suggested next steps:

• Check whether the VA supply pin is connected to the output pin of the 7805.
• If yes, that explains why VA is low without 12V input. Try feeding 12V and 5V at the same time into the card and measure VA and VD again.
• If VD is taken from +5V and VA is taken from +12V using a regulator, there should be no really low impedance path between VA and VD. If you see less than 5 ohms between VA and VD, and you also see VA is coming from the 7805, something that is connected to both VA and VD might have an internal short in that chip.

mkarcher wrote on 2024-05-20, 10:43:

The low voltage on VA is bad. There might be two reasons for it: Either VA is generated from +12V, and you need to also supply + […]
Show full quote

The low voltage on VA is bad. There might be two reasons for it: Either VA is generated from +12V, and you need to also supply +12V to get a good VA supply, or something is shorted on VA. Of course, both could be true at the same time. Looking at the whole card, I identify VR2 as 7805 regulator. This chip makes a very clean 5V supply from a minimum input voltage of around 7 to 8V. It is very common on sound cards to generate the +5V for analog stuff (VA) this way from the +12V voltage on the ISA bus. Maybe everything was fine, and you were "just" missing the 12V supply. In that case, the CS4216 might have tried to supply VA for the whole card from VD, and got hot due to the current it needed to supply.

So my suggested next steps:

• Check whether the VA supply pin is connected to the output pin of the 7805.
• If yes, that explains why VA is low without 12V input. Try feeding 12V and 5V at the same time into the card and measure VA and VD again.
• If VD is taken from +5V and VA is taken from +12V using a regulator, there should be no really low impedance path between VA and VD. If you see less than 5 ohms between VA and VD, and you also see VA is coming from the 7805, something that is connected to both VA and VD might have an internal short in that chip.

Apologies for the delay. Appreciate you helping me with this. I managed to get some time today to continue this.

[*]Check whether the VA supply pin is connected to the output pin of the 7805.[/b]

Confirmed, VA is connected to the out pin on the 7805

[*]If yes, that explains why VA is low without 12V input. Try feeding 12V and 5V at the same time into the card and measure VA and VD again.[/b]

Unfortunately, I don't have a way to connect both 5v and 12v at the moment.

[*]If VD is taken from +5V and VA is taken from +12V using a regulator, there should be no really low impedance path between VA and VD. If you see less than 5 ohms between VA and VD, and you also see VA is coming from the 7805, something that is connected to both VA and VD might have an internal short in that chip.[/b]

zuldan wrote on 2024-05-25, 06:56:

mkarcher wrote on 2024-05-20, 10:43:

If VD is taken from +5V and VA is taken from +12V using a regulator, there should be no really low impedance path between VA and VD. If you see less than 5 ohms between VA and VD, and you also see VA is coming from the 7805, something that is connected to both VA and VD might have an internal short in that chip.

(image showing around 13 kiloohms between VA and VD on the PAS16 without the CS4216 installed)

That looks fine. There is no short on the card between VA and VD. I guess some other chips (most analog pre-amplifiers) are also powered by VA (and possibly also -5V from the other regulator), so there will be load on VA, but no supply if you don't connect +12V as well as +5V. It's possible that the CS4216 is going to supply VA from VD if you only connect +5V to the card. The chip is not meant to work that way, you should supply VA and VD approximately at the same time.

As you have confirmed that VA and VD are not shorted on the card (both by measuring the resistance and by finding vastly different voltages on VA and VD without the CS4216 installed), you should also check whether VA and VD are shorted on the CS4216. If that's also not the case, possibly everything is fine, and the excessive heat in that chip was due to your test setup with missing +12V. As you don't have a setup to power multiple supply voltages (you should supply +5V, +12V and -12V for proper operation), I recommend you resolder the CS4216, put the card into a computer, and power it up for around 10 seconds. If the CS4216 stays cool in that setup, you should be fine for a proper test in a computer.

mkarcher wrote on 2024-05-25, 08:46:

As you have confirmed that VA and VD are not shorted on the card (both by measuring the resistance and by finding vastly different voltages on VA and VD without the CS4216 installed), you should also check whether VA and VD are shorted on the CS4216. If that's also not the case, possibly everything is fine, and the excessive heat in that chip was due to your test setup with missing +12V. As you don't have a setup to power multiple supply voltages (you should supply +5V, +12V and -12V for proper operation), I recommend you resolder the CS4216, put the card into a computer, and power it up for around 10 seconds. If the CS4216 stays cool in that setup, you should be fine for a proper test in a computer.

I’ll check if VA and VD are shorted on the CS4216.

Do you think it’s worth first powering on the card in the PC without the Crystal chip installed and seeing if I can get some basic function? I could also keep thermal camera on it and see what heats up.

zuldan wrote on 2024-05-25, 09:20:

Do you think it’s worth first powering on the card in the PC without the Crystal chip installed and seeing if I can get some basic function? I could also keep thermal camera on it and see what heats up.

That's actually a good idea, because if that fails, you can be quite sure there is another fault on the card. It seems the CS4216 is in fact only used for recording and playback of digitized sound, but it is not used as mixer. The FM music and the digital sound produced by the CS4216 will be mixed at some other point on the card. I expect that without the CS4216, FM music playback should work. Diagnostic utilities should be able to recognize the card, and the mixer utility should work as well to control the volume of FM music. If all of that works, its likely that there is no further critical fault on your card.

After getting a baseline without the CS4216, if you get more issues after soldering the CS4216, it is likely that the issues are caused by the CS4216.

mkarcher wrote on 2024-05-25, 19:34:

zuldan wrote on 2024-05-25, 09:20:

Do you think it’s worth first powering on the card in the PC without the Crystal chip installed and seeing if I can get some basic function? I could also keep thermal camera on it and see what heats up.

That's actually a good idea, because if that fails, you can be quite sure there is another fault on the card. It seems the CS4216 is in fact only used for recording and playback of digitized sound, but it is not used as mixer. The FM music and the digital sound produced by the CS4216 will be mixed at some other point on the card. I expect that without the CS4216, FM music playback should work. Diagnostic utilities should be able to recognize the card, and the mixer utility should work as well to control the volume of FM music. If all of that works, its likely that there is no further critical fault on your card.

After getting a baseline without the CS4216, if you get more issues after soldering the CS4216, it is likely that the issues are caused by the CS4216.

Agree with this, sometimes chips pop due to internal failures. It popping doesn't guarantee there is another problem on the card.

mkarcher wrote on 2024-05-25, 19:34:

That's actually a good idea, because if that fails, you can be quite sure there is another fault on the card. It seems the CS4216 is in fact only used for recording and playback of digitized sound, but it is not used as mixer. The FM music and the digital sound produced by the CS4216 will be mixed at some other point on the card. I expect that without the CS4216, FM music playback should work. Diagnostic utilities should be able to recognize the card, and the mixer utility should work as well to control the volume of FM music. If all of that works, its likely that there is no further critical fault on your card.

After getting a baseline without the CS4216, if you get more issues after soldering the CS4216, it is likely that the issues are caused by the CS4216.

So I plugged the card into a motherboard. When I powered it on I heard a bang and sparks went everywhere 🤣. One of the tantalums exploded.

I decided not to be lazy and replaced all 16 tantalums. 3 out of the 16 were bad according to the component tester. I also reinstalled the CS4216. There is still 30 ohms between 5v and ground so confirmed that's by design.

SHES ALIVE!! Here she is in action https://youtube.com/shorts/3M-B9fXW8L4

Thank you so much for all your help.

Unknown_K wrote on 2024-04-04, 07:03:

The MVA416 is the codec chip. As you can see from the first model revision (650-0022-01), it's actually just a rebadged Crystal CS4216-KL codec.

https://www.ebay.com/itm/324735163523

Thank you Unknown for that link to the chip!

Glad it worked out.

Cool. Glad you got it working. I have 3 of these cards myself, so far no issues with them. Fingers crossed though as they have have lots of tanties on them as you've noticed. 😀

Congratulations 😀 It sounds great from that video, both FM and digital audio!

I thought for sure that thing was toast since the other popped PAS16 I saw had the main chip burned as well.

pan069 wrote on 2024-05-31, 21:55:

Cool. Glad you got it working. I have 3 of these cards myself, so far no issues with them. Fingers crossed though as they have have lots of tanties on them as you've noticed. 😀

I think you should be fine. The problem with this card is that it had never been used. It was practically new old stock. I read somewhere that if tantalum’s don’t have any current go through them after many many years, they can go bad. Not sure how true this is but that seemed to be the case with mine.

Thermalwrong wrote on 2024-05-31, 23:49:

Congratulations 😀 It sounds great from that video, both FM and digital audio!

I thought for sure that thing was toast since the other popped PAS16 I saw had the main chip burned as well.

Yeah I thought she was gone. Especially after the bang. Was shocked when I heard sound.