SoundBlaster Pro 2.0 recapping

[stj]

Re: SoundBlaster Pro 2.0 recapping

most caps on ebay are fakes

[momaka]

Re: SoundBlaster Pro 2.0 recapping

Hi and thanks for your reply,
my idea is to make this card sound the best it can, I'm aware it's an old technology and by no means is hi-tech, (it was basically for video games). I want to make sure that it doesn't lack any frequencies due to bad caps.

In that case, the caps around the TEA2025b integrated amplifier (U24 on your board) are probably the most important. In particular, check if capacitors C93 and C94 are connected to pins 2 and 15 of the TEA2025b chip. If they are, those are the output caps for your Line Out jack. If you are using it to power passive speakers or headphones, then you might want to increase their capacitance to at least 1000 uF.

The caps around the TDA7284 are probably the second that matter the most, as that IC also has amplifier inputs and outputs.

Of course, you are recapping the whole sound card, if I understood correctly, so that shouldn't matter too much.

I didn't do a lot of research yet but it looks like Japanese (Rubycon) caps are among the best, and easy to find on e-bay.

Rubycon, Panasonic, Nichicon, and United Chemicon are all good choices.

I wouldn't get them from eBay, though. As stj said, there are a lot of counterfeits there. And it's not going to be that much cheaper (if at all), so it's not worth risking it. Since you are located in the USA, Digikey.com and Mouser.com would be where I go for genuine Japanese caps. Additionally, Badcaps.net has a capacitor store as well, if you are able to find all of the values for those small caps (as the store is suited more towards motherboard capacitors).

[eccerr0r]

Re: SoundBlaster Pro 2.0 recapping

"and easy to find on e-bay." -- I'd be a bit cautious of this statement.

I think Rulycons are easy to find on ebay

I'd first start off with the question: How bad does it sound now?

Bad caps on the sound card would sound like bad caps on any other audio device, you lose bass mostly, plus you'd possibly get high frequency noise from the rest of the system.

For the heck of it I took a look at my SB16 CT2940, caps are made by Nichicon, [A](?), and Elgen. The Nichicons were all 4.7uF, a couple of [A] branded 10uF, Elgens were the larger 100uF and 470uF.

I think most of them are still in passible condition though this board is 4 years newer (1995). Some are clearly worse than others.

[mockingbird]

Re: SoundBlaster Pro 2.0 recapping

Hi and thanks for your reply,
my idea is to make this card sound the best it can, I'm aware it's an old technology and by no means is hi-tech, (it was basically for video games). I want to make sure that it doesn't lack any frequencies due to bad caps. I didn't do a lot of research yet but it looks like Japanese (Rubycon) caps are among the best, and easy to find on e-bay.
Please feel free to correct me if I'm wrong or if you have any ideas, that's why I came here for.
Also, here is a picture of it,

Thanks for your time!
Dusko.

Post a list of the caps in this manner:

Voltage, Capacitance, diameter (width x height)

I think most of them are still in passible condition though this board is 4 years newer (1995). Some are clearly worse than others.

I just ordered an ESR meter (finally), one of those inexpensive microcontroller jobbies so I'm eager to test the old caps on the Sound Blasters... I've also got some other "vintage" sound cards I plan on re-capping and selling....

I would honestly just replace them. 1995... Even 1995 is almost 25 years old... Funny, old Panasonic butyric acid caps in ancient boom boxes work great after 40 years, but 20 year old caps leak elecrolyte and dissolve solder mask and copper traces (our brilliant member Wester547 might have some input on this).

[Wester547]

Re: SoundBlaster Pro 2.0 recapping

I would honestly just replace them. 1995... Even 1995 is almost 25 years old... Funny, old Panasonic butyric acid caps in ancient boom boxes work great after 40 years, but 20 year old caps leak elecrolyte and dissolve solder mask and copper traces (our brilliant member Wester547 might have some input on this).

Hah, I’m hardly knowledgeable, nevermind brilliant. But thanks for the vote of confidence.

Generally, the use of tetraakyl quaternary ammonium salts caused that problem. It raised the pH balance of the (then alkaline) electrolyte which expedited the deterioration of the rubber bung to the point of electrolytic leakage, after which the very acidic and conductive electrolyte does away with the solder mask, PCB traces, et al, in what sometimes can be a spectacular manner, as you noted. The quaternary salts are still used in double-layer electrolytics.

It also seems that there was a batch of Rubycon YK, Nichicon VR, and Chemi-con SM/SME series capacitors that had this issue as well, manufactured from the early to mid 90s (possibly late 80s too). All being standard 85ºC capacitors, they should not have behaved as the quaternary capacitors did, but it’s possible some other concentration of an unknown compound in the electrolytic solution was resulting in the same problem. It could have been some sort of solder residue or flux eating away at the rubber bung as well, but sometimes it would even happen with capacitors retaining their original leads so it could have been down to the quality of the bung itself. Either that or no one admitted to anything and silently resolved that conundrum (at least all the major brands admitted that no quaternary salts are used anymore in their regular electrolytics).

I think 40-year old electrolytics are still fine because either they use epoxy-coated rubber or polymeric (metal, ceramic, glass, etc) seals which last much longer than hermetic (rubber) seals. It could just be that the γ-butyrolactone/carboxlyic acid electrolyte is less aggressive to the sealing materials. That said, the suggestion to replace all the electrolytics at nearly 23-years old is a good one, especially if some of them are Elgens. As others have, I recommend keeping away from eBay as even -genuine- looking capacitors can be counterfeits. Authorized distributors like Digikey, Mouser, Farnell, etc, are recommended.

[mockingbird]

Re: SoundBlaster Pro 2.0 recapping

Hah, I’m hardly knowledgeable, nevermind brilliant. But thanks for the vote of confidence.

Such modesty...

I think 40-year old electrolytics are still fine because either they use epoxy-coated rubber or polymeric (metal, ceramic, glass, etc) seals which last much longer than hermetic (rubber) seals. It could just be that the γ-butyrolactone/carboxlyic acid electrolyte is less aggressive to the sealing materials. That said, the suggestion to replace all the electrolytics at nearly 23-years old is a good one, especially if some of them are Elgens. I recommend keeping away from eBay as even -genuine- looking capacitors can be counterfeits. Authorized distributors like Digikey, Mouser, Farnell, etc, are recommended.

Isn't Nichicon PV series exactly that? Just Nichicon PJ with epoxy seals... PJ has 2000-3000 hours on the smaller case sizes while PV has a 5000 hour rating.

[Wester547]

Re: SoundBlaster Pro 2.0 recapping

Isn't Nichicon PV series exactly that? Just Nichicon PJ with epoxy seals... PJ has 2000-3000 hours on the smaller case sizes while PV has a 5000 hour rating.

As far as I can tell, they use gray, flat rubber bungs rather than the darker, thick rubber stand-offs / gas escape end-seals characteristic of most miniature through-hole Nichicons. Sanyo OS-CON SP solid capacitors used epoxy-coated rubber seals (not to be confused with their polymer line, which uses a functional conductive polymer cathode layer rather than TCNQ complex salts).

Nichicon PV having a higher endurance rating could mean that the materials are better suited to higher temperatures.

The only advantage I can think of the rubber bungs is they allow more hydrogen to pass through the seal, which reduces the pressure build up inside the case, in the event that H2 is generated.

[Wester547]

Re: SoundBlaster Pro 2.0 recapping

As an addendum and a correction to my third to last post in the thread, I meant to say that rubber seals are polymeric and that ceramic/glass/metal seals are hermetic.

Also, looking at the photo Dusko75 posted, I can definitely see some Elgens on that card. Those look to be 25+ years old now, whereas eccerr0r’s card is closer to 22+ years old. I’d replace them either way and hope that none have begun leaking from the bung.

[eccerr0r]

Re: SoundBlaster Pro 2.0 recapping

I'm going to wait until I really need that card again, right now I have other sound cards that I can use that don't have a mass of electrolytics on them =o

I guess the other thing is that these older sound cards more likely have power amps on them unlike newer cards and on-motherboard sound. The power amps, though only a watt or so, are single ended and thus need DC isolation caps which can dry out. They indeed are the marginal/bad ones on my card. Alas for many many years I've used external amplifiers and the onboard caps for the PA don't matter. But the ones for signal passage do matter...

The newer sound cards that don't have PAs have switched to all MLCC's (for cheaper assembly automation) and probably don't degrade as fast... or at all. They may start off sounding pretty crappy however.

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

My GOD! First, thank you VERY MUCH to all of you, I didn't know about all these replies. I was thinking the forum will send me email alerts or something, I have to check the settings.
I'm going to read all these messages and get back ASAP.

Again, Thank You.

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

Additionally, Badcaps.net has a capacitor store as well, if you are able to find all of the values for those small caps (as the store is suited more towards motherboard capacitors).

Momaka, thank you for the detailed reply, definitely will check on that.
I didn't know there's such a thing as fake caps!! so I'll go for one of your suggestions, I'm just realizing there's a store here so will check on that too.
How about Allied Electronics? or Parts Express? I'm mentioning them because I think I bought from them before, nothing else.
All this process will take me some time because I need to build or get a 486 PC first to test it before and after, unless it sounds great from start without touching it ... (don't think so)....

Thanks again,
Dusko

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

I'd first start off with the question: How bad does it sound now?

Hi eccerr0r, I don't know how it sounds because I haven't test it, as I mentioned to "momaka", I will build a 486 computer whenever I get all the parts and before doing anything, see how good or bad the card sounds as is, then I'll decide. Mine also has Elgen caps except 3 blue ones that have an "R" inside a square. Just curious, is Elgen a good cap company?

Thanks,
Dusko

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

Post a list of the caps in this manner:

Voltage, Capacitance, diameter (width x height)

Hi Mockingbird, I'll get all the info together as soon as I get a hold of a 486 computer to test it as is. Didn't know the values do not dictate the size?

I'll get back with all the info,
Thanks,
Dusko

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

As an addendum and a correction to my third to last post in the thread, I meant to say that rubber seals are polymeric and that ceramic/glass/metal seals are hermetic.

Also, looking at the photo Dusko75 posted, I can definitely see some Elgens on that card. Those look to be 25+ years old now, whereas eccerr0r's card is closer to 22+ years old. I'd replace them either way and hope that none have begun leaking from the bung.

Awesome Wester547, thanks for all your input, that's a lot of knowledge right there!. Yes, as I mentioned my card has mostly Elgen caps and 3 blue with an "R".

I forgot to mention, the two big caps (near the TEA2025B) both look like they have a dried residue on the top (leakage?), also the plastic cover doen't seem as tight as it should (like it was melted a little, or eaten by acid). The card itself has no residues, it actually looks amazingly preserved.

Thanks,
Dusko.

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

In that case, the caps around the TEA2025b integrated amplifier (U24 on your board) are probably the most important. In particular, check if capacitors C93 and C94 are connected to pins 2 and 15 of the TEA2025b chip. If they are, those are the output caps for your Line Out jack. If you are using it to power passive speakers or headphones, then you might want to increase their capacitance to at least 1000 uF.
capacitor store as well, if you are able to find all of the values for those small caps (as the store is suited more towards motherboard capacitors).

Yes, Those two caps are connected to those pins in the TEA2025B, they are the bad looking ones I mentioned in the previous post. I can't wait to test this thing!

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

Here's a picture of the two caps...

[mockingbird]

Re: SoundBlaster Pro 2.0 recapping

Hi Mockingbird, I'll get all the info together as soon as I get a hold of a 486 computer to test it as is. Didn't know the values do not dictate the size?

I'll get back with all the info,
Thanks,
Dusko

The markings indicate the capacity... The physical size needs to be measured by hand.

[kc8adu]

Re: SoundBlaster Pro 2.0 recapping

do yourself a favor and avoid buying caps on ebay.too many fakes!

Hi and thanks for your reply,
my idea is to make this card sound the best it can, I'm aware it's an old technology and by no means is hi-tech, (it was basically for video games). I want to make sure that it doesn't lack any frequencies due to bad caps. I didn't do a lot of research yet but it looks like Japanese (Rubycon) caps are among the best, and easy to find on e-bay.
Please feel free to correct me if I'm wrong or if you have any ideas, that's why I came here for.
Also, here is a picture of it,

Thanks for your time!
Dusko.

[momaka]

Re: SoundBlaster Pro 2.0 recapping

Yes, Those two caps are connected to those pins in the TEA2025B, they are the bad looking ones I mentioned in the previous post. I can't wait to test this thing!

Doesn't look like any leakage is coming from them. When caps leak, they either do it from the top after their vent bulges and opens or they leak from the bottom, in which case the board would look in very bad condition near them (which I don't see on your board, so that's not the case). So whatever residue they have on them is from something else.

Since those are the caps on the output pins, I suggest going with 1000 uF caps in the spot. That way, if you directly connect headphones or small non-amplified speakers to the sound car (on old cards, you can totally do that), you'll get better bass response.

[Dusko75]

Re: SoundBlaster Pro 2.0 recapping

Thank you momaka and everyone else's comments,

it's been over a year I posted about my Sound Blaster Pro recap project, didn't forget about it!. Before getting into the recapping thing, I needed to get a 486 to do all the testing, I ended up buying 4 motherboards (2 ISA VLB and 2 ISA - PCI), I did that because they were having issues or not working at all, one of them had a bad BIOS, etc. etc., at the end I fixed all 4, it took me several months to figure out what was wrong with each of them! Also bought some CPUs and some video cards (all from Ebay), did have a hard time getting all that to work: jumpers, finding the mobo's manuals online, etc. also, I only have one tower to do all the testing (and/or playing around). The Sound Blaster Pro was super noisy on 2 of the mobos while the other 2 were surprisingly clean, even at full volume (actually, no noise at all). I thought recapping it would get a cleaner sound from the other two mobos but didn't happen that way.

Recapping the Sound Blaster Pro:

So, figured out all the caps from the board and bought all legit brands: Panasonic, Rubycon and Nichicon, got them from Digikey, (I'm just realizing "BADCAPS" SELLS CAPS!). This is my first recap project so it took me like 4 hours to finish, very happy with it, the board looks great but I haven't noticed any major improvements in sound quality, maybe a little more bass but not so sure about that. I'm going to need to do some recordings with my other Sound Blaster Pro (which is unmodified) and compare it to this one. The noise is still present on the other two mobos so it should be an ISA controller thing, maybe I'll end up recapping the mobos too. I got a ESR meter and the old caps seem fine (@120Hz) but I'm not an expert in electronics and caps are tricky to test, I found a couple of ESR charts online with different values so I don't know
I'm still happy about this project, it's something I wanted to do for a long time, I'll post back after comparing the two boards, there probably is an improvement in sound quality that I haven't really noticed, it's not Hi-Fi audio anyway.
Here are some pics...