VOGONS


Reply 40 of 85, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie

I've tried to analyze the pictures to form some kind of mental schematic of the card.

I don't know how much you know about IC pin numbering etc, so here is a crash course. Every chip has pin 1 marked with dot or dent; also every chip on card has the same orientation, i.e. they all happen to have pin 1 on their lower left corner, also marked by square copper pad on PCB. Then they are numbered counter-clockwise.

I can see the resistor color codes and ICs markings from your photos and other photos, but I cannot see any other component values. Can you help me a bit and make a list of what markings are on the components? I am mainly interested in the blue volume pot R14, transistor Q1 (black three-legged device), C18 value (220uF 16V in some other picture?), diode D1 markings if you can find, and all the rest of the smaller capacitor markings (C1 through C26 or how many there are).

So here are some things for you to measure.

1) check if the RC4136 op-amp U8 has power. -12V from ISA bus goes to C21, C5 and to pin 7 of U8. +12V from ISA bus goes to C22, C4 and to pin 11 of U8. Measure the IC pins with multimeter, the black lead going to any nearby ground, even PC case should work.

2) check if the YM3014B DAC U7 and op-amp work together. First measure DAC supply voltage, pin 1 of U7 should have +5V.

3) Pin 7 of U7 is the half-supply voltage that comes out, is filtered with C3 and goes to U8 pin 9 for buffering. This should measure +2.5V.

4) Buffered half-supply voltage comes out of U8 pin 10, goes also to U8 pin 9 for feedback and ends up back to U7 pin 8. This should measure +2.5V as well.

5) U7 pin 2 is DAC output which goes to U8 pin 5 for buffering. When no audio is playing, it should read +2.5V.

6) Buffered audio comes out from U8 pin 4 and goes also to U8 pin 6 for feedback. Should read +2.5V when no audio playing. This pin is where something should be heard if connected to very small speaker or preferably some kind of amplified speaker line input. Connect amplifier ground first to PC, then this point, or potential difference of ungrounded devices may kill the op-amp.

7) Next the audio is filtered with two second-order filters with some gain. If I am not mistaken, the first filter stage has gain of 2.2 so a 2.5V input to U8 pin 2 should make U8 pin 3 have 5.5V, and that fed to second filter stage, U8 pin 13, with gain of 1.15, should provide 6.325V on U8 pin 12. Audio should be heard on a speaker if connected to U8 pin 3 or U8 pin 12. I do not know how loud it is.

8) After the op-amp, there is the mute and volume control stage. C10 removes DC component from filter stage output, so DC voltages are no longer important to measure. Audio should still be there if it works.

9) after volume pot, the audio goes to U9 LM386 power amplifier input pin 3. It should be amplified with a gain of 20, and U9 pin 5 is the output. The big capacitor C18 blocks DC so amplified audio is available after it and on the audio output connector.

Please post your findings so I can comment on them.

There is no visible marks of anything being broken. From what other people have said, the RC4136 op-amp may have died. But it is unknown if it has taken something else with it, like the DAC or power amplifier chip or even the YM3812 OPL2 chip through DAC pins.

Reply 41 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t

Wow, thanks for your help! I've squinted through a magnifying glass with a light nearby, and I can see many of the markings that you need, but there are a few that I could not see. I could not see some markings, because the component was not angled correctly for me to see its side.

R14 = 91D1DW32D15
10K MEXICO
J9141X

Q1 = 2N
39()4 I wonder if the () is a zero symbol?

C18 = I cannot see any markings.

D1 = Sorry, no idea

C19=K5M/104
All of these are the same as C19: C20,C24,C25,C26,C1,C2,C4,C5,C14,C16

C21=106/+16K
All of these are the same as C21: C22,C11,C3,C15, and C23 is likely to be the same, too.

C6=K1K/472
All of these are the same as C6: C7,C8,C9

C10=475/+16K

C12=102/A1K

C13= I cannot see the side with the markings on.

C17=K5M/473

Tomorrow morning, I will begin your task list, numbered 1 to 9. Please can you give me a very quick refresher course, as to how to perform the very first task? That task is "Check -12V from ISA bus goes to C21".

So, I get a multimeter (MM), and:

1) Do I set the MM's dial to the "V section" (top left section), and then set the dial to the 20 value?
2) Do I then put the Adlib card in to a mobo, and turn on the mobo, to get to the DOS C:\ prompt?
3) Do I sellotape the MM's black probe to the case screw of a grounded computer?
4) Where do I put the MM's red probe? Do I put the red probe here? (see the yellow arrow in the photo below)

Thanks a lot!

probe.jpg

Reply 42 of 85, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie

Thank you very much for the info.

R14 is 10K audio volume control pot from Bourns. 10K is also in line with LM386 examples for volume control pot.

Q1 can be 2N3904, it's a common transistor. If D1 is assumed to be a regular small-signal diode, the transistor forms a start-up mute circuitry with C11 so it keeps the signal muted to ground for 40-50 milliseconds after 5V power is connected.

Capacitors with "104" marking, they are 100nF capacitors (10*10000 pF), typical value for local power supply bypass near chips. Non-polarized, could be ceramic capacitors. Also suggested value to audio amplifier negative input.

Similarly "106" means 10 microfarad (uF) capacitor, used mostly for board-level power supply bypass or bulk capacitor.
Polarized, could be tantalum capacitors.

"472" means 4.7 nF, used in the audio filters. With 2.2k resistors it makes filter cutoff frequency somewhere between 15-16 kHz, so it makes sense. Non-polarized, could be ceramic capacitor.

"475" for 4.7 uF, DC blocking capacitor for audio signal between filter and power amp stage. Polarized, could be tantalum capacitor.

"102" for 1nF, may be a part of simple low-pass RC filter to remove very high frequency noise (above 72kHz). Non-polarized, could be ceramic.

C13, maybe same purpose, might be even lower in value. Non-polarized, could be ceramic.

C17 "473" is 47nF, very near suggested 50nF value for LM386 amp output.

OK, so back to your measurements.

1) Yes, set selector for 20V, it is enough. Make sure your red lead is connected to V input for voltage measurement, not the A input for current measurement.

2) Yes, you have to have the Adlib connected to motherboard and power turned on to do these measurements. So be careful not to poke any two pins together for short-circuits.

3) Well sort of. Taping the negative black lead to PC screw/case may not make a good connection if the tape gets loose. I would just connect it to for example nearby PC screw with another hand. Or perhaps plug it inside hard disk power connector ground pin (black wire).

4) Exactly where the arrow points, that is the -12V side of C21. The main point however is to make sure there is -12V on U8 pin 7 though, I just mentioned C21 and C5 so you can follow the copper traces on PCB from ISA bus to op-amp U8. Another way is to just measure if -12V is at any of the points and do another measurement when Adlib is not in the computer to verify the copper trace is OK, for example using the resistance measurement mode to see if there is near 0 ohms between U8 pin 7 and C21 negative lead. But still, dirty ISA bus connector might prevent voltage ending up on the Adlib card, so it is best to measure voltage when powered on.

What annoys me about the op-amp is that it seems to have an odd pin-out order compared to today's commonly available op-amps, so it cannot be directly replaced with more modern commonly available op-amp like TL074 just for testing if it needs replacing. Farnell still sells real TI RC4136N though.

Reply 43 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t

Awesome. Please note that I will begin testing later on today - I've just got to set up my "test bench" a bit better, so that I can work on the back of the Ad Lib sound card more easily. In the meantime, I have a couple of questions please -

I have a typical desktop PC speaker, with red and black wires attached to it. I guess that it must not be an amplified speaker, because it looks so basic. I am happy to buy an amplified speaker (from ebay) for the purposes of these testing experiments. Please can you give me a rough idea of what component to search for, on ebay? For instance, if I type in the search "amplified speaker" in to ebay, thousands of results are returned. Is there some keyword that I can append to my search, in order to narrow the search down, so that I can buy the most appropriate item for my testing requirements? Thanks a lot.

Also, "luckily" I recently broke an old Creative Sound Blaster 2.0 sound card. Hopefully, this broken item can provide essential parts for this project. However, please note that at this time I do not have any tools such as a soldering station, or any knowledge in how to use one. But perhaps I can "cross that bridge" when I need to come to it? 😉

Reply 44 of 85, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie

Any amplifier+speaker configuration will do. The system must have some kind of power supply (mains, transformer, batteries, etc) and line-level input connector. What do you currently use to hear any sounds from your PCs? If you must buy something, nearest markets may sell Logitech/Creative brand or similar desktop multimedia speakers, get the cheapest.

But because the audio signal on board the Adlib rides over a DC voltage bias, and the filter stages have a huge amplification, you should only connect the amplifier input to buffered DAC output before the filter stages. I have to check what pins those are. Who knows how the amplifier behaves when it has DC voltage bias applied to its input, so at least it should be connected to a point where DC voltage bias is smallest.

Sound blaster could provide all components, depends on how you broke it. Some models even have the FM chips on a IC socket so you can detach them without soldering. They may not have the same op-amp though.

Reply 45 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t

Regarding my current PC speaker system, I have a pair of powered multimedia speakers, with a mains power cord/plug. Also, they have an audio cable attached to them, and on the end of that cable is a 3.5mm headphone connector. Please see attached photo below.

Unfortunately, the SB2 sound card that I broke has not got a socketed FM chip. However, I do have another (working) SB2 sound card, with a socketed FM chip. I could "borrow" it, for testing purposes..

This is the broken SB2:
sb2b.jpg

Attachments

  • phones.jpg
    Filename
    phones.jpg
    File size
    10.61 KiB
    Views
    1770 views
    File license
    Fair use/fair dealing exception

Reply 46 of 85, by jwt27

User metadata
Rank Oldbie
Rank
Oldbie
Jepael wrote:

But because the audio signal on board the Adlib rides over a DC voltage bias, and the filter stages have a huge amplification, you should only connect the amplifier input to buffered DAC output before the filter stages. I have to check what pins those are. Who knows how the amplifier behaves when it has DC voltage bias applied to its input, so at least it should be connected to a point where DC voltage bias is smallest.

Most if not all audio amplifiers have a DC blocking capacitor on the input so that shouldn't be a problem. You might hear a short 'pop' but that's it.

The jack plug is easy: first ring is ground, second ring is right channel and the tip is for the left channel. So connect the first ring to your PC case or one of the black wires from your PSU, and use the tip to probe around the card.

Reply 47 of 85, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie

Multimedia speakers should do, but do use multimeter first, so we know the voltages are sane, before connecting multimedia speakers anywhere. I could make a program to output some sine waves out of the Adlib, so you could measure with multimeter. Or you could just run a game to get audio out and see changes in voltage.

That is a standard 3.5mm stereo plug or tip-ring-sleeve connector. Tip is left input, ring is right input, and the biggest metal sleeve is ground. You only need ground and the other input.
Since your multimedia speakers most likely has 2-pin non-grounded mains plug, make sure you connect PC and speaker grounds firmly first, before the input signal.
You could buy some connectors or cables which you could use to convert whatever connector on your speakers and splice up the cable to separate signal wires. You could then screw the ground wire under PC case screw, and have a signal lead hanging loose which you could poke around the card.

Too bad you don't have access to borrow/use an oscilloscope..

Yes that SB does have OPL2 and its DAC but it is surface-mount (SMD) version. I think I have never seen SMD OPL2s before, only the bigger non-SMD versions. Interesting. No compatible op-amps there.

Reply 48 of 85, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie
jwt27 wrote:

Most if not all audio amplifiers have a DC blocking capacitor on the input so that shouldn't be a problem. You might hear a short 'pop' but that's it.

Yes I should have mentioned that, but who knows how much bias they can handle. If something is broken and speakers are connected to +/-12V, the pop is short but very loud and could break something.

DAC output has 2.5V bias, and after the filter stages it is about 6V. I have no guess what the bias is at amplifier input, but output could be 6V, before it is removed for output connector. Also the signal level is way above normal line level output. If I recall I could not understand the datasheet if YM3014B has 2.5Vpp or 1.25Vpp output, I am more familiar with OPL3 chips and their DACs and they have 2.5Vpp output.

I think most audio amplifiers could cope with 2.5V bias.

Reply 49 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t

Please note that I am located in the UK, and we use 3 "prong" mains power plugs. They look like this - see attached image. In the image, the "terminals" are where the "prongs" are located, that go in to the mains power socket on the wall. Does this information alter any of the advice/information that has been provided in any of the previous messages? Thanks a lot.

Attachments

  • uk mains.jpg
    Filename
    uk mains.jpg
    File size
    28.25 KiB
    Views
    1736 views
    File license
    Fair use/fair dealing exception

Reply 50 of 85, by Jepael

User metadata
Rank Oldbie
Rank
Oldbie

It does not guarantee that the device actually connects the earth prong anywhere. Most multimedia speakers and external hard drives etc have just used unearthed 2-prong plugs here. Even the ones which come with UK plugs.

You can use the multimeter and measure if audio plug ground is connected to earth prong as well like computer case is connected to earth prong.

Reply 51 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t
Jepael wrote:

You can use the multimeter and measure if audio plug ground is connected to earth prong as well like computer case is connected to earth prong.

I am sorry about this, but can you please tell me how to do this? I am not from any engineering background, and so all of this is completely alien to me.

Please can you tell me what dial I set on the MM (multimeter). Then, do I switch on my MMS (multi media speakers), and then with the MMS switched on, where do I put the red and black probes? And when these two probes are in position, what reading on the MM am I looking for? Thanks a lot for any help. BTW, I'm nearly ready with my reorganised workbench, so I can begin testing later on this afternoon. Thanks.

Hopefully the MM reading for the MMS is "good", in as much as the MMS are grounded, and therefore I do not need to earth them with a piece of wire, because I don't have any suitable wire in the house. However, if the MM reading for the MMS is "bad", and I do need to earth the MMS, then I can buy some wire from ebay. Any advice as to what type of wire I need to buy would be very welcome, because you mention "splicing". Thanks a lot.

Reply 52 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t

Quick update. I have successfully done the first couple of measurements. They are -

Test #1) check if the RC4136 op-amp U8 has power. -12V from ISA bus goes to C21, C5 and to pin 7 of U8. +12V from ISA bus goes to C22, C4 and to pin 11 of U8.

Notes #1) I only checked pin 7, and also pin 11 of U8. I did not check the "C" elements of the card.

Results #1)

Pin 7 = -11.40
Pin 11 = +12.40

Reply 54 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t

Quick update 2. I've done the first 6 numbered tests, excluding the multimedia powered speaker test. All of the results are as expected:

1) -11.40, +12.40 (test results repeated from my previous post, above)
2) 4.92
3) 2.46, 2.46
4) 2.47, 2.47
5) 2.46, 2.46
6) 2.47, 2.47

In order to continue testing, I need to understand how to safely ground my multimedia speakers. I have an idea. Would this work -

Remove the black probe lead from the socket in the multimeter. Put one end of the MM black probe lead in to the black ground socket hole of the PSU molex cable. Then take the other end of the black probe lead, and physically hold it tightly against the ground section of the multimedia speaker connector - this is the sleeve section of the "tip-ring-sleeve" connector. Is this dangerous to either myself, or the computer equipment? If anyone has a better idea about how to ground the multimedia speakers, please suggest it. I could buy some wire from ebay...

Reply 56 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t
h-a-l-9000 wrote:

Alligator clips

Thanks for your suggestion. I have searched on ebay, and seen these items with various "properties" associated with them. These properties are "5 amps" and "insulated". Do I require insulated alligator clips with 5 amps, for example? Thanks.

Reply 58 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t
h-a-l-9000 wrote:

Thanks a lot! I really appreciate you finding that auction. Please note that the economy delivery on that item will take several days, but I will order it anyway because they are useful and cheap, but I have an idea that I could try right now. What about this idea?

In the photo, there is a label marked "1". This is where a bent paperclip is wrapped around the ground sleeve section of a multimedia speaker plug. The paperclip is held fairly tightly in this section, because I have wrapped sellotape around the speaker plug, where some of the bent paperclip is resting. So, the bent paperclip cannot move out of this ground sleeve position.

The other end of the bent paperclip goes in to the black multimeter cable socket - the area which usually plugs in to the multimeter device. BTW, the other end of the black multimeter cable, where the probe is, will be plugged in to the black wire ground section of a molex cable. Will this idea work OK? If so, I could try it right now! 😀

speak2.jpg

Reply 59 of 85, by retro games 100

User metadata
Rank l33t
Rank
l33t

I haven't tried the "paperclip method" mentioned in my previous post, because I don't know if a) it's safe and b) it will work, so I ordered some crocodile clips/leads from ebay, but with a faster shipping method, so I should receive them by Monday..