Sound problems and more.

[Viserion]

Recently got an Amiga 1200 that I've been struggling to install OS3.9 on.

After finally gotten the leads I needed to finish the job, I fired up a game, and the sound was disappointing to say the least.

So I opened it up, in search of leaky capacitors, and found that someone has been tinkering arund with it.

On the attachment I have marked in red where I think there may have been tinkering or parts missing.

Particularly the wire that goes from bottom to top of the motherboard I doubt is stock.

Any idea what that does?

There isn't any obvious leak of capacitors, but then my eye is untrained.

It works fine, so the only thing I want to do is get the sound working again.

Sadly I've not much experience in this., so any pointers would be good.

[demolition]

I think that wire is a factory fix on that model, and I don't see any obvious signs of modifications on it.

Regarding the sound, how do you hear the audio problems? Is it distorted, does it have noise on it and does it have the same problem in both channels or is it mainly in one channel?

If it's the same in both, it could be that Vref is wrong. If it's different, the audio caps might need replacing as well.
Check the two aluminium caps just above the keyboard connector. They are the audio output DC blocking caps. The solder pads on them should be shiny, so if they're not, it is a sign of leaking.

[Viserion]

That was a suprise to me. I was certain it was a mod.

The right side was better than the left. Only a little distortion. The left however was just noise.

Took a look at the caps you mentioned, and they are dull. The two under the keyboard connector also look a bit dull, but the rest looks ok.

I need to order some capacitors then. Never done that before, so any tips are welcome.

No idea what Vref is, but as it is so late I will google it tomorrow.

[demolition]

The lower two caps that you mention are the decoupling caps for the audio opamp.

You can download a schematic for the Amiga here.

You probably have a bad connection between the output audio caps and the output of the opamp, so before you solder ind the new caps, check for continuity between the cap + pins and pins 1 and 7 on the opamp (U15). If they are both good, check between the cap - pad and the RCA center pin. It should be 1 kOhm (check the schematic).

Regarding where to buy the caps, I usually get them from a Chinese ebay seller. Postage is usually free, but you'll have to be patient as it may take up to a month before you receive your order. If you're in a hurry, there's some EU sellers as well. Use the words 'SMD', 'electrolytic', 'aluminium' or variants.

Edit: Here's a picture of the 1D1 motherboard and as you can see, the wire is standard.

[Loedown]

I use 105 degree type capacitors and they're all Jamicon. With the 22uF 16V caps in the audio path you replace C324 / C334 with non polarized parts, finding SMD caps aren't that easy ( or they weren't in the past ) so I use through hole parts and modify them to fit onto the pads where the old ones were.

If you're missing audio on one side only then there's a good chance that one of the tracks under the LF347 has been chewed away or a via and the part will need to be lifted off the board and cleaned, the tracks underneath it repaired and then refitted.

The normal process for sound diagnosis is to follow the audio path using an oscilloscope and working out where it has failed and also by doing resistive checks between one side of the audio and the other. If you're all new to this and you try to repair it thinking you will save a few dollars I can promise you it's not an amateur repair, you will screw the board up and you will pay through the nose to have it repaired professionally after you've made a real mess of the repair. If you do manage to get it fixed I can also say that the chance of it not breaking down again in the near future would be remote at best.

Cosmos used to sell sets of Amiga caps here, drop him a PM if you're still going to have a crack at it.

[Viserion]

Thank you demolition. Have downloaded the schematics, and ordered the SMD caps.

I used the post from e5frog on this thread http://eab.abime.net/showthread.php?t=44138 . To find what I needed to buy. I will study the schematics and do the measurements you suggested, and practise on old unused hardware.

Loedown: As you see from the link above I found the SMD caps. I ordered a bit more than I need.

I'm fully aware that this probably will go very wrong, but this is a unit a bought on ebay about 2 months ago. So I don't have the same attachment I do to the old A500 whos floppy drive look to be shot.

Whether I will buy a new unit, go for a more modern solution, or give up on Amigas for now, we will have to wait and see.

It would be a pithy to ruin one of these great machines, but no catastrophe. At least not for me.

Labor costs in my country is so high that I strongly doubt it would pay for it self to do this professionally anyway. It didn't cross my mind, so I will search around, and see if I can find a shop that still do this kind of repairs and doesn't charge an arm or a leg for it.

[Viserion]

Forgot to mention that on the lefdt channel there is simple noise. On the right you can here what it's supposed to be, but it isn't quite right.

I don't have an oscilloscope either, so that is out of the question.

[demolition]

Originally Posted by Viserion

Forgot to mention that on the lefdt channel there is simple noise. On the right you can here what it's supposed to be, but it isn't quite right.

I don't have an oscilloscope either, so that is out of the question.

An oscilloscope makes fault-finding quite a lot easier, but by replacing the electrolytic caps and ohming out the connections, you might be able to find the problems without too much trouble anyway.

AVref is found on pins 10 and 12 on U15 and should measure around 2.5 V (Vcc/2). In my A1200, it was just 200mV, causing significant distortion in both channels. This was due to C305 being shorted and replacing it fixed the problem. It's not common for ceramic caps to malfunction, but it can happen.

[Viserion]

Have not been feeling well lately, so haven't done much.

Having trouble to "transfer" the schematic to the real mother board. I can't even find U15, after having looked for it, for ages.

[demolition]

Originally Posted by Viserion

Have not been feeling well lately, so haven't done much.

Having trouble to "transfer" the schematic to the real mother board. I can't even find U15, after having looked for it, for ages.

U15 is the 14-pin IC right below the keyboard ribbon connector. It doesn't look like they could fit 'U15' anywhere on the board, but the chip says LF347 which is a quad op-amp.

[Viserion]

Wel that was a relief. Was starting to doubt my eyesight there

[midway]

I have a similar problem in an a4000, the 403 should measure 2.5v on all but pin 4 and 11 but i get only 0.6 v. Vcc and vee are fine. The one who had the board before me said he replaced the caps...
It looks clean but some solderpads of the 402 and elsewhere look dulled.

The psu is okay. When connecting cdrom out to a4000 audio in or the delfina sound is clear...

I like to listen to mods and watch demos, but it does not sound good at the moment. My 2000 still sounds good, when comparing both audio outputs

Any ideas?

[Loedown]

Originally Posted by midway

I have a similar problem in an a4000, the 403 should measure 2.5v on all but pin 4 and 11 but i get only 0.6 v. Vcc and vee are fine. The one who had the board before me said he replaced the caps...
It looks clean but some solderpads of the 402 and elsewhere look dulled.

The psu is okay. When connecting cdrom out to a4000 audio in or the delfina sound is clear...

I like to listen to mods and watch demos, but it does not sound good at the moment. My 2000 still sounds good, when comparing both audio outputs

Any ideas?

It's obvious that the board has had some leakage on it, vast majority of problems occur under and around the LF347 and LM833 areas of the board, get a cheap multimeter and start buzzing out the connections and checking resistances.

[Viserion]

A little update on my problem. After finally getting the parts I ordered. I chickened out, and is about to send it to Amigakit for repairs. Annoying to have to do so, but I would have ruined it.

Next prokect that I will do my self is to see if it's possible to get my old joystick to work again. It's an old Quickjoy TopStar where the fire button works, but any movement don't.

[Loedown]

Originally Posted by Viserion

A little update on my problem. After finally getting the parts I ordered. I chickened out, and is about to send it to Amigakit for repairs. Annoying to have to do so, but I would have ruined it.

Next prokect that I will do my self is to see if it's possible to get my old joystick to work again. It's an old Quickjoy TopStar where the fire button works, but any movement don't.

A much smarter move, it's painful to have to pay for the repairs but a good technician will always do a better job than a novice.

In regard to the joystick, if no movement is registered then you most likely have a broken wire on the movement switch common line.

[midway]

Hope they can figure it out! How much do they charge just for my curiosity

I will map out the resistors and maybe c403 has been damaged over time...

Would that lead to low voltage(about 25% of its nominal value over all pins except 7 and 11 of u403 and 25% of the 1/-1v of the u402....

is paula putting the r/l audio out at 2.5 v or is this being upconverted on its way to the opamps

I am not an electroengineer just getting into reading schematics and the basics, but i liked physics and electricity in high school (sometimes)

[demolition]

Originally Posted by midway

is paula putting the r/l audio out at 2.5 v or is this being upconverted on its way to the opamps

Paula is outputting a signal centered around 2.5V (PWM using Gnd and Vcc) which is why it needs dc-blocking caps.

[Viserion]

Opened up the Joystick, and indeed one wire was teared straight of. Soldiered it on again, and tested it.

It sort of is a bit better and worse at the same time. Better, because I got movement, worse in that it doesn't last long before nothing works, including the fire buttons.

So will have to tear it apart again tomorrow.