Distorted voice - which component to replace?

[dvdborn]

Hi all, and especially Dr. C.  ,

My EIII has one voice that sounds like it has a bit reducer effect on it.

You can listen to it here:
http://apple.phl.be/~dvdborn/EIII/EIII-voice-problem.mp3.

First you'll hear 6 sine tones with different pitches. The first tone is from a voice channel that sounds OK, the second tone is the faulty voice channel. After the 6 sine tones, you'll hear a piano sound. The left channel has the faulty voice, where you'll hear distinctly the bit crushing effect.

Here's what I already tried to analyse and solve the problem (without any succes)

- I know that the faulty voice is voice #9. If I switch my 2 voice cards, the faulty voice becomes voice #1.
- I replaced the CEM3387 filter chip, but that didn't solve the problem (as I suspected)
- I replaced the L1A2538 - TAA8743 chip. That's the long one. Again without any improvement.

I do have a space voice card with some working voice channels on it. So now my question is. Which component is responsible for the distorted voice?

Since all these chips are soldered onto the board I want to be sure that I start to unsolder the correct chip from both voice cards.

I'm guessing it's the PCM53JP-I Burr-Brown chip (the D/A-converter if I'm not mistaken). Or can it be one of the smaller IC's?:
- NE5532N QFP0030 8725VD
- M8718 LF 356N
- ???CD74HCT4053E ???CA H 737 - I can't read the first letters from this chip because there's another component glued on top of it.

Cheers,
David

[dr.c]

Sorry, but there are no six tones !

Please, make a file with :
one time the piano with working channel
one time the piano with faulty channel, same musical note !

Please make mono file !

[dr.c]

I would immediately change the 4053. These series : 4051, 4052, 4053, are analog multiplexers, and they are a pain, a real pain. For the price it costs, its not a great loss if you give it a try and change it for nothing.

[dvdborn]

Hi Dr.c,

Here's a recording made with the faulty voice channel:
http://kunst.phl.be/~dvdborn/EIII/Faulty-voice.wav

And here are the same sounds recorded with a good working voice channel:
http://kunst.phl.be/~dvdborn/EIII/correct-voice.wav

Both files are in .wav format, 16 bit and in mono.
They'll start with 3 sine tones at different pitches followed by some piano notes.

[dr.c]

I have to take these files to a place where I can listen in good condition. I am working here on my mac, so...

[softflow]

Hi all,
I actually do have exactly the same issue with one voice that sounds distorted and is panned to only one side.

Have you found a reason for that behaviour?

Cheers
Ingo

[dvdborn]

Hi Softflow,

I haven't had the time yet to start soldering my voice cards and replacing component. So at the moment I still have the 'bit crushed' voice.

I don't have a problem with one voice being panned to only one side. I'm guessing that the other side in your case is the faulty voice that isn't working.

David

[dr.c]

Softflow : can you pan it if you want, or not ?

[dr.c]

I don't have any channel board diagram.

Send to :

dr.c@keyboardsrecording.fr

[dvdborn]

After 6 hours of debugging, desoldering and soldering I have finally fixed my distorted voice.

I changed one component at a time so that I could keep track of what effect it would have on the distorted voice.

1. I replaced the Burr-Brown chip with another one from a space voice card: no change
2. I replaced the 4053 chip: no change
3. Switched the SSM2030's over to see if another voice would be affected: no change
4. replaced the Emu chip: no change
5. Replaced the Burr-Brown chip with a second one. I heared a difference! The voice distorted more :-(
6. Replaced the Burr-Brown chip with a third one. The distortion was gone. I'm now back at using 16 clean voices.

The Burr-Brown chips that I used all came from a spare voice card. There were already some of them replaced on this spare card. So I'm guessing that this chip often fails.

Why did it take 6 hours? The voice card I had to service was the bottom one in the card cage. So each time I had to disassemble and reassemble the whole card cage. Definitely not funny. But in the end it paid off.

David

[dr.c]

You would have to recalibrate your machine now.
You would need the UST diskette.

Does anyone has such a diskette ?

[dvdborn]

Hi Dr. C,

I do have the UST disk. EIII has already been recalibrated.

David

[dr.c]

This is great !

Congratulations !

[dlmorley]

After 6 hours of debugging, desoldering and soldering I have finally fixed my distorted voice.

I changed one component at a time so that I could keep track of what effect it would have on the distorted voice.

1. I replaced the Burr-Brown chip with another one from a space voice card: no change
2. I replaced the 4053 chip: no change
3. Switched the SSM2030's over to see if another voice would be affected: no change
4. replaced the Emu chip: no change
5. Replaced the Burr-Brown chip with a second one. I heared a difference! The voice distorted more :-(
6. Replaced the Burr-Brown chip with a third one. The distortion was gone. I'm now back at using 16 clean voices.

The Burr-Brown chips that I used all came from a spare voice card. There were already some of them replaced on this spare card. So I'm guessing that this chip often fails.

Why did it take 6 hours? The voice card I had to service was the bottom one in the card cage. So each time I had to disassemble and reassemble the whole card cage. Definitely not funny. But in the end it paid off.

David

I'm really glad you sorted it out as you can come and fix mine now 

[dr.c]

What's wrong with it ?