1040 STE no sound....quest accepted

[Samuraik]

Dear all,
similar to many others I found this site when looking for more information on a problem I have with my lately purchased Atari 1040 STe.

As the title of this post tells you the Atari is all silent. I could now get rid of it and try getting another one but I see this as a quest.

It drives me mad to not know why is this also because everything else seems to work fine.

I've looked through the 208 topics in the HARDWARE ISSUES corner of this forum but related to my problem I could only find "TT no sound" and read into it. There it seems to me that the TT had a bit more problems rather than in my case where everything is working fine except for the audio output.
The TT's PS had good voltages and the LMC1992N had its 8.2v pin good. In my case it's a little bit different (no on-board speaker, no dip switches, no blown off board traces, etc.) and I'm not sure if this complicate things or simplifies them.

But lets have a look on the machine and what I could find until yet:

01

01.jpg (139.29 KiB) Viewed 2718 times

02

01a.jpg (62.02 KiB) Viewed 2718 times

Sysinfo of the machine, 4MB Ram, Blitter on, DMA Sound available, cookie jar has some sound hardw.?

03

03.jpg (91.79 KiB) Viewed 2718 times

From exxos i bought the diagnose cartridge and got me the manual STE_Field_Diagnostics_v19.pdf.

03a

03a.jpg (96.02 KiB) Viewed 2718 times

Unfortunately I had not all of the needed fixtures, no oscilloscope and not all hardware (high-res-screen, hdd).

03b

03b.jpg (98.24 KiB) Viewed 2718 times

So I went through the tests as far as was able to (green = pass, red arrow = lack of fixture/hardware, red cross = fail)

In the audio test I should hear at least the PSG sound sweep from low to high frequency, which isn't hearable.
For the DMA Sound test I would need an oscilloscope, which I don't have but I tried some games....


03c

03c.jpg (119.41 KiB) Viewed 2718 times

03d

03d.jpg (165.34 KiB) Viewed 2718 times

About the games I loaded it is written that they definitely have DMA sound implemented (Xenon 2, Obsession, Dynabuster+)
But all silent still.

03e

03e.jpg (84.44 KiB) Viewed 2718 times

After a bit of research in the web I came across this post and started measuring as advised...

YM2149F IC has its +5V.
LMC1992N IC has +0.3V ?!?!?

so I searched for the LM78L82ACZ transistor and...

03f

03f.jpg (144.27 KiB) Viewed 2718 times

...found it

On the LM78L82ACZ I could measure +12V going in on the right side and the +0.3V leaving on the left side watching on the transistor like in the picture above.

03g

03g.jpg (180.27 KiB) Viewed 2718 times

So I desoldered this thing and ordered a new one from Nikko Electronics. Soldered the new one in and....

Voila! ....same result as before, 12V in & 0.3V out.

Here I am standing now.

20a

20a.jpg (132.51 KiB) Viewed 2718 times

What I truly not understand: The transistor seems to stand pretty much isolated from the rest of the wiring and seem to has only one job = getting 12V giving 8.2V further on.
I mean isn't it irrelevant if everything past that transistor is malfunctioning or is the behavior of this assembly part influenced like "I'll give you only 8.2V if you are asking for from behind".


Advice from the "TT no sound" topic....
1. Replace DACs.
-> Are they relevant to my problem as described?
-> When it comes to desoldering IC's I would clearly have to raise the white flag

2. Check if chips related to sound have 5V and clock.
-> Voltage I can check. With what type of equipment would I check clock? (but clock should be there because of field test 'timing' = pass?)

3. Also, check the power supply. One of the first things to go when my old PSU was on the blink was the sound.
-> On several places on the board I can find +12V and +5V so I think it's okay.
-> Bought another PSU which arrives in some days

4. I would check the ym chip outputs show something on key presses to see if there is at least a source signal...
-> The 1040 STe has no key-click right?
-> If it would have keyclick where could I find / activate this and after it's on especially how to measure this?

5. For troubleshooting get the schematic and trace the sound from the PSG to the output.
-> I 'think' I got all schematics available but have no "storyboard/plan of procedures" to trace whatever.


So that's all and a lot to my little riddle. I would be really thankful for any hints on that.

Yours sincerely
Raik

[exxos]

I suggest to lift the 8.2 volt side of the regulator and see if your 0.3v then rises to 8.2v.

If it does, then with the machine turned off, measure the 8.2v pad on the motherboard to 0v on resistance and see what value you get..

[Samuraik]

Posted on my morning break and already got an answer at lunchtime, wow, you're faster than me.

First thing that I'll try when I get home today!

thanks
Raik

[rubber_jonnie]

Before going any further, can you qualify a few things please.

Firstly, I assume you have connected a pair of speakers or headphones to the phono connectors on the back of the STE and hear nothing through those? If not it is worth checking the solder on the phono ports to be sure that the connections have not become loose in the PCB after many years of plugging in/unplugging.

Secondly, if you are using a SCART lead, check the audio connections are correctly hooked up in that lead by continuity testing with a multi-meter. Not all SCART leads are made equal, so be 100% sure the lead is good.

Lastly, if you are using something other than SCART, like VGA, that requires a separate audio hookup and headphones/speakers either to the audio connector of the video socket or the phono sockets. I can't tell from your picture, but I'd be interested to know.

BTW, for the odd repairs, the whole test harness isn't necessary, you can do plenty with just the diag cart and a multimeter.

For clocks you really need an oscilloscope, but since the machine boots OK and passes the timing tests, I think there is a good chance that the main clocks are fine. As for key clicks, yes the ST range does have key clicks, but ideally you'd need an oscilloscope to check all the sound outputs I think.

I'm not trying to be awkward here just understand your setup and the extent of your testing of the audio outputs and how they are connected, so that I can then try and help some more.

[tzok]

Q500 is not a transistor, but a voltage regulator. If it has 12V on the input, proper GND, and nothing on the output, then something shorts its output to ground.

Sound Hardware seems to be recognized correctly - value of 3 indicates, both Yamaha and DMA are present, so the problem is likely in the analog part. The only chip powered from 8.2V rail is the LMC1992N, so it is likely to be faulty.

[Samuraik]

@rubber_jonnie
First of all thank you for your concerns. I'm glad to notice that someone takes care.

Secondly to your questions...

23

23.jpg (162.09 KiB) Viewed 2618 times

I have connected a pair of speakers to the phono connectors yes. (cable+speaker do great with the amiga 500 cinch-outputs)
I checked the soldering of the phono ports, all fixed well in position.

24

24.jpg (61.35 KiB) Viewed 2618 times

Regarding the 'SCART lead' I've messed that up a bit in the sketch above. I'm not using the TV output. I'm using the Monitor output with a "Monitor to Scart adapter" which are available these days on ebay (its a new manufactured cable). The SCART plugs into an Orion TV (so no VGA is involved). To my understanding the sound (if there would be some) should be hearable parallel on both Monitor connector (pin 1) and the phono ports.



@exxos & tzok

22a

22a.jpg (69.09 KiB) Viewed 2618 times

I lifted the 8.2 volt side of the regulator and then it rises to 8.2v! (I'll get +12.66V between leg3 and leg2 / I'll get 8.25V between leg2 and leg1)

When I turn off the machine and measure between leg2 and 1pad I'll get 0.5 ohm.


So the conclusion is that something shorts the output in the area around the LMC1992N or even the LMC1992N itself?

[exxos]

I lifted the 8.2 volt side of the regulator and then it rises to 8.2v! (I'll get +12.66V between leg3 and leg2 / I'll get 8.25V between leg2 and leg1)

When I turn off the machine and measure between leg2 and 1pad I'll get 0.5 ohm.

So the conclusion is that something shorts the output in the area around the LMC1992N or even the LMC1992N itself?

Something on the 8.2V line is shorting then.. The simple things is to remove the caps as they are easy.. then remove the LM chip...

[Samuraik]

Okay exxos I'll do so and get back here after but I think this will take some time.

I've also ordered a 'new' LMC1992CCN from Littlediode which hasn't arrived yet. But getting off the actual IC will be a really mess. (Saw you on "Desoldering the Falcon NVRAM")

I think I'll need some help from my father but sitting together isn't easy these days, maybe when he has got his immunisation or I can get me some rapid test to be able to visit him.

thank you nevertheless
Raik