1200XL has audible "buzz" on TV. What might be the source?

[ACML]

I've got better photos and here's a recap of what I modified.

 

1) I converted to 28 pin OS and have the 600/800XL and Omniview XE operating systems burned on a pair of 27128 EPROMS.

2) I use the Channel 3/4 selector switch to select which OS to boot using Pin 26 (hi/lo) on the 27128s.

3) I installed the ClearPic2002 and the video is awesome.

4) I replaced the 74LS14 with a 74AC14 and replace C7 with 15 uF capacitor to remedy the cold start issue.

5) R63 jumper.

6) Removed the RF modulator and L11 (noise was present before modulator removal).

 

I have done all six multiple times on previous projects.  Very reliable.

 

Key:

GOLD = 28 pin OS mod

BLUE = RESET fix modification

PINK = RF modulator removal

GREEN = R63 +5VDC fix

RED = ClearPic2002 mod

 

Edited February 20, 2023 by ACML

[_The Doctor__]

Okay this is more helpful,

First things first, get some IPA and an old tooth brush and some swabs, clean up all of the flux, debri, and other stuff you see.

You can use a meter to make sure there are no opens, shorts, or oddball readings tracing out the modifications connections. I would start in the clearpic and audio sections moving out after that. Re examine with a magnifier after that. We are looking for shorts, breaks(open), double check components are correct including orientation. Any components that look bent and stressed need to be checked they could be damaged/fractured inside. Check Amplifier GTIA section for marginal components. That's the connections responsible for keyclick/beeper music

Edited February 20, 2023 by _The Doctor__

[ACML]

29 minutes ago, _The Doctor__ said:

We are looking for short breaks, double check components are correct including orientation.

I know what a short is, but what is a "short break"? Not familiar with that term.

[+bob1200xl]

Audio comes from multiple sources in the Atari, SIO, POKEY, and the like. Try each source by itself and see if any of them do not work. That will probably be the source.

 

Bob

 

[ACML]

_The Doctor_,

     I cleaned up the flux with 91% Isopropyl and no change.  I looked over all the solder joints with magnifier and everything looks good. 

 

Bob,

    I ran the self test audio and all four channels work fine.  The noise can be heard without anything connected to the SIO port.  Also, whenever there is a cartridge installed, the noise is twice as loud.  The noise is there plugging in a BASIC cartridge at READY prompt. Any cartridge causes the noise to be louder.

 

Audio is normal in games except for the constant hum/buzz.

 

Edited February 20, 2023 by ACML

[Keatah]

Can you show a waveform of the noise both with and without a cartridge?

[zylon]

I just repaired a 2600 6er, that had same noise. It would overpower the audio at times. I replaced the two circled caps (purple). and the 4050(purple). Once those were changed, it cleared right up. There was a sort of internal shorting, inside the 4050.

Edited February 20, 2023 by zylon

[_The Doctor__]

On 2/4/2023 at 3:30 PM, _The Doctor__ said:

Grounding, filter caps, maybe 4050.  I'd scrutinize the Power Supply section.

1 hour ago, _The Doctor__ said:

Okay this is more helpful,

First things first, get some IPA and an old tooth brush and some swabs, clean up all of the flux, debri, and other stuff you see.

You can use a meter to make sure there are no opens, shorts, or oddball readings tracing out the modifications connections. I would start in the clearpic and audio sections moving out after that. Re examine with a magnifier after that. We are looking for shorts, breaks(open), double check components are correct including orientation. Any components that look bent and stressed need to be checked they could be damaged/fractured inside. Check Amplifier GTIA section for marginal components. That's the connections responsible for keyclick/beeper music

24 minutes ago, zylon said:

I just repaired a 2600 6er, that had same noise. It would overpower the audio at times. I replaced the two circled caps (purple). and the 4050(purple). Once those were changed, it cleared right up. There was a sort of internal shorting, inside the 4050.

that's reasonable as it touches the GTIA and is in line with earlier guidance about capacitors, components and audio/video, also all are located in proximity to where work was done.

Edited February 20, 2023 by _The Doctor__

[ACML]

Swapped out the 4050 and no change.

 

[Keatah]

Can you show a waveform of the sounds, both with and without the cart? From a point where the audio is leaving the PCB.

[ACML]

46 minutes ago, Keatah said:

Can you show a waveform of the noise both with and without a cartridge?

I don't have a scope.

[ACML]

50 minutes ago, zylon said:

I just repaired a 2600 6er, that had same noise. It would overpower the audio at times. I replaced the two circled caps (purple). and the 4050(purple). Once those were changed, it cleared right up. There was a sort of internal shorting, inside the 4050.

 

The 4050 has been ruled out.  That leaves C96 and I'm guessing C97 as they look the same.  They are 820 pF (25v) capacitors (C061336-03).  Are these part of the sound circuit?

 

I guess the presence of the cartridge effect is the biggest clue.  Wouldn't that typically be a ground issue?  I'm an Aero and am definitely out of my depth wrt electrons. 

 

Edited February 20, 2023 by ACML

[Keatah]

Oh ok. With a 'scope and cart/no-cart I'd be able to tell you if it's in the power supply section or logic section.

[zylon]

10 minutes ago, ACML said:

The 4050 has been ruled out.  That leaves C96 and I'm guessing C97 as they look the same.  They are 820 pF (25v) capacitors (C061336-03).  Are these part of the sound circuit?

 

I guess the presence of the cartridge effect is the biggest clue.  Wouldn't that typically be a ground issue?

 

yes, those silvery caps are part of the sound circuit, and often fail. A couple things you can try, are pressing down on them with pencil eraser. If sound changes, then you have a clue. You can also lift one leg of each, and determine what side of them, the issue might be on. Something bleeding into the sound, can be before or after them.

I work on lots of 5200 systems, and used to have a 1200xl. Those components do the same job, in same way for all their earlier models. I had a 4sw 2600 in, that had those caps fail. the thing used to pick up AM radio, through the bank of resistors nearby. Depending where I touched it, you got different stations, lol

Edited February 20, 2023 by zylon

[ACML]

I really appreciate everyone's input.  Thank You.  I did call my older brother, he's a EE.  While talking to him about my issue, it struck me that when I acquired the board, the previous owner appears to have tested or replaced the really large cap downstream of the rectifier bridge. My brother said that if part of the rectifier failed, it could cause a ripple in power that could be the culprit (AC bleeding to cap).  The addition of the cartridge would increase the load (i.e. voltage drop) and exacerbate the volume of the noise.  Just a theory:  Maybe the previous owner knew it had this issue and removed the big cap to test it and returned it once it tested good.  That would explain why there is flux on the three posts of the big cap.  My brother recommended that I test for both AC and the DC voltage getting to the big cap.  If the rectifier is working good, the AC voltage should be just mili-volts and if its damaged, you might read a few AC volts, meaning it's bleeding AC into the system.  If I get the chance, I will check voltages tonight.

 

[_The Doctor__]

that sounds like scrutinizing the power supply. looks like you've got plenty of ideas now.

[zylon]

7 minutes ago, _The Doctor__ said:

that sounds like scrutinizing the power supply. looks like you've got plenty of ideas now.

definitely sounds like something, not dead shorted, but leaking enough, that increases when power demand goes up. ie- using a cartridge vs sitting idle. I'd be curious as to whether the sound is also manifesting as a video effect onscreen. Such as the light diagonal wave, when an "ingot" is about to kill.

[Keatah]

Sounds to me like something is not able to smooth something out. Or something is generating noise above what the original parts can (now) filter out.

[ACML]

UPDATE:  I tested the rectifier and it appears to be working good.  Getting 11.3 to 11.9 VDC to large 10,000uF capacitor.  Also checked the diode bridge with known good board and they act similar.  The two voltage regulators are both putting out 5.03VDC.  So, the power to 5VDC checks out as much as I can check with a multi-meter (no scope).  No smoking gun.  I've looked over the board with magnification and can see no signs of heat damage to board or components.  I even tried compressed air as a last resort (maybe small solder flake under socket).  Hate to say it, but this one has pegged my fun meter.  I've got other things to do and this board has me saying "UNCLE"!  I'm going to put it on the shelf for a few days and if I get curious enough, I'll have another go, but right now this board maybe headed for salvage (POKEY, PIA, SALLY, ANTIC, GTIA, RAM, MMU, TTL chips).

 

 

 

 

[ACML]

I went ahead and changed the big 10,000uF cap next to the rectifier to see if it cleaned up the noise, but it did not make any difference.

[BillC]

On 2/20/2023 at 2:36 PM, ACML said:

The 4050 has been ruled out.  That leaves C96 and I'm guessing C97 as they look the same.  They are 820 pF (25v) capacitors (C061336-03).  Are these part of the sound circuit?

I believe they are part of the circuit to create the audio subcarrier for the RF modulator, since you have already removed the RF modulator I would suggest disabling all audio signals to it.

 

If the Sobola schematic is correct the following modifications are suggested:

Removing R139 should remove power from the collector of Q12.(audio signal to RF modulator)

Removing R142 should remove power from the base of Q13.(audio signal subcarrier for RF modulator)

Edited February 22, 2023 by BillC

[+kheller2]

Unrelated to any of this, but interesting is the March 7, 1984B sticker on the bottom for a board made in early 83.  

[Mathy]

Hello Bill

 

12 hours ago, BillC said:

If the Sobola schematic is correct ...

 

Why not use the original schematics from Atari?

 

Sincerely

 

Mathy

 

[ACML]

Looks like in the end it's the quad bridge diode rectifier.  One leg has failed and some AC is bleeding over.  I have a spare, but that R&R is more time consuming than most parts.  I need to unscrew the two 7805 voltage regulators, de-solder those big contacts, removed heat sink which requires breaking one time use plastic rivets.  This will have to wait as life has me busy again.  Thanks AA for all the help!

 

 

[TZJB]

28 minutes ago, ACML said:

Looks like in the end it's the quad bridge diode rectifier.  One leg has failed and some AC is bleeding over.  I have a spare, but that R&R is more time consuming than most parts.  I need to unscrew the two 7805 voltage regulators, de-solder those big contacts, removed heat sink which requires breaking one time use plastic rivets.  This will have to wait as life has me busy again.  Thanks AA for all the help!

 

 

Wow, who would have thought that? It makes sense now that you found it. I'm not sure if it would be feasible, but you could possibly just bridge the broken rectifier with an 1N5401 diode as a substitute for the failed internal diode, then you wouldn't need to remove it.