Trying to fix an Atari XLF 800 Freddie

[Beeblebrox]

3 hours ago, ataridns said:

I attached the color bars image , notice that when I adjust the white variable resistor pot, the colors are changing and getting into BW.

Regards the colour pot.. That's normal. The colour pot shifts between b/w and the color. 

Edited April 13 by Beeblebrox

[ataridns]

Thank you Beeblebrox, yes, that’s true. There are signs of life in this board. Some parts are working as they should, but something is stopping it.

Poking around last weekend, during a quick search, I found that apart from the correct frequencies, the data signals exist in the address bus and the data buses.

However, at D7, the signal is very strange, unlike the other data: it has high voltage and strange pulse.

Additionally, the CPU provides a correct oscillogram, as per the SAMS PDF.

 

That’s all I’ve found for now. 

 

I must decode how the video signals are created and why they are missing finally and which IC/part is responsible for that.

Also, I tried to find if there is activity at ROMS read/write: there is at the beginning, but then all data signals are frozen.

 

If someone has any idea what to search for, please help.

 

Edited April 15 by ataridns

[TZJB]

On 4/13/2024 at 11:32 AM, ataridns said:

Crystals were replaced, and every frequency at the involved chips is now okay, as TZJB very nicely explained.

I also changed all capacitors except for the three non-polar ones (C19, C20, C22).

Additionally, L3 was removed, but now I’m facing a black screen.

Occasionally, I see a screen with horizontal color bars, but this is very rare.

The voltage levels have improved slightly to 4.83V in FREDDIE and 4.85V at the input.

 

Now, assuming that the volts, reset, and frequencies are okay, I need to search for signals.

I suspect a bad ROM, but I tried using a Super Salt cartridge with no results.

 

* when powering on, for milliseconds, a red screen appears and then black. 

 

 

 

 

 

 

Now that you have shorted out the L3 component, the FREDDIE voltage at 4.83V is lower than I would expect, as it should be connected directly to the 5V supply on the +5B rail and should be very close to 5V.

 

 

Please run through and post your readings of the voltages that I posted from my 800XLF.

 

J7 (DIN INPUT)- 5.00V

C2 & C3 - 5.00V

FREDDIE - 4.99V

ANTIC - 4.98V

SALLY - 4.98V

GTIA - 4.98V

POKEY - 4.95V

PIA - 4.99V

OS ROM - 4.97V

BASIC - 4.97V

DRAM - 4.90V

 

DRAM has the lower voltage due to the L4 10uH inductor still in circuit.

 

50 minutes ago, ataridns said:

Thank you Beeblebrox, yes, that’s true. There are signs of life in this board. Some parts are working as they should, but something is stopping it.

Poking around last weekend, during a quick search, I found that apart from the correct frequencies, the data signals exist in the address bus and the data buses.

However, at D7, the signal is very strange, unlike the other data: it has high voltage and strange pulse.

Additionally, the CPU provides a correct oscillogram, as per the SAMS PDF.

 

That’s all I’ve found for now. 

 

I must decode how the video signals are created and why they are missing finally and which IC/part is responsible for that.

Also, I tried to find if there is activity at ROMS read/write: there is at the beginning, but then all data signals are frozen.

 

If someone has any idea what to search for, please help.

 

 

Please define the strange signal at D7.

 

D7 appears at every major chip so could be caused by any major chip, or a contact with something else on D7 somewhere on the board.

 

For clarity does the D7 strange signal appear at every chip?

 

SALLY pin 26

ANTIC pin 37

GTIA pin 34

PIA pin 26

POKEY pin 6

OS ROM pin 19

BASIC pin 17

J9 pin 28

 

 

[ataridns]

I apologize for my lateness; life has its priorities.

Thank you, TZJB, for your help. I will complete all these measurements and post my photos.

Please bear with me as I have several other obligations demanding my attention.
 

To achieve the best results, I plan to use the highest-quality 5V power supplies I have.

Additionally, I came across an original ATARI 800 XL, and I also intend to use a Meanwell medical-rated PSU.
 

So to have best results I am planing to use the very best 5v power supplies I have:

also I found an original ATARI 800 XL psu, but I also want to use and a meanwell medical rating psu. 

 

I will post my results ASAP. 

 

[ataridns]

 

After I found an original Atari 800XL PSU, I checked it and plugged it in. Its output is 5.1V, and on the board, it’s 4.99V and in Freddie 4,98V.

So AGAIN after the comments of TZJB ... 

On 4/16/2024 at 1:51 AM, TZJB said:

Now that you have shorted out the L3 component, the FREDDIE voltage at 4.83V is lower than I would expect,

my thoughts turn to the awful soldered DRAM socket. The person who tampered with this board using incorrect crystals likely affected the socket as well.

I removed the socket, I cleaned very carefully the pads and replaced it with a new one. I test the machine with the new empty socket and black screen again. 

 

But my hunch proved correct!

After inserting a new KM4164 DRAM chip, the “Ready” message appeared with a distorted video, but with a “Ready” !

Yes, sirs! The board has come back to life!

 

   

 

There are some more faulty DRAM chips that need to be replaced, and I hope the distortion of the screen will disappear. 

 

          

 

So, THANK YOU AGAIN, TZJB, FOR ALL YOUR HELP AND EFFORTS.

I wish to help you in the future as well!

This weekend, I am planning to socket all the DRAMs and identify the guilty component.

 

 

 

 

 

 

 

Edited April 17 by ataridns

[TZJB]

20 hours ago, ataridns said:

 

After I found an original Atari 800XL PSU, I checked it and plugged it in. Its output is 5.1V, and on the board, it’s 4.99V and in Freddie 4,98V.

So AGAIN after the comments of TZJB ... 

my thoughts turn to the awful soldered DRAM socket. The person who tampered with this board using incorrect crystals likely affected the socket as well.

I removed the socket, I cleaned very carefully the pads and replaced it with a new one. I test the machine with the new empty socket and black screen again. 

 

But my hunch proved correct!

After inserting a new KM4164 DRAM chip, the “Ready” message appeared with a distorted video, but with a “Ready” !

Yes, sirs! The board has come back to life!

 

   

 

There are some more faulty DRAM chips that need to be replaced, and I hope the distortion of the screen will disappear. 

 

          

 

So, THANK YOU AGAIN, TZJB, FOR ALL YOUR HELP AND EFFORTS.

I wish to help you in the future as well!

This weekend, I am planning to socket all the DRAMs and identify the guilty component.

 

 

 

 

 

 

 

 

This is looking promising.

 

Atari 800XL PSU are normally somewhere between 5.1V and 5.3V so yours has a normal potential.

 

Good call on the dodgy chip socket. You have made some good progress and now identified faulty DRAM which was similar to some of my issues with six KM4164 going faulty, so it's not just MT DRAM that can randomly expire, it seems Samsung can do it too.

 

Regarding screen distortion, what display and method of connection are you using?

 

[Beeblebrox]

Might also be worth looking at one of the Q transistors (2N3904), as IIRC a faulty one in the video circuit can cause distortion like this. 

 

But definitely get the ram and or their sockets sorted as a first port of call.