Odd 600xl issue

[seastalker]

Posted in the wrong spot before- my apologies.

Just got two fully socketed 600xls. Using a new power supply, one 600 is not working and I swapped all chips out (and thus reseated). The good 600 found three bad chips from the bad: ANTIC (CO21697), CPU Sally 6502C CO14806 and the CO60472-D (Delay). After putting them aside, I put the GOOD versions from the working board into the bad hoping I'd just need the three chips to fix it. No dice. Now, it powers up to an odd screen (see pix) I was hoping not to desolder willy nilly.

*I followed the Black screen/Red screen guides from http://atari.boards.net/thread/893/black-screen-problems-fix-checklist so all steps from Star Raiders to channel 2-3 moving tried.

Anyone know my next troubleshooting steps to try with that screen on a fully socketed 600xl with all tested and working chips? Does knowing those three chips failed reveal anything from the board's unknown backstory? Bad PSU? Loose socket? Bad crystal?

[Rybags]

Hard to say. Perfect vertical stripes are usually PM graphics and are usually random and disappear in under a second on a normally working machine.

Does anything happen if you hit Reset? Do you get a click through the audio at powerup and reset?

 

I had a quick look at the linked guide - sort of generic. The info on "diag mode" carts isn't right. Just because a cartridge flags as diag like Star Raiders, Asteroids and a few other old games - doesn't mean that they necessarily do any sort of diagnostics. Generally they do a very minimal init of the system. The reason they're more likely to work is that they bypass the XL memory test and rely less on the OS Rom working so are more likely to function on a faulty machine.

 

But, they are a good way (along with other carts) to test something like your problem. If behaviour is different it can be an encouraging sign that at least the CPU is probably OK.

Edited June 6, 2018 by Rybags

[flashjazzcat]

That misinformation regarding diagnostic carts is corrected later in the thread, but you have to go looking for it.

 

Worth having a close look at the machine sockets, since I had one with a break on one of the OS ROM pins as well as a bad delay line IC. The latter seems to be a pretty common fault on the 600XL, so in itself isn't significant of much aside from age.

[seastalker]

UPDATE: Well, my working 600xl I posted in another thread I buggered (only temporary I hope) doing a 64k upgrade. HOWEVER, my original buggered one I now have working! Anyone care for detective work? Based on my OP, I put an Antonia in on the chance it may bypass part of the board where one of flashjazzcat's mentioned problems may be and it worked! Plus I didn't have the ebay a sally CPU in the process. Needless to say, I 'borrowed' the antic and the CO60472-D (Delay) from the working (formerly til today) and it fired up right away! A twist of the color pot and purple went to blue!

 

So, any tech detectives knowing Antonia specs care to guess what my underlying issue is and what caused it?

 

A. ANTIC (CO21697), CPU Sally 6502C CO14806 and the CO60472-D (Delay) WERE BAD.

B. Replacement ICs gave above screen, but Antonia 4MB (replacing CPU and MMU) fixed everything!

 

Does whatever magic the antonia does via bypass reveal what was broken?

[JR>]

Antonia completely replaces all RAM and ROM. Probably all of the memory management chips as well, unless they have some function beyond just memory control.

[seastalker]

I'm actually having a hard time trying to find an image of a 600xl motherboard with all the chips labeled. Has anyone who has stripped down an A8's redundant IC chips listed the IC by number or by picture which I can take out and store away? I'd figure the two ram chips that you WOULD replace when doing the 64 upgrade.

 

EDIT: I may have an answer soon- I am able to at least boot into the ready screen with full sound with the antonia board with almost all of the small chips removed! I'll complete the list with pics but only problem now is holding option at power up does not initiate the SELF TEST. Which IC chip stores that? Then, would I even still need that if Antonia has built in configurations?

Edited July 12, 2018 by seastalker

[+Nezgar]

Here's a quick pic I of an NTSC 600XL mobo i have open. (A few chips were borrowed, and it has a quick hack audio/luma monitor port. ) I labelled the main component locations, and put squares around the chips I know can be removed with Antonia present. (Maybe more can be removed?)

[+DrVenkman]

http://www.atarimania.com/documents/Atari_600XL_Computer_Field_Service_Manual_Rev_1.pdf

[seastalker]

Thank you DrVenkman for the manual. I did find it prior and have been reading it. Now I hope to find EXPECTED multimeter number values since I'm working on two boards with problems without a perfect 600xl. Also, please see the attached pic - I added to Nezgar's labeled white boxes with my own unlabeled ones. I can confirm the Atari still goes to boot screen with ALL white boxed chips removed! That is not to say that is ideal or not yet as I DID notice that when I hold option it no longer boots to the Self Test. If the Self Test is stored (likely) in either the BASIC ROM or OS ROM chip, I suspect that would be a natural side effect and would guess I'd have to configure Antonia's settings which would include that. With no Antonia manual, I have yet to research that.

 

That said, it is AMAZING to me how much of the A8 can be removed and still boot with Antonia! THAT it also replaces my need to get a new CPU or DELAY is also fantastic... AND... remember that "bad" ANTIC? I learned never throw ICs away as they may be misdiagnosed. Though a working 600xl did not accept it, for whatever reason, putting the original 'bad' ANTIC back in actually works now (at least to get to the boot screen)! Maybe Antonia bypasses some ANTIC pins or features and only used 'just enough' ANTIC to work?

 

I hope Simius sees this thread as he would be the authority on Antonia to answer this.

 

I also need to test joyports and SIOtoPC functionality, but with all the above chips removed, I'm happy for anyone's ideas of other tests to run so I can confirm these ICs, though nice to save in protective anti-static storage, removing them won't hinder functionality beyond what can be observed by a simple boot up screen.

Edited July 12, 2018 by seastalker

[JR>]

You need to put the 2 4051's (far right ) back. They are for your keyboard and would explain why the Option key isn't working!

[seastalker]

... AH! There's that SELF TEST now! :)

[seastalker]

So, after putting the 2 4051's (far right ) back as suggested, the SELF TEST loads and passed all tests. Any way to conclusively test if the other 6 chips I white boxed are needed or made redundant by Antonia?

[JR>]

I'd start by testing all the ports....Cartridge, SIO, Joysticks, PBI too if you have any such devices. Extended memory checkers.

[BillC]

You need to put the 2 4051's (far right ) back. They are for your keyboard and would explain why the Option key isn't working!

The START/SELECT/OPTION keys don't connect to POKEY thru the 4051s, they connect to CTIA/GTIA thru 470 ohm resistors.

[_The Doctor__]

well if it started working then it it brings the idea that there is or was a short or open on the board and when putting the chips back in the problem was correct... if moving stuff around does not re create the key problems... whatever was wrong has been dislodged and or a re flow corrected it. I suggest making sure to vacuum or empty out any loose material that might be floating around in the machine...

[JR>]

The START/SELECT/OPTION keys don't connect to POKEY thru the 4051s, they connect to CTIA/GTIA thru 470 ohm resistors.

 

Yeah.....I realized that after I responded, but they needed to go back in regardless. Not really sure what happened with the option key, but all's well that ends well.

[seastalker]

@JR> I'm always a curious one - Do I have this right? You said they need to go back regardless: So they STILL control the keyboard other than the special keys (START SELECT OPTION) which are handled by the 470 ohm resistors?

[JR>]

Yes, they decode the keyboard matrix for all of the keys except reset, option, select and start.

[seastalker]

Wondering if the U1MB also replaces the need for the same chips as the Antonia does?

[flashjazzcat]

Wondering if the U1MB also replaces the need for the same chips as the Antonia does?

No. The Ultimate 1MB replaces only the MMU and OS ROM. 64KB is required and all the corresponding logic on the motherboard is still required to drive that.

[seastalker]

Ok, then this thread is great in that though a U1MB is better in functionality and expandability, the Antonia, though 3mb more than the Ultimate 1mb that you will rarely utilize, it is a quick and easy solderless ram upgrade for A8 machines, AND a saving grace to bring MANY boards back to life from it's incredible bypassing abilities of several ram and rom chips. I really am happy, as this shows a great market for BOTH products!!

[flashjazzcat]

Antonia has a 65C816 as well, although this may introduce corner-case software incompatibilities (unpatched MYDOS and any games using illegal opcodes, etc). Regarding the 'solderless' aspect: I believe Antonia requires a socketed MMU and CPU, so on many XLs and all XEs, you have more work to do than you would for an U1MB on the same machine (desoldering and socketing a 40 pin and 20 pin IC for Antonia; desoldering and socketing a 28 pin and 20 pin IC for U1MB, plus four jumper wires).

[seastalker]

Ok, very good point too! If you have a fully socketed XL or XE, It is good to have an Antonia on standby if your machine ever fails. I just think this added benefit should get higher recognition for those that may just want to fix a board that otherwise won't work without it as a SIMPLE fix.