SOLVED !! not a happy ql , but what causes it ??

[techy]

i have looked closely at the boot sequence of a good working ql ,
and there you can see the same pattern just before it goes to the colorfull pixel-screen .

it seems my defective ql is stuck in the first phase , and does not continue to the colerfull pixel-screen .


Any idea's ??

[bwinkel67]

Sorry, this will be of no help. I'm just surprised that the black and white bars appear before the random pattern. I have never noticed that before. I will give see, next time I'm on my QL if I can spot it...thanks.

[techy]

this is how a boot looks on a good functioning ql ,

the stripy pattern is just visible before the color-pixely screen takes over .

https://streamable.com/xx83r0

[Popopo]

Hi!
Just to give you some points to consider, and perhaps they may help.

1. Minerva RAM fail test is useful when an IC RAM fail completely. It means the IC doesn't work in any way.
But Minerva will fail to detect a partial RAM IC faulty. To confirm a faulty or working RAM IC you need a deeper tester. I started with one (programming it) but I had to pause it.
2. Multiplexers also may provoke similar issue since it is related with RAM.
3. Another IC (CPU, ZX8301, ROM) may provoke it.
4. DCDC converters would be in problems, just in case... check them with oscilloscope and very important the power supply.
5. Capacitors, the most usually DCDC in memory area.
7. Clock signal.

IMO, what is happening is that the process starts fine, but once it try to execute part of the ROM code fail and get "reset" or hang.
I have wrote in order what I would check first.

I would suggest to see with an oscilloscope the power lines, specially the ones for RAM.
Since TrumpCard has its own ROM, and also is failing with this card, I would left the ROM for the end.

Tomorrow I will read again the whole thread and I am interesting to learn from this problem (very interesting), so forgive me if I am not so advance QL user to go directly to the point, but I will be bringing ideas that I consider reasonable.

[techy]

found the problem - not happy - ;

8302 defective .

after replacing the modified 8302 from the defective ql ( hardwired rtc mod wirh exploded deac cell ) , and replacing it with the 8302 from my own ql it works as it should.

now searching for a 8302 ............

[Popopo]

found the problem - not happy - ;

8302 defective .

after replacing the modified 8302 from the defective ql ( hardwired rtc mod wirh exploded deac cell ) , and replacing it with the 8302 from my own ql it works as it should.

now searching for a 8302 ............

Congrats!
Another thing that I have learnt.
Thanks

[furellax]

found the problem - not happy - ;

8302 defective .

after replacing the modified 8302 from the defective ql ( hardwired rtc mod wirh exploded deac cell ) , and replacing it with the 8302 from my own ql it works as it should.

now searching for a 8302 ............

I've bought some genuine QL ICs from these guys:
https://www.ebay.com/itm/166705815480

Good luck!

[t0nyt]

now searching for a 8302 ............

I have several untested spare ones, if you get really stuck I can send one over for you to try (can hopefully test it before sending)

[bwinkel67]

The ZX8302 is used for controlling microdrives an serial ports (and maybe sound), right? Though the 8049/Hermes also works with serial ports so I guess you need both?

It's not a chip that will likely ever be re-created if its focus is microdrives. I hadn't thought about it but I guess I assumed you could live without it and the QL would still boot...that doesn't seem to be the case. So we have a ZX8301 replacement but still not a ZX8302. Probably the closest to creating one might be Charlie Ingley since he created a replacement chip for the Spectrum Interface 1 and I'm assuming they share some logic since both deal with microdrives.

[techy]

@t0nyt , i will keep that in mind !!

@ bwinkel67 , as it was the only chip i did not replace in my first tests it was literally my last hope -> and suddenly the ql came to life .
I have placed it back , and presto error came back too .