Re: ZX Spectrum - can't get it repaired - help!



GdB wrote:
Hi,

I know I can buy a used one for EUR 5 or so, but that's not the fun ;-)

Problem is that I can't get the ZX to boot... black and white big
vertical bars, in some horizontal lines all kinds of coloured
flickering blocks and some thin (often red) vertical lines...

I have a ZX Spectrum issue 4A. I have now swapped almost all components
with a working computer, that way I know (or at least I think I know,
correct me if I'm wrong) that the following components are ok:

o Swapped Z80A - correct;
o Swapped ROM (not 100% usre if that one works, but it came from
another ZX Spectrum) - both seemed non-Hitachi, so the links could
remain in the N-position, not the H-position, Rom says something like
AMI bla bla and somewhere 128, so it should be allright. The original
was a Nec ROM;
o TR4, 5 and 6 were changed, although the 5V, 12V and -5V seemed ok
(-5V seemed a bit lower). That should mean that the coil, the
transistors, PSU, 5.1V zener diode, etc. are fine;
o All 4116 RAM's are 100% ok; all connections were checked, upperside
and bottom side are connected, pins 2 and 14 are connected and all
other pins are in the "chain" (so pin5 of IC6 connects to pin5 of IC7,
etc.);
o 7805 is 100% ok;
o ULA works (6C001E-7 - works great in a 3B issue);
o Even changed the 74157 that is used for the basic 16K computer;
o Two of the electronic capacitors were changed that play a role in the
power part (9V, 12V, 5V, -5V);

Problem is that I forgot what caused this, it was years ago. Must have
been a peripheral taken of while the power was on, but that usually
damages the TR4/TR5 parts...


What else can I check / change? Can the 32K (the eight 4532's + five
74xxx) "extension" do any harm, do I need to take it out (all other
IC's are in IC holders now and that was A LOT of work, so that would be
13 more chips times 16 pins!!).

If the "upper memory" does not need to be taken out, I would say it
doesn't need to, what else can I check. The PCB looks, with an
eyeglass, quite ok and replacing resistors doesn't make much sense. I
can't imagine the 1N4148's are damaged either... Capacitors?

It CAN still be something in the power circuit, as the sounds of the
coil is not 100% normal... Normally when you switch it on, it makes
sound A and after a second or so, it makes sound B and it keeps sound
B. In my case, it seems like the computer knows it can not boot and
keeps on trying, although that's a wild guess, nothing spectacular
changes on the screen, flickering blocks keep on being flickering
blocks, no change after that first second either...

Well, that was a long story, please share your thoughts ;-)

Cheers,


G

It`s been a long time, but I used to repair Spectrums regularly, and what you describe is symptomatic of one of the memory ic`s being faulty, have you tried it with the 32k extension removed? are you absolutely sure that all the memory chips are ok? I might still have a load of old speccy bits laying around somewhere.

Ron(UK)


--
Lune Valley Audio
Public Address Systems
Hire Sales Maintenance
www.lunevalleyaudio.com
.



Relevant Pages

  • Re: maybe a bit OT: EEPROMs and PCs
    ... | Ahh, okay. ... | BIOS chips. ... For all ROM code I'd use ... The things inbetween these brackets are the memory address. ...
    (alt.lang.asm)
  • Re: IIgs peculiarities
    ... permanently shadowing from bank 0 would have corrupted whatever was ... Same for a ROM 1 or ROM 3. ... Memory cards which support more than four "rows" do so via a mechanism ... so a memory card which uses this technique is ...
    (comp.sys.apple2)
  • Re: Large virtual memory, small working set
    ... We're basically opening all these connections and receiving data. ... Almost immediately upon starting and loading configuration information, the application starts opening up connections and the memory usage immediately skyrockets. ... The strange thing is - only the virtual memory increases this drastic. ... Is it some sort of side-effect from having that number of open TCP connections that we are stuck with? ...
    (microsoft.public.dotnet.languages.csharp)
  • Re: Self booting systems with out bootloader?!!!
    ... A non-fragmented memory just means that every byte from the start of memory ... This means that after a reset is made, ... in processor ROM. ... executing the 'work program'. ...
    (comp.arch.embedded)
  • Re: Flash CPU level / ROM question
    ... I have not had problems getting the background sound ... to work on the Green Flipper ROM version of Flash. ... Dan Gutchess said that all Flash games were supposed to have the background ...
    (rec.games.pinball)