128k woes - Sinclair

This is a discussion on 128k woes - Sinclair ; Hello again! There's an issue with my Speccy 128k for which I hope you might have some possible solutions in store. Every now and then (mostly when loading from tape, but it has occured at any given time) the machine ...

+ Reply to Thread
Results 1 to 16 of 16

Thread: 128k woes

  1. 128k woes

    Hello again!

    There's an issue with my Speccy 128k for which I hope you might have some
    possible solutions in store.

    Every now and then (mostly when loading from tape, but it has occured at any
    given time) the machine crashes in such a way that the whole screen is
    scrambled/distorted, sometimes with odd sounds, and the video signal goes
    beserk. Often, not even resetting the machine would help, only "pulling the
    plug" would do the trick.

    Regarding my earlier post, "PSU question", I've replaced the lead going from
    the PSU into the computer. But maybe this is a power brick issue after all.

    Any takers?

    Thanks,

    Frank



  2. Re: 128k woes


    "fsh" napísal

    >
    > There's an issue with my Speccy 128k for which I hope you might have some
    > possible solutions in store.
    >
    > Every now and then (mostly when loading from tape, but it has occured at
    > any given time) the machine crashes in such a way that the whole screen is
    > scrambled/distorted, sometimes with odd sounds, and the video signal goes
    > beserk. Often, not even resetting the machine would help, only "pulling
    > the plug" would do the trick.
    >
    > Regarding my earlier post, "PSU question", I've replaced the lead going
    > from the PSU into the computer. But maybe this is a power brick issue
    > after all.
    >
    > Any takers?


    Beautiful English, pleasure to read

    B


  3. Re: 128k woes

    Bohus Král wrote:
    > Beautiful English, pleasure to read


    You wot, mate? The rest of us talk like fahking scum, do we?!

    Bollocks.

  4. Re: 128k woes


    > You wot, mate? The rest of us talk like fahking scum, do we?!


    Before we go any further ( f*ing 'bout), may I just point out (you stoopid
    c**tz), that there's a (s*dding) problem with me Spectrum at hand?

    Now go solve it, or else...

    (((-;



  5. Re: 128k woes

    On May 23, 4:28 pm, "fsh" wrote:
    > > You wot, mate? The rest of us talk like fahking scum, do we?!

    >
    > Before we go any further ( f*ing 'bout), may I just point out (you stoopid
    > c**tz), that there's a (s*dding) problem with me Spectrum at hand?
    >
    > Now go solve it, or else...
    >
    > (((-;


    And this is when I come in to suggest that you might want to ask this
    question on WOS's hardware section.

  6. Re: 128k woes

    Assuming its the toast rack 128, take a look at the connection to the
    regulator, which is a push on connector in many. When it gets old, this can
    be very intermittent. The 128 can be very prone to noise as the fan out for
    the chips is a bit high and almost anything seems to crash it. I used to
    find just an inch of ribbon cable on the expansion port was enough to do
    this. It was so bad on some models that we used to fit NEC Z80s with a
    military spec to get the data signals up.
    Of course as we all know dry joints tend to come with age, so it could be
    all sorts of things. Intermittent faults are a real swine to find.

    Brian

    --
    Brian Gaff - briang1@blueyonder.co.uk
    Note:- In order to reduce spam, any email without 'Brian Gaff'
    in the display name may be lost.
    Blind user, so no pictures please!
    "fsh" wrote in message
    news:7f77c$48372c82$4e367a3f$22919@news.usenext.de ...
    > Hello again!
    >
    > There's an issue with my Speccy 128k for which I hope you might have some
    > possible solutions in store.
    >
    > Every now and then (mostly when loading from tape, but it has occured at
    > any given time) the machine crashes in such a way that the whole screen is
    > scrambled/distorted, sometimes with odd sounds, and the video signal goes
    > beserk. Often, not even resetting the machine would help, only "pulling
    > the plug" would do the trick.
    >
    > Regarding my earlier post, "PSU question", I've replaced the lead going
    > from the PSU into the computer. But maybe this is a power brick issue
    > after all.
    >
    > Any takers?
    >
    > Thanks,
    >
    > Frank
    >
    >




  7. Re: 128k woes

    On 2008-05-24, Brian Gaff wrote:
    > Of course as we all know dry joints tend to come with age, so it could be
    > all sorts of things. Intermittent faults are a real swine to find.


    I'd also add that capacitors age, too - electrolytics in particular.
    Aged electrolytics whose capacitance has fallen and ESR has increased,
    tends to lead to excessive ripple on the 5 volt power supply rails.
    Sometimes, the machine will function fine, then for whatever reason, a
    slightly higher power demand is made - and the ripple gets a little too
    much and *crash*.

    I've seen it with modern PCs too, we had randomly crashing machines at
    work - checked the 5v rails with an oscilloscope and found 0.5v worth of
    ripple. It's a wonder they worked at all. When I took the PSU apart I
    found bulging and leaking electrolytic capacitors. But they aren't
    always so obvious when they've gone 'sub par'.

    An outside possibility is memory gone bad, I've seen memory (on PCs as
    well as on a Spectrum) suffer from 'bit fade' - they work OK for a time,
    then you get some random corruption. If this happens where the Z80's
    stack happens to be, *crash*.

    Note that the modern CMOS Z80 (they are still manufactured in 40 pin DIP
    which just drops into a Speccy) has MUCH better fan-out capability than
    the original NMOS Z80.

    --
    From the sunny Isle of Man.
    Yes, the Reply-To email address is valid.

  8. Re: 128k woes

    Dylan Smith schrieb:

    > I'd also add that capacitors age, too - electrolytics in particular.
    > Aged electrolytics whose capacitance has fallen and ESR has increased,
    > tends to lead to excessive ripple on the 5 volt power supply rails.


    I know that from switching PSUs, but the capacitors on the Speccy's 5V
    filter only mains' 50 or 60 Hz ripple. The whole PSU circuit of the
    Speccy is rather low power, there shouldn't be so much stress on the
    capacitors to cause them to go bad. It can't hurt to check, though

    > Note that the modern CMOS Z80 (they are still manufactured in 40 pin DIP
    > which just drops into a Speccy) has MUCH better fan-out capability than
    > the original NMOS Z80.


    Got a manufacturer / part code?

    Hanno

  9. Re: 128k woes

    On 2008-06-02, Hanno Foest wrote:
    > I know that from switching PSUs, but the capacitors on the Speccy's 5V
    > filter only mains' 50 or 60 Hz ripple.


    It's also needed for the 7805 regulator to work stably, and it's not
    just power filtering, either - all those gates switching puts
    considerable noise (ringing) on the power rails (although the ceramic
    decoupling caps should take care of that).

    The 48K and toastrack 128K models also have switch mode power supplies
    for the -5v, +12v and -12v rails (although I think in the 128k
    toastrack, the +12/-12 is only used for the analogue side).

    50 Hz ripple is still not what you want, though!

    >The whole PSU circuit of the
    > Speccy is rather low power, there shouldn't be so much stress on the
    > capacitors to cause them to go bad. It can't hurt to check, though


    It doesn't really matter if it's low or high power - electrolytic
    capacitors simply degrade with age.

    >> Note that the modern CMOS Z80 (they are still manufactured in 40 pin DIP
    >> which just drops into a Speccy) has MUCH better fan-out capability than
    >> the original NMOS Z80.

    >
    > Got a manufacturer / part code?


    Zilog - Z84C0008PEG. Farnell order code 1081890.

    --
    From the sunny Isle of Man.
    Yes, the Reply-To email address is valid.

  10. Re: 128k woes

    Dylan Smith schrieb:

    [ringing]
    > although the ceramic
    > decoupling caps should take care of that).


    Right.

    >> The whole PSU circuit of the
    >> Speccy is rather low power, there shouldn't be so much stress on the
    >> capacitors to cause them to go bad. It can't hurt to check, though

    >
    > It doesn't really matter if it's low or high power - electrolytic
    > capacitors simply degrade with age.


    All components fail eventually, those stressed fail earlier.
    Electrolytic capacitors mostly degrade because of heat, stressing them
    (especially in switch mode PSUs) heats them up - and considering how
    common switch mode PSUs are these days that means you see lots of
    failing electrolytic capacitors is you do electronis repairs. But those
    not stressed are fine for a long time, e.g. those in lots of older TV
    sets. The same goes for the capacitors in a ZX Spectrum, they shouldn't
    really have a noticeable failure rate after only 25 years.

    It still *may* happen. But that would be really bad luck.

    Hanno - who still has a few broken Spectrums to fix.

  11. Re: 128k woes

    Dylan Smith wrote:
    > The 48K and toastrack 128K models also have switch mode power supplies
    > for the -5v, +12v and -12v rails (although I think in the 128k
    > toastrack, the +12/-12 is only used for the analogue side).


    Doesn't RS232 require +/- 12v?
    --
    | |What to do if you find yourself stuck in a crack|
    | spike1@freenet.co.uk |in the ground beneath a giant boulder, which you|
    | |can't move, with no hope of rescue. |
    | Andrew Halliwell BSc |Consider how lucky you are that life has been |
    | in |good to you so far... |
    | Computer Science | -The BOOK, Hitch-hiker's guide to the galaxy.|

  12. Re: 128k woes

    Yes (well the full spec says so anyway)

    Andy

    "Andrew Halliwell" wrote in message
    news:6ublh5-ob6.ln1@ponder.sky.com...
    > Dylan Smith wrote:
    >> The 48K and toastrack 128K models also have switch mode power supplies
    >> for the -5v, +12v and -12v rails (although I think in the 128k
    >> toastrack, the +12/-12 is only used for the analogue side).

    >
    > Doesn't RS232 require +/- 12v?
    > --
    > | |What to do if you find yourself stuck in a
    > crack|
    > | spike1@freenet.co.uk |in the ground beneath a giant boulder, which
    > you|
    > | |can't move, with no hope of rescue.
    > |
    > | Andrew Halliwell BSc |Consider how lucky you are that life has been
    > |
    > | in |good to you so far...
    > |
    > | Computer Science | -The BOOK, Hitch-hiker's guide to the
    > galaxy.|




  13. Re: 128k woes

    The message <6ublh5-ob6.ln1@ponder.sky.com>
    from Andrew Halliwell contains these words:

    > Dylan Smith wrote:
    > > The 48K and toastrack 128K models also have switch mode power supplies
    > > for the -5v, +12v and -12v rails (although I think in the 128k
    > > toastrack, the +12/-12 is only used for the analogue side).


    > Doesn't RS232 require +/- 12v?


    RS232 specifies a maximum bi-polar signalling level of +/-24v but lower
    bipolar voltages, such as +/-12 and +/- 5 volts can be (and are) used.

    It's even possible to use 0 and +5 volts with most RS232 transciever
    chips if the resulting limited distance and data rates are accepted.

    Most, if not all, PC versions of RS232 use +/- 12v for the transciever
    chips which will accept signals recieved over a short link (c 2 metres)
    from a device employing only the 0 and +5 volt levels at data rates of
    115Kbd or less.

    --
    Regards, John.

    Please remove the "ohggcyht" before replying.
    The address has been munged to reject Spam-bots.


  14. Re: 128k woes

    Johnny B Good wrote:
    > The message <6ublh5-ob6.ln1@ponder.sky.com>
    > from Andrew Halliwell contains these words:
    >
    >> Dylan Smith wrote:
    >>> The 48K and toastrack 128K models also have switch mode power supplies
    >>> for the -5v, +12v and -12v rails (although I think in the 128k
    >>> toastrack, the +12/-12 is only used for the analogue side).

    >
    >> Doesn't RS232 require +/- 12v?

    >
    > RS232 specifies a maximum bi-polar signalling level of +/-24v but lower
    > bipolar voltages, such as +/-12 and +/- 5 volts can be (and are) used.
    >
    > It's even possible to use 0 and +5 volts with most RS232 transciever
    > chips if the resulting limited distance and data rates are accepted.
    >
    > Most, if not all, PC versions of RS232 use +/- 12v for the transciever
    > chips which will accept signals recieved over a short link (c 2 metres)
    > from a device employing only the 0 and +5 volt levels at data rates of
    > 115Kbd or less.
    >


    my reply seems to have been eaten by the server
    I tried to say, when did sir clive ever follow standards
    iirc the plus three doesn't swing negative on it's serial pins, it goes
    from no volts, to some volts...
    but of course sir clive didn't make the +3 so that nullifies my comment
    really

  15. Re: 128k woes

    Guesser wrote:
    > my reply seems to have been eaten by the server
    > I tried to say, when did sir clive ever follow standards
    > iirc the plus three doesn't swing negative on it's serial pins, it goes
    > from no volts, to some volts...


    What does the +3 have to do with sir...

    > but of course sir clive didn't make the +3 so that nullifies my comment
    > really


    Yeah, that.
    --
    | spike1@freenet.co,uk | "Are you pondering what I'm pondering Pinky?" |
    | Andrew Halliwell BSc | |
    | in | "I think so brain, but this time, you control |
    | Computer Science | the Encounter suit, and I'll do the voice..." |

  16. Re: 128k woes

    The message
    from Guesser contains these words:

    > Johnny B Good wrote:
    > > The message <6ublh5-ob6.ln1@ponder.sky.com>
    > > from Andrew Halliwell contains these words:
    > >
    > >> Dylan Smith wrote:
    > >>> The 48K and toastrack 128K models also have switch mode power supplies
    > >>> for the -5v, +12v and -12v rails (although I think in the 128k
    > >>> toastrack, the +12/-12 is only used for the analogue side).

    > >
    > >> Doesn't RS232 require +/- 12v?

    > >
    > > RS232 specifies a maximum bi-polar signalling level of +/-24v but lower
    > > bipolar voltages, such as +/-12 and +/- 5 volts can be (and are) used.
    > >
    > > It's even possible to use 0 and +5 volts with most RS232 transciever
    > > chips if the resulting limited distance and data rates are accepted.
    > >
    > > Most, if not all, PC versions of RS232 use +/- 12v for the transciever
    > > chips which will accept signals recieved over a short link (c 2 metres)
    > > from a device employing only the 0 and +5 volt levels at data rates of
    > > 115Kbd or less.
    > >


    > my reply seems to have been eaten by the server
    > I tried to say, when did sir clive ever follow standards
    > iirc the plus three doesn't swing negative on it's serial pins, it goes
    > from no volts, to some volts...
    > but of course sir clive didn't make the +3 so that nullifies my comment
    > really


    Thanks for providing a 'classic example' of the use of only a 5 volt
    rail with RS232 transciever chips. Although the +3 only provides
    unipolar voltage level signalling on its RS232, it can quite safely be
    connected to a system box using bipolar voltage signalling (even up to
    levels of +/-24v, not just 12 or 5 volt) since the transciever reciever
    input is designed to cope with the standard bipolar voltages in spite of
    only being powered off just a single 5 volt supply.

    The distance and data rates will be severely limited but the typical 2
    metre data cable length will permit reasonable data rates to be used
    with little or no problem.

    To be fair, it wasn't just Sir Clive's products which used the '5 volt
    only budget option' designed into the RS232 chips, other products which
    only needed to support 'short haul' modest data rate link ups have also
    taken this option (and for pretty much the same reason - elimination of
    the expense of a bipolar supply and no necessity to send data over more
    than a couple of metres of cable).

    --
    Regards, John.

    Please remove the "ohggcyht" before replying.
    The address has been munged to reject Spam-bots.


+ Reply to Thread