ACA500plus startup issue

Caution: Non registered users only see threads and messages in the currently selected language, which is determined by their browser. Please create an account and log in to see all content by default. This is a limitation of the forum software.


Also users that are not logged in can not create new threads. This is a, unfortunately needed, counter measure against spam. Please create an account and log in to start new threads.

Don't Panic. Please wash hands.

    I have the following issue:


    - On power-on, the ACA500plus will always blink briefly with the led(s).

    - However, the system more often than not stays dark at this point.

    - Often I have to flip the power switch 7-8 times with a certain timing to get the whole system to reach the "spinning" animation on the DISMO.

    - I then see 00, followed by FF and the well-known dark blue screen. (Happens always if I reach this far)

    - Ctrl-C-A kbd reset always gets me into the ACA menu and normal operation from then on.

    - No other problems with warm-reset, it seems to be a problem exclusively happening at POR.

    - My Amiga 500 is a Rev. 3 "type 2" with the 4-pin oscillator.

    - Original PSU that came with the A500, the heavier model with the transformer core.

    - Only other device installed is a Gotek, replacing my DF0.


    I have replaced the C712 (10uF el.) capacitor near the 555, after reading some old threads.

    This did not fix the problem, unfortunately. I'm seeing the same pattern.


    What else can I do?

    Easy test: remove the Gotek from the setup


    Also, if you didnt already do this, clean the connector and expansion port with alcohol - a surprising number of problems can be tracked down to dirty connectors.

    I'll try without the Gotek. But it's likely to be compatible, right?


    I have already cleaned the expansion port card edge with isopropanol.


    I think I will also try the 1kohm pull-up on the reset line.

    Tried the 1k resistor and it didn't help.


    I did however replace the 2.5A Amiga 500 supply with a 3.0A one from an A600, and I'll be damned if it doesn't seem to start up more often with that one.


    I haven't had time to unplug the Gotek yet, will try that as well.

    Unfortunately, that's a common problem with some power supplies: When they get old, their startup behaviour is not deterministic. I also have a few old PSUs where I have to launch the system with a keyboard reset. It's just a thing you have to live with when using 35-year old computers.

    Is there something that could help it in the PSU itself, like replacing the caps?

    Remove corrosion from the connector?


    And please note, I still get an FF on power-on and have to do a keyboard reset - with either PSU.

    Quote from cadahl

    Is there something that could help it in the PSU itself, like replacing the caps?

    Remove corrosion from the connector?

    Two points worth investigating, yes.



    Quote from cadahl

    And please note, I still get an FF on power-on and have to do a keyboard reset - with either PSU.

    That indicates a problem with the rising edge of the reset signal. Did you try the pull-up resistor on the reset signal? You wrote you've replaced the timing cap of the NE555 circuit, but nothing more in terms of cleaning up the reset signal.

    I did try the 1k resistor in the connector on the ACA500plus, as shown by you in another thread. At that point I didn't see an improvement but of course no soldering was involved so I can't be 100% there was contact.

    Using an online calculator I see that the 47k (?) / 10uF on the 555 gives about 0.5s of reset pulse. Is that what the ACA500plus expects?


    If I measure the actual reset signal on my board with a logic analyser, what should I look for? Or would I need an analog scope to see the rise time?

    The NE555 is not a particularly good reset generator, as it's not starting at low-level when powered up. It's taking a bit of time to go low, then does it's delay, then release the reset line. If you really want to measure, you need to measure from the moment that the power rail is rising.


    It may be simpler to use a MAX809-type reset generator and be done with it. However, you may be lucky by increasing the 47k resistor, prolonging the reset pulse and maybe getting rid of the keyboard reset requirement.

    An additional circuit that makes sure that reset is held low while voltage is ramping up can even be done externally. I agree about keeping a Rev.3 as original as possible, because they're rare. Do you have a genuine one with the Cherry keyboard? Just asking out of interest, has nothing to do with this case.


    The exact part number to try for an additional reset chip would be MAX803L - this will produce an open-drain reset that's really pulled low while power supply is below 4.63V and only release it when power is stable for at least 140ms. This would fix the "reset is high on power-up" problem of the NE555 circuit (=start of the pulse), and the higher resistor value would make the pulse longer, moving the end of the pulse.


    An alternative part number would be OnSemi NCP803SN463.


    At this point, I don't exactly know which one is the problematic one (start or end of the pulse), and it's not easy to test with the keyboard only. One possible way to test this would be to hold the CTRL-A-A keyboard reset while powering on, then quickly release the keyboard reset, ideally within half a second, so the end of the pulse is generated by the NE555. Since these three keys are hard-wired to reset (with a 3-input OR gate), this should pull low safely during startup. If you release quickly, you can be sure that the NE555 circuit still pulls low. Might be hard to time, and will require a few attempts to be able to tell that it really fixes it (I have a picture in my head where I'd have two hands on the keyboard and switch the PSU with my toe). If it does, you might even fix it with a capacitor between VCC and the output of the NE555, as that will raise the base of the reset-transistor during the moment of startup, potentially fixing the start of the reset pulse.


    We could do more investigations, as different NE555 manufacturers might behave differently on power-up. You're not the only one having this problem, but so far, each and every customer didn't follow up when I suggested to go deeper on this - and this is since introduction of the ACA500plus almost 6 years ago, so it either doesn't bother people too much, or they found a solution that wasn't reported.

    Interesting reading!


    I'm willing to try things until it works, seems like it should be fixable somehow.. it's just a pulse after all.


    And yes, it's a Rev 3 with the "space invaders" mechanical keyboard. Hi-Tek I think is the manufacturer. I was very pleased to discover my childhood Amiga was this variety, as I always loved the keyboard on it, so I'm treating it with care now. And it's fun that it's quite rare, too. :)


    I've done a reversable audio filter mod (100 ohm resistors across the opamp) and replaced that reset cap, but otherwise it's original.

  • The last reply was more than 365 days ago, this thread is most likely obsolete. It is recommended to create a new thread instead.

    • :)
    • :(
    • ;)
    • :P
    • ^^
    • :D
    • ;(
    • X(
    • :*
    • :|
    • 8o
    • =O
    • <X
    • ||
    • :/
    • :S
    • X/
    • 8)
    • ?(
    • :huh:
    • :rolleyes:
    • :love:
    • 8|
    • :cursing:
    • :thumbdown:
    • :thumbup:
    • :sleeping:
    • :whistling:
    • :evil:
    • :saint:
    • <3