Ph2 Ringing on Apple 1 ACI

2 posts / 0 new
Last post
Offline
Last seen: 1 year 1 month ago
Joined: Sep 4 2009 - 21:04
Posts: 127
Ph2 Ringing on Apple 1 ACI

 Fixed Title

It was observed that the the ACI was having problems on the Extender Board when writing and I was able to trace the problem to ringing on Ph2 causing the write FF to get extra edges which I was able to fix by adding a 47 pF capacitor after the 100 Ω resistor already on the Ph2 line on the extender.

 

To understand why there was a problem I set up the Apple 1 clone with the ACI installed as normally used with no Extender board attached so this is simply the Apple 1 with ACI in normal operation and I looked at the Ph2 line on the ACI when running the ACI software which is shown in the fist photo.  It is clear there is a lot of ringing.  This type of ringing where different traces have different waveforms is what happens when you have a long line with lots of crosstalk, the noise will be dependent on the signals on other lines.  Looking at this you would expect there to be clocking failures because some ringing clearly goes below threshold but the next photo shows the Ph2 line with the FF Clock line and this time synching on the FF Clock line instead of the Ph2 line so only the Ph2 signals that occur that would reach the FF are seen and we see that the Ph2 signals that are used are making it.

 

Adding the Extender to the mix it is reasonable that the ACI on the Extender might have problems with added line length.  However, adding the 47 pF Capacitor on the Extender adds a 100 Ω series resistor followed by 47pF shunt capacitor to the Ph2 line so if we now connect the modified Extender to the Apple 1 and look at the Ph2 line on the ACI located on the Apple 1 we see that the Ph2 line is much better.  This is the same picture as the first photo and shows the effect of the 100 Ω 47 pF termination.
Offline
Last seen: 18 hours 16 min ago
Joined: Apr 1 2020 - 16:46
Posts: 913
It's not only the PHI2 clock that's ringing ...

... but almost all signals on the Apple-1 bus connector do ring to some extent. Adding any additional length of PCB traces by plugging in daughter cards (or extenders) only makes it worse. I had great trouble making my Floppy Disk Controller work due to the ringing on the clocks (this FDC is seen in the following thread: https://www.applefritter.com/content/uncle-bernies-woz-machine-based-apple-1-floppy-disk-controller).

 

Your fix is an interesting idea which does not change the timing on the clock line (s) by adding the fix too much.  This may rescue my FDC project... not that it does not work (it does) but the hack I have used so far to tame the clocks is not viable for producing more than one hand tuned prototype. Of course I know many circuits to cure corrupt clocks, but all would drive up IC count, which is not desirable.

 

Maybe your damping circuit helps my FDC, too ! Just doing it with a few passive components is always more elegant than to add PLLs etc.

 

(What a can of worms the Apple-1 is. But without its infamous quirks, they probably would not have gained notoriety and desirability among certain collectors bidding up to 1 Million US$ on them. The whole bizarre phenomenon not only has an engineering dimension, but also a psychological dimension which eludes me ... I always wonder if anyone would want to build an Apple-1 on a modified PCB that fixes all the issues. Probably not !)

 

Keep up the good work ! Yet another great contribution of wsander to the "remedy pill box" for Apple-1 diseases.

 

 

Log in or register to post comments