• Please review our updated Terms and Rules here

SBC6120-RBC problem

thunter0512

Veteran Member
Joined
Sep 27, 2020
Messages
861
Location
Perth in Western Australia
I have been using the SBC6120-RBC for a few years, but recently Chris Tersteeg has created a nice FP6120 front panel clone of Bob Armstrong's original. I have now noticed that there may be a problem with the SBC6120-RBC because the following little count program does not behave as expected on the SBC6120-RBC + FP6120, but works fine on Bob Armstrong's original:

0: 7001
1: 2101
2: 5001
3: 5000

On the original when you select the AC with the rotary switch, the LEDs continuously count from 0000 to 7777 forever.
On the RBC version the LED display freezes at about 7777 and then resumes what is approximately one full count cycle later.

When while in the frozen state I hit break on the console or toggle the Halt switch, I see that the program is running fine - just the LEDs are not updating.

In an effort to narrow down the culprit I have now replaced the SBC6120-RBC with Bob Armstrong's original SBC6120 and plugged that into Chris Tersteeg's FP6120 and it behaves correctly.

When I plug the SBC6120-RBC into Bob Armstrong's original FP6120 the data LEDs freeze as before. I also tried a second SBC6120-RBC and it too has the "freezing" behaviour.

This seems to indicate that there is some problem with the SBC6120-RBC as the problem follows the SBC6120-RBC.

When the display is frozen I see that the CPREQ signal keeps pulsing at approx 30 Hz.

The ROM Monitor in both SBC6120 (original) and SBC6120-RBC is version V320.

The original SBC6120 and both SBC6120-RBC run OS/8 without a glitch. There is no noticably freezing of the LEDs with OS/8 on any of the 3 SBC6120.

I wonder what could cause this:

  • The SBC6120-RBC is a double layer board whereas the original is a 4 layer board. Possibly when AC wraps from 7777 to 0000 it causes some noise on the 2 layer board which disturbs something.
  • The PAL firmware may be subtly different.

Could some of you please try this if you have any FP6120 (original or Chris's version) and a SBC6120-RBC.
 
It turns out that if I enter and start the 4 word program above using the following console commands, the "freeze" does not happen:

d 0 7001,2101,5001,5000
st 0

The FP6120 AC display nicely counts from 0000 to 7777, wraps around to 0000 and continues counting up to 7777 forever.

Doing the same via the FP6120 switches causes the AC display to freeze as soon as AC reaches 7777.

Nevertheless Bob Armstrong's original SBC6120 works correctly under all circumstances with both original FP6120 and Chris's FP6120.
 
Hmmm - during the "frozen" state a single console key press unfreezes the system and the count continues without ever freezing again. The key-press shouldn't be serviced by the monitor program in EPROM, but somehow the pending key-press affects subsequent behaviour.

Any thoughts?
 
Back
Top