[Date Prev][Date Next][Thread Prev][Thread Next] - [Date Index][Thread Index][Author Index]

Re: Special Bulletin ANS-063.02 BOD Meeting

>> There was the minor matter of a LIU wiring error.

>I keep hearing hints about this, but it semms conspicuously absent from 
>the history of the project. Anybody wanna share the story?

A CD4015 8-bit serial-to-parallel converter listens to the engineering
beacon generated by the IHU.  When the IHU sends the proper firing
code prefix (think of the CRM-114 device in "Dr. Strangelove"), the
converter captures a subsequent 8-bit fire duration value, loads it
into two CD4029 4-bit counters that form an 8-bit counter, and begins
firing the motor.

The counters are clocked by the 400 Hz beacon clock divided by 1024 in
a CD4040. This results in one count every 2.56 seconds. When the
counter reaches zero, the motor valve is closed. The actual valve
control pulses are generated by a pair of 555 timers (one to open, one
to close).

The problem? The PC traces carrying bits 5 and 6 from the
serial-to-parallel converter to the down-counter were crossed.  So
when the desired burn duration of 29 (hex) was sent to specify a burn
duration of 105 sec, the counter was actually loaded with 49 (hex),
giving a burn duration of 186.88 seconds. [Most of this information
comes from my copy of the AO-10 LIU schematics and PC images. I didn't
have the actual burn counts, so I regenerated them from the 2.56sec
clock rate and the planned and actual durations given in Peter's

One thing to note about this bug is that there was no good reason to
do the burn timing in dedicated hardware in the first place, as it
could have easily been done in IHU software. Ariane safety
requirements did reasonably dictate a hardware "lock" on the motor to
make it impossible for an "insane" IHU to fire it until the
appropriate keys were uplinked after launch. But this need only have
taken the form of a simple interlock that, when enabled, allowed the
IHU to fire the motor under direct software control. This would have
avoided quite a few parts and, as it turned out, an opportunity to
make a wiring mistake.


Via the amsat-bb mailing list at AMSAT.ORG courtesy of AMSAT-NA.
To unsubscribe, send "unsubscribe amsat-bb" to Majordomo@amsat.org