SMOC
I'll get back to you a bit later on the PCA/CA parameters.
Ideally they should be going off before you get an RA to avoid the sort of conflicting advice from ATC and TCAS that we have seen in past incidents/accidents.
From memory the lookahead time is 60 seconds and the minimum lateral and vertical closest point of approach values vary depending on the distance from the radar head and the applicable separation standard, but I need to check what they are now as we changed them a little while back.
As far as the plate is concerned, that is basically the crux of the matter that others are complaining about.
The chart and text do not indicate a restriction at 5,000 but the coding table attached (not shown in the AIP for some reason) does include hard altitude constraints.
It is normal practice these days for the procedure designer to attach an FMC coding table to remove any ambiguity for the coding companies (Honeywell etc) as to what the procedure designer actually meant.
In the past we saw a lot of different nav behaviour as different coding companies would interpret the procedures differently. For instance, sometime they would use a "Direct to Fix" path when the procedure called for a "Track to fix" to be flown.
Unfortunately, for the controllers, they do not normally see the coding table, only the chart, so for quite a while were berating pilots for asking if the restrictions could be cancelled. As far as they were concerned, there were no restrictions.
Although it has now been explained to them and they normally cancel the restictions when climb is given, it is still not fixed properly.
As a further example, If you look at AIP ENR 1.5(
http://www.hkatc.gov.hk/HK_AIP/AIP/ENR/HK_ENR1.5.pdf) and go to pages 27/28, you will see the segment of a Macau SID within the HK FIR.
Here you can see the coding table that is produced and sent to the coding companies to produce the procedure in your FMC.
Apart from the normal alt/speed/flyover/flyby restrictions, the Path Terminator is the code that tells the FMS how the track from one waypoint to the next should be flown. Its not as simple as just "DCT". Watching Airbus and Boeing aircraft flying supposedly the exact same coding is quite instructive sometimes.