Re: [foxboro] One IND block per compound -- still true?

OK, thanks for all the replies.  It must've been one of the 3 or 4 old 
wives tales I heard about I/A but never had any real problems with in my 
own experience.

"One thing I'd caution you, be it that you are a neophyte at HLBL. The
default BPCSTM is (IIRC) 100. If you don't change that default to
something reasonable *and* if your codes loop infinitely without a WAIT
statement, you'll drag your CP to its knee in no time flat."


In some testing of another program I wrote awhile back, I discovered the 
BPCSTM caution you mentioned.  It endlessly looped, and with the BPCSTM at 
the default of 100, it reduced the IDLETM of a CP on our test system 
something like 40% (!)


"1 line of HLBL code is roughly the same load as an AIN block.

Internal block references don't suspend the block, but they are slower 
than you might expect and will have an impact on your loading if you have 
enough of them."


Good to know, and relates to the BPCSTM issue as well.  I so far have 
tried not to set BPCSTM higher than 20 or 25, or put dummy WAITs in the 
code where they can be tolerated.  In my rolling average program, though, 
I think I come out ahead block-equivalent-wise using the IND block :)


I'll still reserve HLBL for those problems that the other blocks can't 
solve practically, as Alex suggests.  I've seen "regular" interlocking 
(i.e., where a CALCA would do fine) done in HLBL, but I don't see the 
advantages in that approach.


I'll also post my hack to the wiki, though whether it is a "best practice" 
I'll leave to the Paulas, Simons, and Randys that read it.


Corey Clingo
BASF Corporation





 
 
_______________________________________________________________________
This mailing list is neither sponsored nor endorsed by Invensys Process
Systems (formerly The Foxboro Company). Use the info you obtain here at
your own risks. Read http://www.thecassandraproject.org/disclaimer.html
 
foxboro mailing list:             http://www.freelists.org/list/foxboro
to subscribe:         mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe:      mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave

Other related posts: