Experimental Physics and
Industrial Control System

[email protected] (Ned Arnold) · Fri, 16 Aug 1996 09:42:59 -0500

> I've got a situation where I need to cause a PV to process when a second
> PV makes a transition in one direction.  In otherwords, I need to fire
> PV1 when PV2 changes state 1->0, but not on the 0->1 change.  If it matters,
> PV1 is a BO to a GPIB device.  A particularly ugly GPIB device, at that.
> 
> I'm sure I've seen this done (short of writing a state machine in SNL),
> but it escapes me.  Probably just another case of "Berkeley Brain Rot."
> 
> Anyhow, suggestions, comments, flames, etc. appreciated.
> 
> 
> Disclaimer:  Any opinions are my own and have	    |  -bill
>     nothing to do with the official policy or the   |   [email protected]
>     management of L.B.L, who probably couldn't      |   Berkeley, CA
>     care less about employees who play with trains. |   aka [email protected]

I can think of two ways ...

  Use a CALC record and utilize the LA, LB, LC, ... fields that tell you
  the previous value of an input. An edge could be detected as (A&!B) where
  A points to PV1 and B points to the CALC_RECORD.LA .

  OR, you could use a WAIT record which is much like a CALC record but has 
  "conditional" processing of an output link. One of the conditions is when
  the result of the calculation changes from 1->0. BE CAREFUL ! The input
  links on a WAIT record are NOT PV LINKS (until 3.13, that is). They are
  raw text strings that indicate the PV name, which can change during
  run time.

  Have fun ...

  	Ned

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System