Experimental Physics and
Industrial Control System

"Allison, Stephanie" <[email protected]> · Fri, 22 Nov 2013 17:08:02 -0800

Hi Emmanuel,

> Is RVAL supposed to be an integer?

Yes.

> How can fix my current example?

I'm sure other people have a better solution but I use a subroutine record.  The logic looks something like this (I snipped code that you don't care about so maybe it doesn't compile):
...
#include <cvtTable.h>         /* for cvtRawToEngBpt      */
...
long subBLMagField(struct subRecord *psub)
{
  /*
   * Inputs:
   *          A = Gap      (mm)
   *          D = Breakpoint Table Index (index in menuConvert.dbd)
   * Outputs:
   *          H = Value from the Breakpoint Table
   *          J = Previous Breakpoint Table Index
   *       DPVT = Pointer to Breakpoint Table
   *          L = Last Breakpoint
   *        VAL = Magnetic Field    (T)
   */
  short lbrk;        /* Last breakpoint              */
  long  lbrkStatus;  /* Breakpoint conversion status */

    if (psub->d != psub->j)
    {
      psub->j = psub->d;
      psub->dpvt = 0;
    }
    lbrk    = psub->l;
    psub->h = psub->a;
    lbrkStatus = cvtRawToEngBpt(&psub->h, (short)psub->d, 0,
                                &psub->dpvt, &lbrk);
    /* If at a breakpoint table limit, OK to use the limit (status = 1) */
    if ((lbrkStatus != 0) && (lbrkStatus != 1)) return -1;
    psub->l   = lbrk;
    psub->val = psub->h;
  return 0;
}

Hope this is helpful,

Stephanie Allison

> -----Original Message-----
> From: [email protected] [mailto:[email protected]] On Behalf
> Of Emmanuel Mayssat
> Sent: Friday, November 22, 2013 4:49 PM
> To: [email protected]
> Subject: Breakpoint tables: Raw value must be integer?!?
> 
> 
> I would like to replace a few of my existing CALC records (polynomial transformation)
> with simple breakpoint tables.
> Let's assume that my raw values are voltages between 0 and 5V, I could therefore build a
> transformation table as follow:
> 
> breaktable(bptPosSlope) {
>         raw=0.000000 slope=9.000000e+00 eng=-27.000000
>         raw=1.000000 slope=1.800000e+01 eng=-18.000000
>         raw=2.000000 slope=1.800000e+01 eng=0.000000
>         raw=3.000000 slope=1.400000e+01 eng=18.000000
>         raw=4.000000 slope=3.000000e+00 eng=32.000000
>         raw=5.000000 slope=3.000000e+00 eng=35.000000
> }
> 
> Although the table allows for double/float transformation, to my surprise, the RAW value
> stored in the AI record is an integer.
> Maybe is it because of my INP link...
> 
> This record is supposed to be the raw value
> clsXdev:6175>dbpr RECORDS:BrkpntTable:Rdbck:RawAI
> ASG:                DESC: Raw readback  DISA: 0             DISP: 0
> DISV: 1             NAME: RECORDS:BrkpntTable:Rdbck:RawAI   RVAL: 0
> SEVR: NO_ALARM      STAT: NO_ALARM      SVAL: 0             TPRO: 0
> VAL: 1.111
> 
> Which is transformed in another ai record (which has its INP field pointing to the above
> record).
> 
> clsXdev:6175>dbpr RECORDS:BrkpntTable:RdbckAI 3
> ACKS: NO_ALARM      ACKT: YES           ADEL: 0             ALST: -18
> AOFF: 0             ASG:                ASLO: 1             BKPT: 00
> DESC: Readback      DISA: 0             DISP: 0             DISS: NO_ALARM
> DISV: 1             DTYP: Raw Soft Channel                  EGU: mm
> EGUF: 0             EGUL: 0             EOFF: 0             ESLO: 1
> EVNT: 0             FLNK:CONSTANT 0     HHSV: NO_ALARM      HIGH: 0
> HIHI: 0             HOPR: 0             HSV: NO_ALARM       HYST: 0
> INIT: 0             INP:DB_LINK RECORDS:BrkpntTable:Rdbck:RawAI.VAL NPP NMS
> LALM: -18           LBRK: 0             LCNT: 0             LINR: bptPosSlope
> LLSV: NO_ALARM      LOLO: 0             LOPR: 0             LOW: 0
> LSV: NO_ALARM       MDEL: 0             MLST: -18
> NAME: RECORDS:BrkpntTable:RdbckAI       NSEV: NO_ALARM      NSTA:
> NO_ALARM
> ORAW: 1             PACT: 0             PHAS: 0             PINI: NO
> PREC: 3             PRIO: LOW           PROC: 0             PUTF: 0
> ROFF: 0             RPRO: 0             RVAL: 1             SCAN: 1 second
> SDIS:CONSTANT       SEVR: NO_ALARM      SIML:CONSTANT       SIMM: NO
> SIMS: NO_ALARM      SIOL:CONSTANT       SMOO: 0             STAT: NO_ALARM
> SVAL: 0             TIME: 2013-11-22 16:44:26.374133281     TPRO: 0
> TSE: 0              TSEL:CONSTANT       UDF: 0              VAL: -18
> 
> 
> Now for some reason the RVAL of this last record is = to 1 (and not 1.1111)
> As a result, the VAL is not calculated correctly.
> 
> Is RVAL supposed to be an integer?
> How can fix my current example?
> 
> Thanks,
> --
> Emmanuel

Experimental Physics and Industrial Control System

Experimental Physics and
Industrial Control System