Re: [foxboro] PIDA-FBTUNE configuration

        I copied this right out of the Ver 6.2.1 user docs available on the
Foxboro Customer Satisfaction website located at:
http://ips.csc.invensys.com/

Many of the questions asked recently on this list can be found there.  Are
there a lot of members on this list that don't have access to this website?
We use it on a daily basis but I think that you have to have a current
service contract with Foxboro in order to login to it.

Tom VandeWater
Control Systems Developr/Analyst
Dow Corning Corp.
Carrollton, KY

A value of SETTLE indicates that the previous response has not been settled
long enough, but the control error is not large enough to trigger a new peak
search.

FBTUNE is an extender block for a PIDA or DPIDA block. FBTUNE can be
connected or disconnected while the PIDA or DPIDA control block is running,
without reinitializing or interrupting the PIDA or DPIDA. This is done by
configuring the BLKSTA parameter of the PIDA or DPIDA block into FBTUNE's
PIDBLK parameter. If FBTUNE is connected to another PIDA or DPIDA, it is
automatically disconnected from the first. If another FBTUNE is already
connected to the PIDA or DPIDA, the second FBTUNE connection is not made and
the block is declared undefined.

55.3.1 Pretune
When PTNREQ is toggled on, Pretune applies a doublet pulse of amplitude
equal to BMP (or the distance to the nearest output limit, whichever is
smaller).
The pulse duration is made slightly larger than the process delay (dead
time). The width is determined by a MEAS change exceeding THRESH in the
direction implied by PIDA's INCOPT. The first MEAS peak must be at least 2.5
times the THRESH value.

When DFCT is not greater than 1, measured variable values and times at
peaks, zero crossings, and output-switching instants are used to calculate
the parameters of a gain-lag-delay process model.

When DFCT is greater than 1, these observations are used to calculate the
parameters of a gain-lag-lag-delay process model. DFCT is updated based on
the result.

These process model parameters determine PR_TYP and together with derivative
factor DFCT are used to calculate PBAND, INT, DERIV, DTIME, and SPLLAG,
using the algebraic tuning method. Pretune also specifies values for ITMAX,
ITMIN, PBMIN, and PBMAX.

When Pretune is completed, PTNREQ is reset. You can choose to have Pretune
leave the controller in Manual by having STNREQ 0. Otherwise when STNREQ is
1, the controller is automatically placed in Auto and Selftuning is started.

Pretune is aborted if the PIDA is placed in Auto, receives a request to
initialize or has a control error. Pretune is aborted and the controller
left in Manual at its original OUT value if the PIDA bump size is zero,
INCOPT is configured incorrectly, the measurement goes out of range, the
first measurement peak is less than 2.5 * THRESH, or the iterative
identification or controller design calculation fails to converge. 

Pretune does not calculate controller tuning if the first measurement peak
in percent of span is not greater than 2.5 * THRESH. This makes the process
deadtime a significant fraction of the Pretune pulse width, necessary for a
good identification. Pretune may not tune the PIDA, even when the BMP is
large and THRESH is small, if the process has very low gain. 


ITMAX 
Maximum Integral Time, in minutes, is a settable input that limits the
maximum integral time and the maximum time for the response peak search.
Default = 200 minutes.

-----Original Message-----
From: Neil Martin [mailto:neil_martin@xxxxxxxxxxxx]
Sent: Tuesday, June 17, 2003 4:28 PM
To: foxboro@xxxxxxxxxxxxx
Subject: Re: [foxboro] PIDA-FBTUNE configuration



If I remember correctly, the FBTUNE in Selftune mode must see at least 3
significant peaks, possibly 4, in a row.  I believe the FBTUNE's THRESH
parameter is used to define what a significant peak is.  Note the value
entered for THRESH is % of the PIDA block range.

Neil Martin
Huntsman Polymers Corporation
2505 South Grandview
Odessa, TX.  79766
ph) 432-640-8436
pager)432-742-4289
email page)9157424289@xxxxxxxxxxxxxxxxxx


 

                      <Budana_Prijadi@f

                      mi.com>                  To:
<foxboro@xxxxxxxxxxxxx>

                      Sent by:                 cc:

                      foxboro-bounce@fr        Subject:  [foxboro]
PIDA-FBTUNE configuration                                           
                      eelists.org

 

 

                      06/17/03 09:10 AM

                      Please respond to

                      foxboro

 

 





Dear I/A users,

I have been trying now for days to set up a PIDA-FBTUNE combo to
automatically tune (selftune) a real level control application in a mineral
processing rougher circuit (I am trying only for a first stage rougher bank
before copying to others).

The question I have is why if I get a confirmed Peak_1 or Peak_2 search and
then the measurement crosses the setpoint value, I always have a "SETTLE"
status?  This brings back to search for Peak_1.

In other words, I know there will be another Peak after crossing the
setpoint value, but the algorithm somehow starts again from Peak_1 rather
than continuing the next peak search.

Please have a look at the lower left quarter screen trend, where we know
that there is a pattern in measurement (cyan color) against the setpoint
(purple).  The output is given in white.

What FBTUNE parameters can I adjust to prevent "SETTLE" when measurement
crosses the setpoint value (line)?

Many thanks for any insights given.

Budana Prijadi
(GMT +09:00)

-- No attachments (even text) are allowed --
-- Type: application/x-zip-compressed
-- File: fbtune_settle.zip
-- Desc: fbtune_settle.zip




_______________________________________________________________________
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






 
 
_______________________________________________________________________
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
 
 
 
_______________________________________________________________________
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: