Re: [foxboro] NONLOP function IMPORTANT READ THIS
- From: tom.vandewater@xxxxxxxxxxxxxx
- To: jclaramunt@xxxxxxxxxxxxxxx
- Date: Wed, 19 Jan 2005 15:49:19 -0500
FYI,
Additional .NONLOP/ZONE Control info that is helpful.
Jaime Claramunt wrote to Tom VandeWater:
"Which one defines meas > setpoint zone, LZONE ? (assuming err=spt-meas)"
Jaime,
This is another good question that is not well defined in any of the
Foxboro documentation that I have read and the answer depends on what value
you have for .INCOPT
If .INCOPT = 0, (INC/DEC or Reverse control action), .ERROR value is
calculated as you say ie:
.ERROR = (.SPT - .MEAS)
example:
.SPT = 500
.MEAS = 550
500 - 550 = -50.0
If .INCOPT = 1, (INC/INC or Direct control action), Foxboro computes the
.ERROR value but reverses/changes the sign:
.ERROR = -(.SPT - .MEAS)
example:
.SPT = 500
.MEAS = 550
-(500 - 550) = +50.0)
.ERROR = +50.0
.LZONE always applies to a negative .ERROR value
.HZONE always applies to a positive .ERROR value
So you can see that the meaning of HZONE doesn't always mean High
Zone above setpoint and LZONE doesn't mean Low Zone below setpoint because
that changes when the .INCOPT value changes.
If .INCOPT = 0
.LZONE is active when .MEAS > .SPT
.HZONE is active when .MEAS < .SPT
If .INCOPT = 1
.LZONE is active when .MEAS < .SPT
.HZONE is active when .MEAS > .SPT
I know, COMPLICATED, but true!
Tom VandeWater
Control Systems Developer/Analyst
Dow Corning Corporation
Carrollton, KY USA
-----Original Message-----
From: Jaime Claramunt R. (Inforsa) [mailto:jclaramunt@xxxxxxxxxxxxxxx]
Sent: Wednesday, January 19, 2005 2:54 PM
To: tom.vandewater@xxxxxxxxxxxxxx
Subject: RE: [foxboro] NONLOP function IMPORTANT READ THIS
Tom,
Thx for this point...now I beginning to get it
Which one defines meas > setpoint zone, LZONE ? (assuming err=spt-meas)
Regards
Jaime
-----Mensaje original-----
De: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx]En
nombre de tom.vandewater@xxxxxxxxxxxxxx
Enviado el: Miercoles, 19 de Enero de 2005 16:29
Para: foxboro@xxxxxxxxxxxxx
Asunto: Re: [foxboro] NONLOP function IMPORTANT READ THIS
FYI,
User ALERT concerning .NONLOP differences between old PIDX/XE blocks
and new PIDA blocks!! Foxboro made a big boo boo when they built the PIDA
block. This may be because the folks that built it read their own
documentation for a PIDX block which incorrectly states:
"HZONE
High Zone is a connectable, settable, real input that defines, in terms of
the error, the upper limit of the zone in which the nonlinear gain option is
exercised. HZONE is expressed as a percentage of the measurement span."
HZONE in a PIDX/XE block is actually defined in Engineering Units instead of
percentage of the measurement span. It is expressed as percentage of the
measurement span only in the new PIDA blocks.
I know this because we are in the process of converting all of our
PID, PIDE, PIDX, and PIDXE blocks to PIDA in order to utilize some of the
new functions provided by the PIDA block, (i.e. setpoint ramping .SPROPT,
deviation alarm delay .DEVTIM, and ZONE control .NONLOP.) It really created
problems because I used the PIDX/XE block settings for .HZONE / .LZONE in
the new PIDA blocks and essentially lost control of the process.
There is a significant issue concerning the .NONLOP because the old
blocks HZONE and LZONE settings utilized engineering unit values and the
PIDA's use % of fullscale for HZONE and LZONE. An example that you can
easily build in an active controller follows:
PIDX block for FC0001 is a steam flow controller that has a range of 0 -
1,000 LB/HR
NONLOP = 1 (USE .HZONE, .LZONE, .KZONE SETTINGS)
HZONE = 100 (LB/HR)
LZONE = 100 (LB/HR)
KZONE = 0 (NO GAIN APPLIED OR OUTPUT CHANGE WHEN .MEAS IS LESS THAN 100
LB/HR ABOVE OR BELOW SETPOINT) (USE GAIN DEFINED BY .PBAND AND INTEGRAL
DEFINED BY .INT WHEN .MEAS IS GREATER THAN 100 LB/HR ABOVE OR BELOW
SETPOINT. i.e. OUTSIDE THE ZONE).
If .SPT = 500, .OUT = 50, and .MEAS = 500 the block output would not
change until .MEAS > 600 or < 400.
PIDA block replacement for FC0001 is a steam flow controller that has a
range of 0 - 1,000 LB/HR
NONLOP = 1 (USE .HZONE, .LZONE, .KZONE SETTINGS)
HZONE = 100 (% of full scale)
LZONE = 100 (% of full scale)
KZONE = 0 (NO GAIN APPLIED OR OUTPUT CHANGE WHEN .MEAS IS LESS THAN 100 %
ABOVE OR BELOW SETPOINT) (USE GAIN DEFINED BY .PBAND AND INTEGRAL DEFINED BY
.INT WHEN .MEAS IS GREATER THAN 100 % ABOVE OR BELOW SETPOINT. i.e. OUTSIDE
THE ZONE, WHICH IS IMPOSSIBLE).
If .SPT = 500, .OUT = 50, and .MEAS = 500 the block output would not
change until .MEAS > 1,500 or < -1000).
-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx
[mailto:foxboro-bounce@xxxxxxxxxxxxx]On Behalf Of Johnson, Alex
(Foxboro)
Sent: Wednesday, January 19, 2005 12:40 PM
To: foxboro@xxxxxxxxxxxxx
Subject: Re: [foxboro] NONLOP function
Please use the PIDA block and not the:
PID
PIDE
PIDX
PIDXE
blocks.
The PIDA does what the others do and much more.
Regards,
Alex Johnson
Invensys Process Systems
Invensys Systems, Inc.
10707 Haddington
Houston, TX 77043
713.722.2859 (voice)
713.722.2700 (switchboard)
713.932.0222 (fax)
ajohnson@xxxxxxxxxxx
-----Original Message-----
From: foxboro-bounce@xxxxxxxxxxxxx [mailto:foxboro-bounce@xxxxxxxxxxxxx] On
Behalf Of Brian Bates
Sent: Wednesday, January 19, 2005 9:35 AM
To: foxboro@xxxxxxxxxxxxx
Subject: Re: [foxboro] NONLOP function
Jaime
I use this to control the level of a tank, that has minor upsets in level
every 15 minutes when filters are backwashed.
When the backwash starts, the tank level starts to drop, and without the
NLOPT, the Prop. action would open the feed valve, which causes a swing in
the feed chemicals (not desirable). This configuration allows the
controller to (almost) ignore the backwash upsets.
The controller is configured as follows: NONLOP = 1, HZONE = 0.2, LZONE =
0.2, and KZONE = 0.1 (PBAND = 100).
When the spt deviation is between +0.2, and - 0.2, the controller gain is
(1.0 X 0.1) = 0.1 (1,000% PB).
If the error is larger than + or - 0.2, the gain is 1 (100% PB).
It is confusing because Foxboro uses PB for setting controller gain, and
the KZONE is a gain factor, so you must convert PB to gain to calculate the
effect of your chosen KZONE.
-Regards, Brian Bates.
"Jaime Claramunt
R. (Inforsa)" To:
foxboro@xxxxxxxxxxxxx
<jclaramunt@infor cc:
sa.cmpc.cl> Subject: [foxboro] NONLOP
function
Sent by:
foxboro-bounce@fr
eelists.org
01/19/2005 11:24
AM
Please respond to
foxboro
Hi List,
Does anybody knows how to work with NONLOP option on PIDXE block ?
Manuals are a little bit confusing...about HZONE and LZONE (percent of
span,
absolute value, of error,)
Do you have an actual case ?
Thx
Jaime Claramunt
INFORSA paper mill
_______________________________________________________________________
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: //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: //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: //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: //www.freelists.org/list/foxboro
to subscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=join
to unsubscribe: mailto:foxboro-request@xxxxxxxxxxxxx?subject=leave
Other related posts:
