Re: [foxboro] pH control
- From: "rys" <rys@xxxxxxxxxxxxxx>
- To: foxboro@xxxxxxxxxxxxx
- Date: Mon, 15 Mar 2010 12:52:41 -0400
Bob,
There are specific issues with using Lime for pH control that
could be playing a role here. The lime slurry is a solid suspension
that must dissolve before it can react. pH reactions for ions in
solution occur exceedingly fast, but reagents need to be dissolved and
mixed first. The solubility of Calcium vs, Magnesium in the lime can
also play an important role. High calcium limes cost more but
dissolve faster. One thing to look for is continued reaction downstream of
the pH control system - i.e. the pH continuing to rise due to residual
lime finally getting into solution. The fact that the system worked
satisfactorily at one time, means something has changed:
1)
Lime composition (Calcium/Magnesium/ and carbonates (limestone content)
may have changed).
2) Particle size of the solids in the slurry has
changed- finer particles dissolve faster, lumps not so fast.
3)
%solids in the slurry has changed. Weaker solutions work better
because more of the lime is already in solution.
4) Mixing in your
process may have changed - even when reagents have dissolved, they must be
mixed before they pass by the pH electrode.
5) pH electrode location
or response time may have changed. Lime can coat electrodes and slow
the response.
6) The buffering in the waste stream has changed - more
buffering near the setpoint is easier to control. This could be caused by
various operational changes like removing certain materials from the
influent stream. Preparing a new titration curve and comparing it
with the old curve would show if this were the problem.
There are tons of things that can be done in the IA configuration and
the PIDA block with the NLNBLK option (more precise than using the NONLOP,
and the setpoint can be changed without redoing the gains) is only one
example, as that feature was specifically developed with pH control in
mind. Few people have configured this. As others have mentioned,
feedforward schemes could also help. In general understanding your
process behavior will require some laboratory work. Fixing or
understanding field problems is often more effective and longer lasting
than adding complexity to the control scheme.
Rick Rys
P.E.
www.r2controls.com
508-339-6633 office
508-369-5186
Cell
_______________________________________________________________________
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:
