pH Control

  • Thread starter Thong Phat Co., Ltd.
  • Start date
T

Thread Starter

Thong Phat Co., Ltd.

Hello,

Waste water are discharged from process having a fluctuation pH from 3 to 12. I prefer to have a simple and cheap pH controller (one pH sensor, one
controller but 2 pumps with flowrate ~20 litters/hr) to neutralize this water by HCl 10% when pH >8 and by caustic 10% when pH <6. Please advice some manufacturers for this application.

Thank you.

Hung Lam
 
B
With such fluctuations (3-12 pH) a large equalizing basin might save you a lot of caustic and HCl through self-neutralization.

Bela
 
R

R A Peterson

get your high school chemistry book out first. I doubt you will be able to control such a wide range of pHs with just 2 pumps. the flow variations required will probably make it near impossible.

Bob Peterson
 
J

Johan Bengtsson

Why impossible?

The tolerances don't seem to be that hard to meet, if this is possible or not is lagely depending on the rest of the setup. If the mixing is done in a relatively large buffer tank the control could quite possibly be a simple on-off-on. There will be very different flowrate needs but that don't necesarily mean it is impossible.

Only controlling the pumps might be hard to do continious, but a pump + a siutable valve and a split range control can handle very different flowrates.

Some questions to be answered by the original poster:
1. Do you have a buffer tank for the mixing?
2. If yes - how large is it compared to the flow?
3. How fast do the incoming pH vary?
4. What can you accept in outgoing pH variations (limits and change speed)?


/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
J
We send all of our process waste water to a large holding tank (approx 2500 gallon). rather than using a ph sensor, which we have found to be to touchy for our processes (there is a lot of liquid
cheese that tends to gum things up), we use a conductivity sensor. Admittedly, we use fairly clean water to start with. ie: water going to the boilers first runs through a reverse osmosis filter. However, we have found that straight water is fairly low conductivity, Acid is higher by about 1 order of magnitude, and caustic is about 2 orders of magnitude higher than neutral water.

You may want to consider conductivity sensing of the process water is not fairly clean.

Just me $.02

--Joe
 
T

Thong Phat Co., Ltd.

Dear Mr. Johan Bengtsson and other list members,

Thank you for all helping. Actually, we do not expect that we can control accurately the pH of waste water with only pH sensor, controller and two pumps.

The is a small buffer tank, about 6 m3. The flowrate is about 20 m3/hr. The pH does not change fast in a stable process, but when we
shutdown or start up a process or have a process fails, pH can change suddenly: 1/ pH lower or higher or 2/ Changing from low to high or vice versa. If such action happens, I believe that we have to use university chemistry book, not high school one as Mr. PETERSONRA mentioned, plus flowmeters, modulation control valves and a
complicated controller to handle it and we may forget this system soon!

Larger buffer tank is useless in the first case, but it will save us some chemicals. The expected pH is not often out of pH range 6-8. That is the reason we would like to look for a simple pH controller having two settable or programmable relays to control two pumps. I am doubtful that there is no manufacturers to make it!

Regards,

Hung Lam
 
R

Roderick Duet

I am currently using a pH controller from Metler Toledo. There are enough auxiliary contacts available to interconnect to a PLC for more control of pumps etc. The model currently used is pH transmitter 2100.
 
J

Johan Bengtsson

Those figures gives you an exchange rate of the water of about 18 minutes. This means the buffer tank can be considered a low pass filter with a time constant of these 18 minutes. I have not
made all the calculations necesary to check what this means for the non linearity of a pH control but it definitely sounds like a possible thing to do. Buy yourself a pH transmitter and two on/off controllers. One controlling each pump and set their setpoints placed somewhere just inside the pH limits you want. The hysteresis could probably be kept moderately low, it does depend on noise levels and how often you can accept the pumps to start/stop while they are working. A higher hysteresis will make it more probable that the controllers will act against each other and that would of course be a complete waste.

The larger you make the tank the more chemicals will be saved if the pH vary much both above and below the average during start/stop. It will also save the relays and the pumps since the start/stop frequency for each pump could (and would) be lowered.

If you are not aware already: Check how often you have to do maintenace on the pH sensor...


/Johan Bengtsson

----------------------------------------
P&L, the Academy of Automation
Box 252, S-281 23 H{ssleholm SWEDEN
Tel: +46 451 49 460, Fax: +46 451 89 833
E-mail: [email protected]
Internet: http://www.pol.se/
----------------------------------------
 
W

Wayne Shimanis Vortex Technologies, Inc.

Mr. Lam:

You may want to check with Devar. Inc. The Model 3265 pH contoller offers (2) 10amp relays that can be set to the desired parameters. You can find out more at their web site http://www.devarinc.com. Good luck with your project.

Wayne Shimanis
Vortex Technologies, Inc.
 
A

Al Pawlowski, PE

If you are looking for continuous PID type control, rather than just bang-bang, look for what is called (at least I do) a "floating"
controller. This type usually has a PID algorithym that operates two separate relay output. Whenever there is an error, one of the outputs pulses closed with a duration proportional to the error and a frequency that is adjustable. One output is associated with positive error; one with negative. Often these controllers also have adjustable setpoint
dead-band. Some even have non-linear gain which can be helpful in pH control.

These units are often used with motorized electric operators on controls valves, gates and dampers. The last ones I used were from F & P and
Honeywell. Honeywell called theirs a "3 step" controller. It is also easy to program that kind of function. I have done it for flow control with Rugid Computer RTU's.
 
B
Neutralization by alternate starting of acidic and basic reagent pumps is not likely to give good control if the influent pH varies outside the 5 to 9 pH range and/or if the agitator pumping capacity on the equalizing basin is
incapable to lower the dead time of the loop to 10 seconds. I am afraid that the moneys you save on tankage, agitation and controls will be spent on wasting neutralizing reagents (not to mention the discharging of off-spec effluent).

Pages 1328 to 1361 of my handbook give some 20 batch and continuous pH control schemes of gradually increasing sophistication for all possible services.

Bela Liptak
 
B
In a way neutralization is similar to controlling the temperature of our home within a comfort zone. Say we start heating at 68 F, start cooling at 78 F, and inbetween we let the temperature float without spending any energy to change it. The on-off control of the heat and coolant sources is feasible, because the controlled process (our home) has a large heat capacity (low
gain) and the relationship between energy and temperature is linear.

What makes on-off control of pH very difficult (and unlikely to succeed) is that it is neither: Near neutrality the pH process is very HIGH gain and is nonlinear (logarithmic). The amount of neutralizing chemical needed to bring the process from say a pH of 2 to 3, is 10,000 times the quantity needed to bring it from a pH of 6 to 7.

Therefore, if the neutralizer pumps are sized for the maximum demand, they are drastically oversized when used near neutrality. One way to visualize this is to consider what would happen to the comfort control in our homes if instead of a 10 degree comfort zone, this neutral zone was reduced to 1/1000th of a degree.

It is this high gain and logarithmic nature of pH which necessitates the complex control strategies described in my handbook. Neutralization (unless
done in a batch manner in very large tanks) in general is not suited to on-off control and if one insists upon using it anyway, the waste of
neutralizer will rise and control quality will drop.

Bela Liptak
 
J

jaideep lahiri

Hi!
I feel that the Yokogawa new PH20 sensor with the 4-wire transmitter PH402G can suffice.
Jaideep
 
C

Controls Engineer

Our application is difficult because we will have to neutralize several different types of chemicals in a day. It is a continuous system that has wildly varying titration curves daily. We are attempting to perform the control in an Allen-Bradley PLC system ... The tough part is that they do not have a non-linear PID equation. We may have to create one from scratch. Or at least try to create some SP and PV characterizers. Any ideas? Thanks
 
I am going through the same scenario. Did you come up with any solutions? I would appreciate any advice. I was trying to create a SP and PV characterizer but would rather not if there is an easier way.
 
Good question,

I too look forward to the gurus on this list stepping up to the plate with some ideas on this one. We have similar scenarios also involving PID control in A-B PLC's.

A similar problem is where we control temperature in a reactor by splitting the CV from a PID between a steam valve and a cooling valve. Of course, the CV is non-linear. I wonder if a solution to the pH control question might lead to ideas to help with this one as well.

Mike Ryan
Aerojet Fine Chemicals

>I am going through the same scenario. Did you come up with any solutions? I would appreciate any advice. I was trying to create a SP and PV
characterizer but would rather not if there is an easier way.<
 
K
On Wed, Sep 25, 2002 at 03:16:16PM -0400, Anonymous wrote:
>
> On March 22, 2001, Controls Engineer wrote:
> > Our application is difficult because we will have to neutralize
> > several different types of chemicals in a day. It is a continuous
> > system that has wildly varying titration curves daily. We are
> > attempting to perform the control in an Allen-Bradley PLC system ...
> > The tough part is that they do not have a non-linear PID equation. We
> > may have to create one from scratch. Or at least try to create some
> > SP and PV characterizers. Any ideas? Thanks
>
> I am going through the same scenario. Did you come up with any
> solutions? I would appreciate any advice. I was trying to create a SP
> and PV characterizer but would rather not if there is an easier way.

Anonymous and Engineer,

Could you perhaps map a function of some sort to the PID output and input?

--
Ken Irving <[email protected]>
 
Top