Wastewater pH control Deadtime

[kdcui]

I am trying to impliment some pH control logic on a wastewater treatment system.

We are useing 93% sulfuric acid and 57% magnesium hydroxide as the adjusting agents, delivered by pumps controled by VFDs.

First stage is coarse adjustment of the pH in a 80,000 US gallon tank, the acid or base is added inline on a recirculation path that feeds the contents back into the tank.

From there the contents can go to a 8,000 US gallon 2nd stage tank for fine adjustment. Acid or base in added directly into the tank at this point.

Both tanks have agitators.

The wastewater will vary in content and compositon, and I don't have any titration curve data for the system. I don't want to use a PID loop, but I am unsure of what kind of deadtime to expect with the amount of wastewater in the tanks and neutralizing agents that are being used (I am familliar with the chemistry, but lack actual experience implementing pH control).

Also, I am unsure of how to calculate reasonable dosing amounts. I am told there are lookup tables for this, but since the wastewater will vary is this approach still valid? Or should I calculate an average dosing amount myself?

Any suggestions? I appreciate the help.

Thanks.

 

[leitmotif]

I am trying to impliment some pH control logic on a wastewater treatment system.

We are useing 93% sulfuric acid and 57% magnesium hydroxide as the adjusting agents, delivered by pumps controled by VFDs.

First stage is coarse adjustment of the pH in a 80,000 US gallon tank, the acid or base is added inline on a recirculation path that feeds the contents back into the tank.

From there the contents can go to a 8,000 US gallon 2nd stage tank for fine adjustment. Acid or base in added directly into the tank at this point.

Both tanks have agitators.

The wastewater will vary in content and compositon, and I don't have any titration curve data for the system. I don't want to use a PID loop, but I am unsure of what kind of deadtime to expect with the amount of wastewater in the tanks and neutralizing agents that are being used (I am familliar with the chemistry, but lack actual experience implementing pH control).

Also, I am unsure of how to calculate reasonable dosing amounts. I am told there are lookup tables for this, but since the wastewater will vary is this approach still valid? Or should I calculate an average dosing amount myself?

Any suggestions? I appreciate the help.

Thanks.

You said
The wastewater will vary in content and compositon
DO you mean volume and composition OR
it will vary in content composition and VOLUME and what about temperature? Then there is BOD COD dissolved solids (TDS) suspended solids etc etc.

In my opinion you need to learn what you have. I grant you since it is waste things will vary BUT you need to know what you are controlling. You need to invest in instrumentation and a data logger for automatic recording. Your best bet is
1. start reading the utility water meter or your own plant water meters so you can get some idea of volume.
2. If dumping chemical tanks an idea of volume how often dumped pH etc etc. Dumping an acid tank and a base tank at the same time can save you lots of chemical treatment.
3. Minute by minute changes in pH -- from this you can estimate hourly daily etc.

Dan Bentler

 

[Peter Nachtwey]

(I am familliar with the chemistry, but lack actual experience implementing pH control).

So you can predict how much of either chemical will bring the pH to the desired level? That is a huge start.

Is this a bleed and feed process or a batch process? I don't do waste water control

Also, I am unsure of how to calculate reasonable dosing amounts. I am told there are lookup tables for this, but since the wastewater will vary is this approach still valid?

Tables work. That is what we used for steam generator chemistry 30 years back. The technicians would sample and make adjustments every few hours. I would think more precise methods and continuous methods are available now. The formulas that were used to generate the table should be part of the PLC program now.

Or should I calculate an average dosing amount myself?

Any suggestions? I appreciate the help.

I suggest that you go through the calculations yourself so you understand the process. Then you can better describe the problem. Try some what if scenarios.

 

[kdcui]

Thanks for the replys.

I should probably mentioned the system has already been designed by another company and is under construction, and we are integrating the controls.

My background is Chemical Engineering, so I am familliar with the chemistry; given a pH and a volume I would likely be able to get a rough estimate on how much of either chemical to add.

The engineer who dropped this in my lap didn't give me much info on the system, besides the P&IDs. The contents will likely be a mixture of juice concentrate, water, and CIP (clean in place) fluid, and possibly other disslved content such as sugar. The process will be continuous, but the volume of the effluent into the wastewater system will vary with time, as will the proportions of the said fluids.

I haven't done much pH work outside of a lab in college, so even if I get the chemical dosage in the ballpark, I still am not sure of the dynamics (lag time between dosage and seeing a pH change, which will vary with dosage amount, strength of the acid/base and treatment volume anyways).

To my knowledge, the system is not monitoring TDS, BOC, COD, etc. and is strictly concerned with pH. As of right now, temp and pressue indicators are field instruments and are not being sent to the SCADA/PLC. Are these factors something that I should be concerned with (will it affect pH)? There is an auto sampler, but until the system is built and running the data will not be available.

It sounds like I will need information before I code anything. It also sounds like some of this might need to wait until start-up, when I am on site.

Thanks

 

[Tom Jenkins]

Please note that the buffering (alkalinity) of the wastewater will play a part in how much chemical you need to add to correct the pH. Also note that, by definition, pH is non-linear.

I avoid PID when I have a process that has a long response time. Peter's solution, with field adjustable time delays, will probably give better results.

 

[danw]

You might want to consult the authority.

McMillan wrote pH Control, published by ISA, in the mid 1980's. He's got a chapter on in-line and tank control schemes. An updated version, Advanced pH Measurement and Control is currently sold by ISA.

http://www.isa.org/Template.cfm?Section=Books3&template=/Ecommerce/ProductDisplay.cfm&ProductID=9943

The used book sellers like Amazon, Alibris, Half.com or Abebooks usually have copies of the older edition in stock.

Dan