Flow Pacing in Sodium Hypochlorite Chemical Feed System

[Yna]

Hello,

I have four metering pumps that widraw neat sodium hypochlorite from bulk storage tanks. Pumps speed will be paced off of the plant flow signal while the pump stroke length will be paced off the total residual chlorine.

I will control the pump's speed and stroke length in the PLC using PID controller. I am confused which parameters (for PV and SP) will be used in speed control. I have the following signals: effluent flow(plant flow), total residual chorine and total flow(discharge flow,the flowmeter after the metering pumps).
Please advice..

Thanks & regards.

 

[Saffa]

The Pump "speed" control will not be PID controlled. as you already say, it's proportional to the flow rate. So you need to know your maximum effluent flow rate, and then pump speed is simply 100% x (Measured Effluent Flow / Maximum Effluent Flow).

That's in simplest terms - you may need to introduce a constant into that as its likely that the pumps will be oversized, and 100% pump speed at 100% flow will result in overdose.

Then for the PID "trim" control, your PV is the total residual chlorine. SP is whatever you need the residual to be at the measuring point.

Well, from what I understand from your description anyway!

 

[Tom Jenkins]

..... Pumps speed will be paced off of the plant flow signal while the pump stroke length will be paced off the total residual chlorine....
..... and total flow(discharge flow,the flowmeter after the metering pumps)....

This makes no sense to me.

This is apparently a positive displacement pump. By controlling stroke or speed you are changing the pump flow, in a predictable and calculable manner. This can be accomplished by changing the stroke alone - there is no advantage to changing the speed as well. That would just add cost, complication, and potential failure points. KISS.

I assume that "discharge" flow refer to the chemical feed pump discharge. That isn't really necessary, but it is a reasonable precaution to verify the pump is working properly.

When I've done this i simply have a dosage rate, x ounces Cl2 per y gallons of water, for example. If this is something like potable water that's all you really need. If it is wastewater or something with variable Cl2 demand then you can use a PID or stepping control loop to adjust the dosage rate.

 

[Yna]

Thank you so much Saffa and Tom for your response.

Our metering pumps have an independent adjustable stroke and speed and designed to have 4-20mA output for speed and stroke. I just want to confirm if no need to consider the dose rate(determined by the PID trim control) in the calculation of pump's speed.

Thank you.

 

[kamenges]

Originally posted by Yna:

I just want to confirm if no need to consider the dose rate(determined by the PID trim control) in the calculation of pump's speed.[/qoute]

That is correct. You don't want to do this.

You are describing a system that in effect is normalizing your PID correction to flow rate by tying dosing pump speed to flow rate. If you didn't change pump speed with flow rate you would need to increase your correction (displacement) for a given residual chlorine with effluent flow rate. You already do this by tying pump speed to effluent flow rate and you do it as an open loop command, which makes it inherently stable. I like the concept.

The only thing I could think of that would make the system more responsive is if the effluent content is consistent enough that you could operate with some level of baseline displacement and trim on top of that. If history tells you that will have to add some amount of sodium hypochlorite anyway you might as well start adding it before you sense the need for it.

Keith

 

[Yna]

Thank you for the confirmation and explanation.

I confirmed the consideration of dose rate in speed calculation because i just confused on the formula that I have read on the other forum:

Dose (pump speed) = dose rate(determined by PID trim control) * flow rate*pump calibration.

where I dont know that pump calibration.

Thank you so much.

 

[kamenges]

That function assumes a constant displacement pump where the only control of dosing you have is pump speed. You have a variable displacement pump and are controlling both speed and displacement, which changes your control relationship.

Keith

 

[Yna]

Thank you very much.