Flow Control for Solution System

[rickco007]

Hi,

Need some advice on flow control for solution system.
I have a system that dispense fluid into a container and the dispense amount is determine by a fiX weight.
Currently, I am facing the problem in controlling the flow which resulted a wide fluid dispensing variance.
In the up stream, a diaphragm pump is use to supply the fluid.
During the dispensing, I observed that when the pump pump, the dispensing pressure and rate increase.
I am thinking of using a flow control valve to regulate the fluid flow.
Will that solve my problem? Is there a better way to regulate the flow regardless of the upstream pressure and flow? Thanks in advance for the advice.

 

[Latexman]

Your words can be interpreted many ways. Attach a sketch. Provide details on the pump and the weighment.

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[rickco007]

Hi Latexman,

Thanks for the advice.
I had attached a schematic.
Not sure if I have got all the symbols are correct.
The dotted line are electrical input and output feedback line.

Best regards
Rickco

 

[LittleInch]

That looks pretty good to me. The only other thing that might be useful is a flow meter to stabilise the flow or perhaps some sort of pulsation damper to reduce any pulses coming from your pump.

It is difficult to say much more without specifics, e.g. volume of the container, speed of delivery, accuracy of amount etc.

IN dispensing you sometime get a variable flow which slows down as the amount limit is approached tobe able to more accurately shut it off - a fairly simple logic loop in the PLC.

It's not entirely clear what your problem is so a bit difficult to say much more...

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[Latexman]

Details on the diaphragm pump - make, model, type, constant speed, variable speed, constant stroke, variable stroke, volume delivered in one cycle, flow rate?

Details on weighment - how much - lbs, kg, liters, cc, grams?

From start to finish, how much time is used for the weighment?

What is the fluid? Water-like or high viscosity?

Good luck,
Latexman

Technically, the glass is always full - 1/2 air and 1/2 water.

 

[Compositepro]

To get a system like yours to work accurately, the dispensing rate must be quite low close to the set point, in order to minimize the effects of measurement response time and variability in flow rate. The usual approach is to fill at a high flow rate to, say, 5% of target and then switch to a trickle, to hit target.
Another issue could be due to tubing design. Does a variable amount of fluid drain from the line after the valve closes?

 

[rickco007]

Thanks for all the advice, LittleInch/Latexman/Compositepro,

Hi LittleInch, a pulsation damper would certainly help.
I am actually thinking of a system that can dispense accurately to +/- 1gram of water with a range of supply pressure.
For example, if the system is installed in plant A, with a water supply pressure of 1bar and the same system is installed in plant B, with a water supply pressure of 3bar. I would like the system to be able to dispense within +/- 1 gram. Dispensing time is not critical. With the same simple loop program, will I be able to achieve that? Or I will need to fine tune the program each time the supply pressure changes?

Hi Compositepro, currently the dispensing logic is similar. When the program is written, I tested using gravity feed supply, which seems fine.
When I add a pump into the system, the dispense amount overshoot. When water supply pressure is at 1 bar and at 3 bar, the amount dispense differ. So I am thinking if I can make it in such a way that the system can be operated between 0.5-4.0 bar without any changes to the program.

I am thinking if I could added a pressure transducer after the variable control valve before the pinch valve and make use of pressure feedback to control the amount of water flow through the variable control valve.

Once again, thanks for the advice, really appreciate it.

 

[LittleInch]

+/1 g per how much in total? 1g out of 50 is doable, but 1g/ 1000 maybe not. at least in any sort of reasonable time.

I'm with compositepro here - I would regulate flow rather than pressure and a you get to within say 90% of your target weight make the program reduce flow rate gradually by 50% or more so that you have very good control over the last bit and can shut off with higher accuracy without compromising the time taken to fill.

Regulating pressure might be enough, but with a pulsating inlet pressure you'll struggle to get smooth flow without some sort of pulsation damper.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[rickco007]

Hi LittleInch,

Got it, I will try that out.
Target weight is approximately 135g +/- 1g.

Thanks for all the advice.

 

[LittleInch]

No problem. There are lots of dispensing systems available off the shelf though so not quite sure why you're doing this yourself, but good luck and let us know how it goes.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[itdepends]

You're trying to control the flow from a positive displacement pump using a control valve- that's not going to end well.

As the control valve closes in to reduce the flow the pump (assuming it is fixed speed) is still trying to deliver the same volume of flow- increasing the line pressure which increases the flow.

A recycle line from thet discharge of the pump back to the tank that is used to control the pump discharge pressure would help, a variable speed pump would be better, replacing the diaphragm pump with something more controllable (peristaltic, or centrifugal) would be better still.

As a chem eng/metallurgist the first part of any answer I give starts with "It Depends"

 

[rickco007]

As expected by itdepends, we could not control the dispensing rate well enough as the fluid supply varies from 3.5 to 4.5 bar.

Unfortunately, we do not have the option to change the pump or add a return line to the tank.
From my understanding, a pulsation dampener needs to be install as close to the pump as possible.
What is the effect if I try to install the pulsation dampener near my equipment which maybe 500m away from the pump?

I am wondering, by adding a pressure transducer and a intelligent controller to automatically control the supply flow using a variable control valve could help to damper the change of pressure?

 

[LittleInch]

Itdepends got it right and I missed it - your pump is essentially a constant flow pump so unless you either control the speed or allow the flow somewhere else to go, trying to control the pressure / flow or stop the flow will just result in higher an higher pressure until the motor stalls....

You really need to change something fundamental here otherwise it will just flow at the same rate an in slightly pulsating way.

The only other way perhaps is to pump into a local tank which is above your dispensing unit and use gravity to feed the dispenser?

So either change pump configuration, change pump type or introduce a break tank or feed it from the mains water supply.

My motto: Learn something new every day

Also: There's usually a good reason why everyone does it that way

 

[rickco007]

I do agree with both of you. Something need to be done in order to solve this problem.
The local tank concept will definitely work, but I will need to add in features for sanitization.
Thanks for all the advice and idea.