Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

Flowmeter Issues with PD Pump 1

Status
Not open for further replies.

polyroly

Chemical
Mar 29, 2007
34
0
0
CA
Hi Everyone,

We are having issues at our facility with a mag meter and a metering pump. The pump is an electric diaphragm pump with adjustable stroke length used for metering ferric sulfate into a tank.

The problem is that we aren't getting accurate readings from the flowmeter because the flow isn't constant. The stroke length is backed off as much as it can be but it doesn't solve the problem. Ideally I could get a smaller pump.........but I thought I would ask some smart people if there is a fix in mind :)

The flowmeter is installed in the vertical position.......

Thanks!!
PR

 
Replies continue below

Recommended for you

Mag meters (like all meters) have an inherent latency. This can range from a few milseconds with square-edged orifice meters to tens of seconds for a turbine meter. You are giving your mag meter a stroke every so often and it has to run to go from no flow to flow, the error is the latency period. Not a lot you can do with the technology you are using. If flow rate is critica, I'd put a scale (for a horizontal tank) or a head switch (for a vertical tank) and use changes in tank contents to control the process.

David
 
Thanks David. The reagent flowrate in comparison to the vertical tank volume is very small so I don't know if there would be an instrument sensitive enough to detect a change in level and convert that into an accurate flowrate.
 
Most metering pumps are designed to be more accurate than most flow meters and are therefore often run without metering. Perhaps you should focus on making sure the pump is calibrated. A positive displacement flow meter would work. They have a design very similar to metering pumps.
 
Have your PD pump discharge to small tank with an overflow line. Overflow line small enough to maintain a constant flow. Put your meter on the overflow line.
 
First make sure that your flow meter position (vertical) is correct. It should but you'll never know...

What is the stroke rate of your metering pump? How many strokes per minute (approx.)? The pulse flow characteristic increases with low stroke rates.

A pulsation dampener might be a solution, too. But that only makes sense if pressure and flow are high.
 
With all due respect, I would not generally suggest installing a PD meter in series with a PD pump with a strongly pulsating flow such as one might get from a single diaphragm pump or indeed where there is a strong pulsation.
This pattern of flow is likely to cause excessive wear and ultimate failure of the meter.

For the most part today we need to look to the flowmeter electronics which are generally much more sophisticated than previously but which may trade off the primary element response time for a longer update time in the electronics.

Then too some meters are inherently slow responding, but not turbine meters.
Turbine meters are actually very fast responding, something like 5-7msec to 95% of a step change in flow and it is why they are used in jet engine test rigs where they require the meter to be able to handle the SNAP tests and turbines were originally used for rocket applications being able to handle the high G forces and high vibration environments.


JMW
 
The flow can be accurately measured from the operation of the pump, via runtime or pulses whichever is easier. You may need to do an initial calibration with a bucket.
 
Dear All,

I would like to ask you to clarify me the below:

For most Mag meters manufacturers they type ( medium control: separate electrode???) what they mean by that??

Thanking you

Ramy
 
The electrodes of a MAG meter are used to induce the voltage. "medium control: separate electrode" means that there is any additional/separate electrode which is used to monitor a minimum flow (= low flow detection).
 
Thanks micalbrch, As I understand that for MAG meter which I spoke about has 2 electrodes so one will induce the voltage and the other will detect the low flow? Am I right?
 
You can install a pulsation dampener on the pump discharge line. The pulsation dampener will help to lessen the peaks in flow.

However, a pulsation dampener will probably not completely fix the problem.

You should consider adding a totalizer to the flowmeter output. The flowmeter will accurately measure the total flow, it will just not display the average flow over time.

If you totalize the flow over a set time period, you can calculate the average flow over that time.
 
You are probably discharging into a system without much back pressure, which is somewhat helpful.

A properly sized and selected pulse dampener will solve the problem completely. If the flow accuracy really matters, verify this with a calibration column on the suction side.

If you had a higher discharge pressure, you'd have check valve efficiency to worry about as well. Think about it: one of the reasons that real PD pumps deliver less volume per stroke as the discharge pressure increases is due to check valve leakage, which implies reversing flow in the discharge line. Without a proper pulse dampener, you're asking a lot of any type of flowmeter to give you an accurate measurement of the net forward flow for a stream which changes direction every stroke.
 
Thanks everyone for your help. I think I am going to try the pulsation dampener option to see if that helps the problem. We need instantaneous flow indication because the pump is on ratio control based on the flowmeter measurement.

Moltonmetal you are correct about the discharge pressure. The pump is connected to a large atmospheric tank. I was contemplating throttling the discharge to see if this had any impact.
 
Status
Not open for further replies.
Back
Top