Product Water Flow

What is the tank capacity? Have you checked the level?

If more is being removed from the tank than is being added, the level should be going down. If it is not going down, then the indication of what is being add or removed or both is not accurate.

tank level is maintained by closed loop control at 90 %. if production changes transfer pumps (vfd)will ramp up or down to maintain the level.

Have you verified tank level is being controlled acceptably?

i check the trend in dcs .level is maintained between 88-90 % all the time.

Have you personally checked the tank?

On second thought, if this is not a huge tank, there is no way it could be the cause of such a large discrepancy.

Looking back can you see when the divergence occurred if either the amount being taken out appears to deviate upward of typical, or the amount added appears to deviate downward of typical?

Are there other ways to get a measure of what should be flowing, like the feed input to RO minus the concentrate outflow?

My calculations: 14.5/14.3 = 1.014. My guess is that the total of the inaccuracy of the two meters sums to 1.4 percent. (Both could be within 1 percent if they err in opposite directions.) Check with you manual to see what the stated accuracy is for the meters.

Here is another thought: How have you determined this discrepancy? If your control system turns on input and output pumps to control water level, this reading would fluctuate up and down. Are these instantaneous readings calibrated to MIG per day, or are they actual volume of water per day? If it's an average of instantaneous values, variations from the control system could give you an apparent error just by chance if there are not enough data points.

Then the tank is too large. The only purpose of a tank is to allow the level to go up and down.

So you're running a 15 Million Imperial Gallon a Day reverse Osmosis plant and your financial management and system layout is so incredibly poor you can't shut down or isolate let alone do any on the go calibration checks and or adjustments to any key sensors in the system?

From what I understand RO treatment of bulk volumes of water costs about .5 - 1 cent per gallon and sells for at least double that so if you are doing ~ 15 million gallons a day that's a net financial profit value of ~ $150,000 - $300,000 dollars per day.

And a volume totalizing discrepancy between input and output flow seems like a bigger issue to be concerned with?

tcmtech

hmmm. you should join finance & management.you are wasting time in instrumentaion

Maybe the water is expanding and contracting....?

http://www.dummies.com/education/science/physics/how-to-measure-volume-expansion-due-to-temperature-increase/

And how is the reject stream accounted for?

>we are facing a problem from last week that Product flow of RO plant per day is 14.1 MIGD and product water transfer flow from tank is 14.3 MIGD.

What were the product flow and transfer flow values a week prior to the 'problem'?

yes.....

you need to know more about RO system

Please describe it, then, because it cannot be seen from here.

Elec conductivity of product water is more than 700 us/cm. according to farady's law of electromagnetic ,any conductor passing through magnetic field will generate emf.so that emf is proportional to flow of the conductive fluid. so we can use mag flow meter in RO system.

<...700 us.cm...> That's new, and pertinent information to this thread. Mentioning pertinent information in the original posting does wonders to keeping threads on-topic.

Quite.

What you said: " loose or corroded wire connections, or a coating on the electrodes could all contribute to less current from the mag meter, I'm surprised it worked at all..."

Appears to me that you were spot on with that answer without being presented critical information, way to go!

It is perfectly OK to use mag meter on RO permeate, provided there is sufficient TDS remaining (perhaps the source water is high brackish or seawater??).

One thing about mag meters, if you have the probe/meter located on a vertical run of pipe, there can be issues when the system is offline, that the meter will keep registering small amount of flow due to thermal currents, air bubbles, etc. in pipe.

It is preferred for absolutely the best results, to place mag meters on a horizontal run of pipe.

this is what instrumentation is. any thing can happened at any time. we need to resolve the problem and come with solution instead of poking and making fun.

Please describe it, then, because it cannot be seen from here.

Then CALL THE INSTRUMENT MAKER and get a field service tech out to the site!

You are taking the worst approach to solving your problem by hoping someone will guess what your problem is and tell you how to fix it for free.

Get a tech out there and be nicer to him than you are being to the forum!

Good luck getting help here with that attitude.!

You overlooked to compensate for tank draw-down, quite possibly. Another thing, please consider most meters are probably going to only be accurate to +/-3%.

3% of 15 =0.45 which is larger than the difference of 14.3 and 14.1.

It is either nothing or everything. If it is everything, then it has to be one of those little miracles.

Good. At least down-time has been found to enable basic maintenance and process improvement to take place.

Note that, on the basis of the information that the Original Poster has supplied to this point, no-one else in the forum was able to guess the cause of the problem apart from the Original Poster.

I'd guess a contributing factor was the original, inadequate piping design. It was designed on the cheap, with no provision for PM.

If the system is designed to be up full time, it does make it hard to keep up with sensor maintenance schedules. However, there are things they could do to help themselves (like have a meter bypass loop so meter run can be isolated, probe cleaned, replaced, etc.).

Like this?#12

Yes, at least we can agree on that.

Post 19, responding to yours...

Always good when OP solves their own problem.

You should do a complete statistical analysis of data to prevent fooling yourself, putting the sensors on a preventative maintenance schedule to ensure system metering accuracy.