Flow Transmitter/Switch Problem

What does the Rosemount technician say when you relate the problem to them?

They can also educate you on flow switches to, as that seems to be something you need as well.

You HAVE spoken, or communicated with them, right? And they have already suggested thinks to try, right?

And you have documented the things you have tried so far, right?

Tell us what you have done, if anything, to solve this problem already.

I'd just use a different flow sensor for each fluid. Seems simple enough.

Plumbed correctly, you could keep the sensors flooded with the two different fluids. You could even switch between fluids automatically, with the proper controls.

If your photo is labeled correctly, then the flow is vertical up, correct?

Several things come to mind:

1) If the damping is set too high, getting a flow rate that is step function up from zero flow rate could take a long time to appear. Damping setting defaults at 2.0 seconds. It can be as high as 255 seconds.

2) 1m3/hr = 4.4gpm
Are you sure you have a minimum flow rate you think you do? The meter body size is not mentioned. Here's the table with minimum flow rates for various body sizes:

3) This is speculation: although absence of a full pipe should appear as a no flow situation, one can't be sure with all the modern micro this and micro that. If the meter can detect a not-full pipe somehow, it could go into fault mode and coming out of fault mode might take a full re-initialization, such as on power up.

4) This is speculation: I don't know what pushing air out of an empty pipe does to a vortex meter set up for liquids. I'm sure the meter can somehow sense the vortices, like it does for steam or gas. Air bleeds might throw the meter into a fault mode if the vortex rate is somehow off the map for the frequency the meter uses for liquids and what you're seeing is a recovery time delay.

The note presented by Shokistan start out in the first paragraph, that the meter is very sensitive to disturbances up and downstream.

What is filling and emptying a pipe, than a huge disturbance?!? Until the new slug of liquid clears up, no meaningful measurement will take place rendering the setup imprecise or worthless.

The maker have tested the meter. I have not seen a setup with drained pipes, nor with high turbidity due bubbles. I would take that, as a strong warning, not to deviate. If this is the case the maker's techs rightly decline to get involved.

My take on the abuse of electronics is, that analog can latch up, slow or never to return. Digital may go into sleep, with a loong return to normal, after a long selfcheck and calibration. If there is a power piezo transmitter is there, it overheats, goes off frequency and may even fail. A well built meter prevents that. Add more selfchecks, before returning operational.

Solution: switch the liquids wet. It may take 2,3or4 valves. The meter will function without interruptions, and the measurements will be correct.

Go in peace, and sin no more against manufacturer's recommendations.

Yes...the flow is vertical up.

Damping is set at 2.0 sec

Other transmitter prarameters are as follows. Please have a look over it & suggest if some thing seems suspecious in it :?:

LRV :0.00 m3/h

URV:5.00 m3/h

PV unit:ACMH

Xfer function:Linear

Reference K factor:295.04

compensated K fact:294.922

Install effect:.00%

service type:Liquid

LFC:0.771 ACMH

Low pass corner freq:6.752000Hz

Filter Trig level:4

Proc density units:kg/cm

Process density:994.274

Pipe ID units:mm

Pipe iD:26.645 mm

Pulse output mode:0.0 off

Pulse scalling :0.00 ACM

Flange type:wafer

Body type:wafer

wetted material :Hast -c

meter body#:107282

Body #suffix:A

Base temp units: deg F

Base temp:59.00

Base press unit:PSI

Base press:14.701

Base compress:1.00

Process temp unit:deg C

Process temp :35

Process pres unit:bar

Process press:3.81

Proc.compress:1.00

Densily Ratio:3.52

Base vol unit:gal

conversion number:1.0000

Base timer unit:min

Alarm jumper:Hi

Write Protected:NO

Manufacture:Rousemount

Where do you get Dematerialized Water? New York could have used several million tons of it about a month ago!

Hi!

As you want to monitor the low flow or no flow condition, try using

Flowphant T DTT31 from E&H

Interesting. This problem seems to stem from insufficient Process Engineering consideration at some point in the past.

Correcting the typo dematerialise demineralise, several quations come to mind:

• How did the plant get built like this without considering this possibility during a full HazOp Study beforehand?
• Why is there a need to swing between two extremes of materials in the same line? Why were they not segregated into two separate lines, each with its own flow transmitter appropriate to the fluid passing through it?
• Why is there a perceived need to drain the line before swinging to the other fluid? What is the risk of these two fluids blending at varying levels before arriving in the process?
• What is the material-of-construction of the line and everything upstream? It wouldn't be the first time corrosion products cause malfunction in a flowmeter, as indicated in the original posting.

Search the web for different types of flow detection switches.

There are types that work on detecting the flow from outside the pipe:i.e. not installed in the stream but from outside, relying in the presence of some bubles or impurity etc. Others use temperature differential or some on accoustics ...

Mechanical float switches performance is very dependent on installation details: Horizontally, amount of obstruction created for the flow to enable it to operate at a particular rateof flow...

Why does the line size change at the flowmeter? What turbulence effects might be expected at an abrupt change of diameter?

If all you need is a simple flow detector, use a pressure switch.Mount it in a low trap to insure it stays wet.Trip sensitivity is up to you to decide when ordering.Sounds like the Rosemount is a bit of overkill for the requirement.

As for the Rosemont installation, a minimum upstream-downstream pipe diameter distance is specified in the installation manual.If this distance is not physically possible, there are straightening vanes available to give a laminar flow thru the meter.

Inside of the meter, there is a triangle (arbitrary description ) shaped obstruction, that creates vortices, like the swirls you see in water flowing around rocks.The number of swirls is linearly related to the velocity of the liquid( and a few other factors which I will not delve into here.)

There is a pick up unit installed above this vortex-shedding device that translates the pulses into an electronic signal that is processed by on board (or remote) electronics to indicate engineering units of the measured media.

That is a lot of technology to simply detect the presence or absence of flow.

A switch is much simpler and easier to maintain.

Good luck.

The manual is probably so prescriptive about the diameters upstream and downstream and has probably completely overlooked the possibility that someone might change the line size abruptly at the meter itself...

Luck donated onwards to the original poster, who would seem to be needing it.

According to my understanding, Pressure switch can not be a replacement of flow switch. Because if flow gets 0 due to any blockage in line then pressure would remain present in the line which would not detect absence of flow. It can lead to potential risk situation.

Can you suggest any suitable type of flow switch that can work on both - Demin & Sea Water.

Anything made of plastic. What about the Platon Flowbits catalogue <usual disclaimer>?

'Pressure switch can not be a replacement of flow switch'...'potential risk situation'.

Your pumps / motors should have pressure switch cut-out protection. In the event of no-flow, your 'blockage' could be on the suction side of the pump, or even if the 'blockage' is downstream, you still don't want the pumps to run, right ?

Install a DP transmitter or Pressure Switch across the pump. Zero DP = zero flow = cut-out signal to motor bucket. Also, HH DP = zero flow = cut-out signal to motor bucket.

Intriguing requirement. Please describe your process. Why produce de-min water at considerable expense, and then run it through a line contaminated by sea-water ? Fits description for a naval craft that demineralises it's own water for steam / cooling service...nuclear ship ?? It's the only likely process scheme that I can think of ...you can only be flushing the de-min water overboard, aren't you?

the subs do something low flow rate measurement (i'm not shure + ?central paciffic?)

The Emerson 8800 meter requires 15 times the meter's inside diameter of uninterrupted, undisturbed, no turns, no fittings flow upstream of the meter inlet in order to function without error.

In addition the meter requires 5 times the meter's inside diameter of uninterrupted, undisturbed, no turns, no fittings flow downstream of the meter outlet in order to function without error

All this being said; This exceptional meter has a tuneable "K" factor that you can manipulate to offset some of the self-inflicted disturbances within meter design limitations.

Below is a path that will take you to the IOM proceedure and hopefully solve the problem. If not, you will have to contact Emerson and seek on-site field technical service to solve your problem(s).

http://www2.emersonprocess.com/siteadmincenter/PM%20Rosemount%20Documents/00816-0100-3250.pdf