Dear Roy Ji,
You may be right when you say that something is flashing off inside the cell so that the diaphragms are loosing contact with the measuring element. But how can I add a purge to the wet leg as both the measuring as well as reference (wet) legs are under suction and not positive pressure.

Regards
singhbvirk

 

Dear Roy Ji,
A. I saw the level go down on the DCS while local gauge glass indicated actual level in the tank. I checked the DP transmitter output using smart field communicator and found it consonant with the DCS reading.

B. Calibration is 4mA with empty tank.

C. LP connects to wet leg and its fine for all other cases except problem being faced in two instances.

D. Yes, in the DP transmitter itself. LRV is set at -1200 mmwc and URV at 0 mmwc.

Regards singhbvirk

 

Dear J. P.,
The water inside the tank is condensed from steam after it has worked in the turbine. The level of liquid is maintained for pumping it to some other tank located at higher elevation.

Yes, we have level glass installed.

Temperaure and pressure inside the tank are almost steady at about 40 deg.C and 0.15 kg/cm2 (absolute). It is the LP side on top of the tank and its working fine in case other numerous DP transmitters. Measuring range is 0-1200 mmwc.

Regards
singhbvirk

 

Dear NG,
1. Yes, the HP side is connected to the bottom of the tank and LP to the top of the tank.

2. Yes, these reading are about 0.15 kg/cm2 and 40 deg.C.

3. We have checked the wet leg and found the level to be OK. But when we checked the impulse pipe connected to lower side of the tank, we found to our surprise the level to be lowered.

4. We compared the level at DCS lower than that at the local gauge glass.

5. mA signal inversion is being done in the DP transmitter itself.

6. The DP transmitter is not HART based but its LRV is set at -1200 mmwc and URV at 0 mmwc.

Regards,
singhbvirk

 

Dear Singhbvirk,

-1200 mmwc sound as a perfect number,

I think that you must check with the Honeywell SFC how much is produced by the wet leg water colum , I think that you will see a value around the -1200mmwc, maybe a little more or less, but no exactly -1200 mmwc

Maybe the following can improve the level reading

1.- Open both filling tee to the atmosphere and check that the liquid in the wet leg be at the middle point of the tee installed in the wet leg, this mean put the HP side to atmosphere and the LP side to the atmosphera but with the wet leg filled, in our case condensate water right?

2.- Check in the Honeywell SFC how many mmwc is sensing the transmitter, then use that value as your LRV, this number have to be close to -1200 mmwc, maybe -1215 or -1198 or something like this numbers

3.- Introduce the URV value making the level calculus, for your case 0 mmwc

4.- Put on line the transmitter and check the level reading

Hope this help
Regards

J.P.

 

singhbvirk,

Some additional checks:

1. Check whether you have an air leak on the dry leg somewhere between the main tank and the transmitter. Check all joints. An air leak in this leg will give a lower reading. The problem is finding the leak under vacuum, as the air is going inside the pipe.

2. Did you change the transmitter with a new one, to see whether the problem persists?

3. If you break the vacuum on the transmitter, i.e. operate it under atmospheric pressure, does the transmitter retain its calibration?

 

Singhbvirk,

You said
> 3. We have checked the wet leg and found the level to be OK. But when we checked the impulse pipe connected to lower side of the tank, we found to our surprise the level to be lowered" <

I think that's the clue. If your transmitter is mounted below the low tank connection and there is air in the impulse tubing of course the reading will be low.

Next thing, where is the air coming from? I would suspect a leaky fitting. Because the tank is under vacuum, the air sucks in rather than the water leaking out. As the air displaces water, the indicated level drops. It could also be gas coming out of the water.

There are a number of things you can do. If it's a leaky fitting, fix it. If it's gas coming from the water, add a small water purge or relocate the transmitter so the measuring element is at the same elevation as the Low tapping point. Re-arrange the impulse piping so that any gas can enter the tank to be replaced by water.

If you need to add a purge, use a small needle valve and rotameter (one for each side) tapped into the tubing or connected where the air bleed/vent valve is screwed into the transmitter body.

Send me a sketch of your installation and I will advise you further.
roy_matson at yahoo dot ca

Regards,
Roy

 

Hello, dear Singhbvirk.

I have been reminded by latest post by roy, yes you should check for leaking at process connection from the tank to the transmitter and not forget also check the signal output of the transmitter is equal to decreasing and increasing of the liquid inside the tank, some reference such as side glass or pressure gauge are needed. If there's a leak try to fix it, that's simple. If the output of the transmitter are equal to the status of the liquid then make sure if when liquid are at low level it should be 4mA at 0 mmWC or empty and 20mA at around -1200mmWC. And that is make me remember something, in what manner you want the Transmitter measure the liquid inside the tank is that when the liquid level are going up the reading have to goes up at the transmitter display or at DCS/PLC or instead if the pressure (Vacuum) goes to rise then the reading are increasing? In that kind of manner you need to make proper adjustment to the transmitter if your measuring system is the first or normal and common level measurement, then treat the transmitter to measure level by the principles of common DP as you know the HP to the bottom and the LP to the top and the LRV set at 0mmWC and URV at the pressure of liquid when it full (I don't know exact nominal of the pressure you should check it your self) so if liquid goes up then the pressure at bottom should increase and the reading rising save only when there's a leak. Second, if you prefer the second manner then arranged the transmitter at opposite sides where's LP side to the bottom and HP side to the top, then give adjustment to the LRV at 0mmWC and URV at around -1200mmWC so when the Pressure (Vacuum) increase then the reading is rising. There's some cases where the URV adjustments are given 0mmWC and LRV to -1200mmWC it depends on the manner of the zero position, roy had mention that in another post is the suppression or elevation. That's my own opinion based by information give by others. I hope this can help you figures out something even a little. And by the way where's Walt? It's about time he gives some comment.

Thank you.

 

Dears,

I express my gratitude to all the caring fellows who contributed their knowledge to the query posed by me. The problem has been solved. There was air ingress from the equalising valve of 5/2 way manifold connecting DP transmitter to the standpipe attached to the tank under vacuum. It was found that this air ingress was causing fall in the level of water in the impulse line connected between DP transmitter and the bottom of the tank. The ingress was detected by isolating the standpipe from the process and applying external positive pressure by using mechanical pump.

Regards,
singhbvirk

 

Singhbvirk,

That's good news. Thank-you for letting us know what you found. Hopefully we all learned something in the process.

Regards,
Roy

 

Dear Mr.Singhbvirk

We want to install the level DP transmitter for the similar application in power plant. I would like to have your help for the installation & calibration of the DPT.

You are requested to contact at control.com or
at instcontrol1 at gmail.com

 

Hi there,

Send us the details of what you need to do. We need as much info as possible in order to give a detailed explanation of the steps to follow for a successful installation and calibration. With vacuum things can get a bit hairy but how hairy will depend mainly on what you use, so the details of what you have available and can use is important.

We need info on at least the following as well:

1) What is the model and make of the transmitter you want to use?

2) What is the ID and size of the vessel and what will the normal operating vacuum be?

3) What type of DP transmitter you plan to use? Tubing or capillary type?

4) The details of the tapping points on your vessel in relation to vessel bottom.

5) The range of the transmitter you plan to use. For example -1000mmWC to +1000mmWC

6) Where should zero and span should be in relation to vessel bottom or in relation to the tapping points on the vessel? Give dimensions in relation to vessel bottom is always the easiest to prevent confusion so find the vessel inside diameter first and work from there.

7) What is the SG of the product inside the vessel.

8) Also let us know if it is possible to modify the current installation or is the current transmitter already installed and cannot be changed. The reason is that we might disagree with your installation and propose a different setup or installation so we need to know if it is worth the effort to propose something else or not.
For instance do you have various types and size transmitters in stock? Do you have a data sheet where a particular transmitter have been specified and must be used or can you use anything you want?

 

Hi there,

Only if I'd seen this earlier the people here would have opined on my method. I've had success with level measurement in reboilers of distillation columns using this technique.

OOps... An image I wanted to include didn't upload. will anyone tell me how to attach a word file.
Regards best wishes

MOUNTING ARRANGEMENT OF LEVEL TRANSMITTER
(MeOH Columns Ankleswar)

Fill Level is the same as Max Level

For Configuration of the DP transmitter, the following steps are done. Note that in the above arrangement, the meter is mounted at 3000 mm below the lower level of the vessel!

1. Un-install the instrument, take it to lab with a known pressure source & comparison meter of higher accuracy.

2. Connect pressure source to the HIG Pressure side of the DP and increase pressure till the calibration pressure is 2000mm(wc) (Corresponding to calibration certificate.)

3. Press Zero & Span buttons on instrument at the same time for 10 secs. Instrument is now in Re-ranging mode.

4. Press SPAN button only for 5 secs. This makes the instrument give 20mA whenever the pressure is 2000mm(wc) greater than the Low-side pressure.

5. With a known pressure ie @ 3500mm (wc) connected to the high side, apply an increasing pressure from second pressure source. The instrument will give 20mA reading. The proof that this is working is like this....whenever this pressure is below 1500mm wc. When this pressure is increased higher than 1500mmwc, the 20mA reading decreases from 20mA to 4mA (when the low side is also at the same pressure).

6. In the PLC program, write a rung to calculate the actual level.
Actual_level = Max_level - DP value.

Finally, If there is a level gauge, to which you have to Match readings from the DP, Matching is done by means of observing the level displayed and the value of the differential pressure measured by the transmitter, adding the 2 values and assigning this addition to "Max-Level" . For example, suppose the DP reading is 250mm and the Level gauge shows a reading of 970 mm , the sum of the two = 1220mm! In the PLC program, the value of Max_level must be constant = 1220mm.

The functioning must be verified and minor adjustments may be required. If the level inside the vessel is fluctuating rapidly, PLC logic must have an averaging algorithm.

Raj S. Iyer

 

I feel I must caution anyone to try doing a vacuum level like this since this man is way off the mark and clearly needs some more training in basic level concepts.

Sorry my friend I am not trying to insult you but you should not post something like this as a international standard way of doing something since it's not.

Don't feel to bad since you are obviously not the only one that found DP level applications a bit confusing at times, just look at some of the previous replies in this thread and you will see what I mean.

 

good day to all.

can i ask for an advise or any probable solution on my problem?

i have a problem on our lvl transmitter. the set-up is 2v3. the 3 LT are indicating fine and close reading to each other. the problem is that there is somewhat like an time based spike on 1 of the LT for about 1.5H.

there is a fluctuation in the reading of 1 LT. after about few seconds, it returns back to its normal reading.

what could be the possible problems and solutions?

Thanks,

 

MarkieM, did you found the solution for your problem?

We are having a similar issue here at our plant, we are planning on changing the DP cell but we aren't sure if this will fix the problem...

Thank you

>can i ask for an advise or any probable solution on my
>problem?
>
>i have a problem on our lvl transmitter. the set-up is 2v3.
>the 3 LT are indicating fine and close reading to each
>other. the problem is that there is somewhat like an time
>based spike on 1 of the LT for about 1.5H.
>
>there is a fluctuation in the reading of 1 LT. after about
>few seconds, it returns back to its normal reading.
>
>what could be the possible problems and solutions?

 

This 2009/2010 thread is statistically statistically, maybe 1 in 200(?), because the original poster.

- maintained a dialogue throughout
- answered questions asked of him
- posted the resolution of the problem (on the 30th post)

Most control.com threads that are not turbine generator threads have neither dialogue nor resolution and I bring up the point because getting information to assess the situation can take an extraordinary number of exchanges.

Are MarkieM and/or chucksomerock willing to have a dialogue? Let's find out.

The original problem in the thread was a level that drifted lower than the true level indicated by a sight gauge over time.

That is not the same problem symptom as a prolonged 'spike' in the level signal.

Hence, I have to ask, what is the "same" about either 'recent' application compared to the original?
Or more to point, what exactly are the 'recent' applications and the exact symptoms of whatever the problem is?

What exactly is the problem? (the more description, the better)

What is the process?
Have the level instruments ever worked properly for some period of time, or is this a situation where it's never really worked right?
Is the level instrument a DP?
What is the DP's cell High side connected to ?
What is the DP's cell Low side connected to ?
Does either leg use a capillary and remote seal?
Does either leg use a wet leg?
What is the internal vessel pressure? Which units, absolute or gauge?
What is the vessel's temp? What units? C or F?
What is the liquid medium in the vessel?
Is there a local indicator on the transmitter?
What is the transmitter output: 4-20? HART? Foundation Fieldbus? Profibus? something other?
Is 2v3 same as 2oo3 voting?
What is the transmitter's scale or range (LRV/URV)?
What is the failsafe setting, upscale or downscale?
Do any or all of these transmitters have the surge or lightning protection option?
Are the transmitter bodies earth grounded?
Do the multiple DPs use a common tapping point on the vessel for all 3 transmitters?
Do the Impulse lines have blocking/isolation valves?
Do each transmitter have a 3 or 5 valve manifold?
Are all multiple transmitters powered from the same power supply ? If not, how many power supplies are involved?
Is one side of the power supply(s) earth grounded?
What is the level signal connected to? PLC? DCS? alarm module?
Is there any other device in the loop? TriLoop, indicator, I/S barrier?

With regards to the 1.5 hour prolonged spike, what is spike's amplitude in relation to the signal?
Signal + 5% FS? signal + 50% FS? Off scale high? Providing actual numbers would help.
When a milliameter is put across the test terminals during the 1.5 hour spike, what does the mAmeter read?
What does the HART communicator indicate during the 1.5 hour spike? Same value as the spike value?
Does this transmitter have smart diagnostics which include some form of a fault log?
What does inspection of the fault log reveal?

 

Dear All

From the long discussion I read regarding difficulties of measuring level in a vacuum service vessel, typically found in power plants (condenser hotwell), we consciously avoid using D.P. transmitter. Wet leg, water ingress, meticulous calibration are the points which has lead to our decision. We have been using Displacer type level transmitter or Guide Wave radar type instruments which show now effect on variation of vacuum.

Theoretically of course D.P. type can be used with all the precautions elaborated above.

 

This has been a very useful discussion pointing out some of the traps to be aware of when doing what seems like a simple DP level.

The pressure in the tank is quite low, it's not far off boiling (50°C)

The fault has been resolved, it seems a mechanical problem with the manifold causing air to leak in, not a design fault at all.

I'm pretty certain there is no "International Standard Way" of measuring a tank under vacuum. It's problems like this that teach us all something.

Cheers
Roy

 

Use electronic DP, problem solved with legs: icing, clogging, leg level inconsistencies, leg leaks, tap leaks, etc..