Dear Friends,

We have two Tanksfor Acetic Anhydride storage of 100 Ton capacity(Each)

It has Tank weighing system. With 04 load Cells at foour corner of it's platform. The load cells are double ended type. The problem is Tank weight varies with the time It goes on increasing during night time & starts decreasing during day time however the reading of any 24 Hr. span are same. What may be the problem.

Pl. help.

 

It might possibly be the same problem as was discussed in the following thread.

http://www.control.com/thread/1242418426

 

> The problem is Tank weight varies with the time ....

My guess is that the tank is not mechanically isolated enough and that some pipework connected (and not flexibly enough) to the tank causes stresses that vary with temperature

 

Is it the same tank which was earlier installed with compression load cells, with indicator 100m away from tank? What is the capacity and resolution.?

Rgds
Sam

 

Sam .. No it isnt the same tank.

Tank capacity is 20000 litres approx (Two separate Stainless Juice storage tanks Insulated tanks side by side)Both with the same problem. Im not sure what resolution (By this , do you mean the value that the litre reading jumps each time as it increases/Decreases. ?

A further development... Today we loosed all the pipework joing onto the tanks.

Result.......... Zero reading before Hot CIP when tank empty. After CIP we had a -330 Litres. I am not sure what happened to the reading after cooling, It was late..I had to go home

Mick - Thank you

 

My question was regading the Ac2O tank. What is the indicator capacity and resolution? that is for eg, 100000kg* 100kg resolution (display).

Readings can be affected by temperature changes (expansion, pipes etc), wind, friction etc.

Tanks weighing would give accuracy of 0.1 to 0.5% typically. (100kg to 500kg on a tank of 100t capacity).

Regards
Sam

 

Dear Sam;

The Indicator resolution is 10Kg The indication is in Kg's.

We have another third tank for the same application. & same range (but from another company) but it has very less variation (50 Kgs) It's load cell are installed at middle of the structure & are of compression type. so does installation affects the readings. Another doubt of piping stress-- I ve to check for that.

 

Hi Vinod,

If your tank capacity is more than 30000 kg (30t), it will be difficult to get 10kg stable reading on a 24 hour basis. A drift of 1/3000 is normal.

I have only one experience with a 150t AC2O tank where the reading was stable within 40-100kg in 24 hours. (But was atleast 7 years back).

Regards
Sam
sam(dot)sensor(at)gmail.com

 

So do u think by incresing the least count from 10 to 30 will solve the problem?

Also what is your openion abt the installation? does shifting the load cell at the centre of the supporting beam will help to reduce the problem?(As we have for another 2 tanks?)

Pl. reply.

 

You may have a number of compounding issues possible.

First, are the tanks outside where they catch direct sunlight?

Second, same question for connecting piping.

Third, if inside are there huge ambient temperature swings in the area the tanks are in. A good idea would be to mount a temperature sensor near the tank and see if the weight change trend tracks the ambient temperature. More than likely, they do, but the additional info will help in talking to the tech support people from the mfg.

Fourth, check the temperatures of the load cells themselves during normal operation and after cleaning. If not thermally isolated you are heating the load cells and the mounts. Not good for many reasons, least of which is an error in measurement.

Fifth- have you checked each individual load cell output? You need to do this empty, and full, before and after the hot CIP. This will let you know if you have uneven loading on the cells. Uneven loading makes the whole system more susceptible to temperature changes as there is different stress on each cell and the change in stress will vary from cell to cell with a change in temperature.

This will also let you know if you have one bad cell causing all the problems. One cell in the process of failing can cause the issues you're having.

Sam W (different guy from the other Sam responding to this thread earlier)

 

Dear Sam;

Thanks for the quick reply.

1. Our tanks are under direct sun light. But as i mentioned The third tank which behaves normal is also under direct sunlight.

2. All pipnig is under direct sunlight.

3. We have taken the readings of tank weight against the temp. & we found the change in weight has some relevance with temp.

4. We had tried to cover the load cell form direct sunlight. But this does not help. The vendor taken the tests for the load cell(Voltage;Resistance) Then after he changed the Load Cell & Indicator & Cable. But the problem is as it is.

5. My another observation is when the tank is less filled the error is less & if more filled ;More the error.

6. Another thing i Have noted that when we open the JB (containig the Load cell connections to the indicator) some time the indication varies drastically. It fluctuates at very High rate. & again come down to original after some time. WE have provided the Earthing to thr JB.

7. Pl. note that I'm comparing this two tanks with another tank whose indicator is working fine.

 

Hi
Your 5th observation makes me think that the "temperature effect on rated output" of the load cell is a major contribution of error.
What is the temp effect on rated output specification of the load cell (It will be like 0.001%/C or so). Compare this specification with the load cell used on third tank which behaves normal.

Are you using a long cable from junction box to indicator. This can affect the temperature effect on rated output (if this is the case , the weight reading will decrease when the temperature increases).
How much does the temperature vary in 24 hours ? And how much does the weight reading vary in 24 hours ? Find this error for 2 different loads (say 25% and 75% filled) and calculate the error and see whether it is proportional to load. Can you please post the results?

Point number 6 can be because of a loose connection or solder joint inside the JB.

Regards
Sam S
[email protected]

 

Sam;

the Temp. Specs are 0.0025%/°C

The temp. Varies @12-15 °C in 24 Hrs.

at 25% load The Indication Variation is 100 kgs & at 60-70 % load it ia @ 300 Kgs.

We have changed the cable Also for 1st indicator the cable length is within 1 mtr & for 2nd it is within 8-10 mtrs. for 2nd indicator we had shifted the indicator to have the cable length within 1 mtr but nothing positive result.

I think this is a issue of platform installation or load cell installation or other connected piping stress.

Pl. reply

 

Hi,

Temp. Specs of 0.0025%/°C will give 40kg error ONLY on a 100t tank for 15C change. It is most likely an installation problem. Do you have some photos of the installation (1st/2nd giving error and 3rd working).

To check piping stress , dis-connect the pipe and monitor the drift in 24 hours (make tank "free" from surroundings).

Read the notes on following link, which may point to possible areas contributing error/drift:
http://www.revere.nl/Application_Notes/English/6_Vesselweighing.pdf

All the best
Sam S

 

Hi,

We are not experts , but we work on one thumbrule use compression type loadcells for any of our applications.

Once to save cost of instruments we have used shear beam type loadcells for 30T bromine tank, rather than going into details of all variations We simply replaced with Compression type.

Do you have any inert atmospher?

 

1-How many tanks are giving problems?

2-Is it the juice storage or the un-hydride tank giving more problem?

3-Most load cells are compensated for upto 40C. Are you operating outside compensation temp range?

4- What is the opinion of your supplier?

5-What kind of load cell make is used? Is it a good make?

6-Is it a 2mV/V or 3 mV/V load cell?

7-You said , one system is working ok. What load cell is used there?

8-In the system working ok, did the supplier face similar problems at its installation?

9-Is there excess bending on platform?

10-Have you checked without any pipe connexion? how was the reading?

11-Is reading stable when the tank is empty (0kg) with auto tracking OFF?

Cheers
Joma F

 

If one tanking is showing correct weight and two tanks do not, show the supplier the first one. It is either faulty load cells or incorrect installation.

Since one tank is working ok, you should be able to get the problematic twos fixed.

Patrick

 

Dear All;

The load cells installed for these two tanks are doubble ended type. & for the Other which is OK are compression type.Our supplier is continously trying to fix the problem. He has changed every possible component of the system.But still fail to fix the problem.

The OK tank does not have any problem right from it's installation.(The supplier is different) Also the basic difference in the installation is the OK tank has the load cells installed at middle of the supporting structure.(at middle of each four supporting beam)& have a kind of shocabsober type arrangement near every load cell. But the faulty tanks have load cell at each corner of the supporting structure.

We did'nt tried to machanically isolate the tank till date. it has flexible hoses for every pipe connection.

i'll post Other load cell details within short period.

 

Replace one of the two faulty weighcells with the other vendor whose product works correctly, and do the installation exactly like the one that works. That will tell you if it is a vendor and/or product related issue.

Walt Boyes
Editor in Chief
Control and Controlglobal.com
www.controlglobal.com

Mailto:wboyes [at] putman.net
Read my blog SoundOFF!! At www.controlglobal.com/soundoff

 

Hi Vinod,

Looks like , I was involved in the installation of your first system (good one) 8 years back (at my previous employment). A150000 kg tank where the customer (Vam) wanted 50kg stable resolution!

It had lot of initial hiccups which was subsequently fixed after a lot of close monitoring and fine tuning. One Tyagi was the Inst Mgr at Vam (jubilant) Gajraula, if I can recall correctly.

If the above sounds like your system, please post your email id, I will email you my phone number.

If it is not, my apologies and kindly ignore this post.

Regards
(name withheld)