I work as a process engineer in a a vinyl chloride (VCM) plant. As a result of the cracking of one 1,2-dichloroethane molecule, a molecule of VCM and one of HCl are produced. Currently, we have a difference of 20 tons/day between the production of VCM reported by the coriolis flow meter and the production calculated estechiometrically with the production of HCl (there is no balance). Furthermore, these 20 tons/day are neither in the tanks or in the flare, so apparently, the sensor is not giving a correct measure. the sensor was changed, there was no difference. the calibration was verified and was ok. the installation was verified by a micro motion technician and was ok. The size of the line is 3", it seems that all doors are closed, i would really appreciate some help, thanks in advance

 

Reminds me of one of our customers.

They were making a product that used some fairly nasty ingredients.

They had been making this product for decades and had not done a mass balance because as long as they were making money no one cared.

The USEPA did care though. Despite the fact that they never were found to have discharged one of the nasty chemicals used in the process, the mass balance showed several tons of the compound was unaccounted for annually based on the input of raw chemicals and output of finished goods.

They spent a bunch of money trying to figure out where it was going. Everyone figured it was going out the exhaust stack or down the drain but long term monitoring showed that was not the case. Ground water monitoring failed to find it either. Still a mystery to this day. They never did figure out just where it was going, and eventually ended up ceasing production.

--
Bob

 

20 tons/day of HCL (lost) is what percentage of daily HCL production? 20% 10%? 1% 0.1% ?

One mass flow meter or more than one mass flow meter are in use?

 

A few questions:
1. How big a discrepancy is 20t/day? (At 1000 t/day thats 1%)

2. Where is the totalization happening? The micromotion transmitter, DCS, PLC, SCADA?

3. How is HCl production measured?

While no instrument is perfect, I'm surprised by the number of times I've run into coriolis meter accuracy being questioned based on some pretty unreliable assumptions. For example if the HCl production is based on NaOH used to neutralize the HCl how accurate is the assumed strength of the NaOH? I've seen instances where the "25%" NaOH varied from 18 - 30% depending on which operator made that batch of NaOH.

 

what is the pipe size, flow rate, p, t, & what fluid are you measuring the dichloroethane or the VCM?