Does a large temperature change affect the accuracy of a mass flow meter when it is in use?
Thread Content
Our company has several TECO mass flow meters used to measure propylene oxide (abbreviated as PO), and now there is an issue with inaccurate measurements. I would like to know if this is related to temperature. The current situation is as follows: The PO is pumped from the tank area to the reactor, with a mass flow meter installed next to the reactor. There are long pipelines between the tank area and the reactor, and these pipelines always remain filled with PO without being emptied. The PO temperature in the tank area is maintained at around 15°C. As it is an intermittent process, at the beginning of each feeding cycle, some PO from the pipeline is introduced into the reactor. Due to the high temperatures in recent days, and despite the insulation surrounding the pipelines, the temperature of the PO still rises to 25°C. It takes about half an hour to an hour for the temperature of the PO to gradually drop to 15–17°C and stabilize; in total, 2–4 hours of feeding are required before the process can be stopped. At the beginning of material flow, I made sure that the flow meter was filled to capacity; although there was no zero drift and the readings were stable, by marking the 0 point, measurement errors still occurred, at a rate of 4‰. Kolon promised that this error could be reduced to 1‰. May I ask, will a temperature difference of 10°C between the front and back of the material cause inaccuracies in the measurement by the mass flow meter? .
I think the main error lies in density; according to the working principle of mass flow meters, temperature does not affect them, it can only affect the density of the medium! As for whether it’s accurate or not, it depends on the manufacturing process. All watches have some degree of error; it’s just a matter of how large that error is. The level of error is related not only to the watch itself but also to the manufacturing conditions. If the error is within acceptable limits, then the watch can be considered accurate. ...Since you have already calculated the measurement error, all you need to do is check whether this error falls within the limits permitted by the manufacturing process. If it does, then you can continue using it in that way; if not, you’ll need to find another solution. It has nothing to do with the error range specified by the instrument manufacturer – what matters is your own assessment! . . . Personal opinion!
Well, it depends on the volumetric expansion coefficient of propylene oxide; the value isn’t very high, at 0.00213. The density should also change little with temperature, but how did you calculate this error? If there is a difference of 10 degrees Celsius, an error of about 4% is acceptable.
Since insulation is required, this medium should have a low boiling point and be easy to vaporize; it’s likely that at the beginning the medium existed in both liquid and gas phases. The accuracy of the instruments is specified under the condition of a full pipe with single-phase flow.
We have seven or eight units here, and many of them are able to achieve this pace under normal conditions, with high accuracy. Recently, with the hot weather, there have been many cases of large measurement errors, and the temperature is the most obvious reason for this. We really can’t take Cologne’s claims too seriously, but when it was used before, calculations and comparisons showed that an accuracy of 0.2% or even 0.1% could indeed be achieved. The error has increased recently, so we need to find out why.
The measurement principle of a mass flow meter is independent of temperature, but the following issue may arise: when the temperature changes significantly, the inner diameter of the measurement tube can change, which might have some impact. However, the manufacturers of mass flow meters should have made adjustments in the software to compensate for such temperature changes.
I’m the type of person who doesn’t like reading and isn’t interested in understanding things in depth; so I’ll just make a random guess here: when the poster says the error is 0.4%, does that mean the measured value is always lower than the actual value? If so, I wonder if it could be that the \"PO\" in the pipeline expanded thermally and shifted slightly backward toward the flow meter by 0.4%? If that’s the case, then it’s probably because of the reason mentioned by Pig Brother—that the accuracy of the flow meter isn’t sufficient. It’s just a guess, my humble opinion; please don’t get angry at Pig Brother! :lol
This is a very good question. A scenario like this is quite illustrative: 1. Temperature itself has no effect on the measurement accuracy of mass flow meters, but temperature changes do have an impact, and it’s a significant one at that； 2. Its mechanism of action is as follows: the elastic modulus of the measurement tube changes with temperature, and the calibration coefficient of the mass flow meter is a coefficient that is directly related to the elastic modulus; in other words, when the elastic modulus changes, this coefficient also changes ； But this change does not require manual intervention; the manufacturer has already incorporated this relationship into the transmitter ； 3. Therefore, if the temperature measurement is inaccurate, the corresponding elastic modulus will also be inaccurate, and as a result the corresponding flow coefficient will be incorrect as well. This inevitably leads to significant errors; the extent of these errors needs to be determined by consulting the manufacturer’s technical staff.