3 replies to this topic

[VELU_BAPCO]

IF AN ONLINE ANALYZER (SAY FLASH POINT ANALYZER) HAS IN-BUILT PUMPS TO DRAW THE SAMPLE AND RETURN IT, IS IT NECESSARY THAT THE POINTS OF SAMPLE DRAW AND SAMPLE RETURN HAVE HIGH PRESSURE DROP BETWEEN THEM TO HAVE FAST LOOP SAMPLING?

[Zauberberg]

Velu,

Leave that to the process analyzer vendor. Why would you bother with it?
You need to specify the following:

- Type of fluid and expected (operating) range of measured quality parameters
- Process conditions (P, T)
- Sampling frequency
- Calibration issues (reference methods, repeatability etc.)

The rest should not be the matter of your concern, at least at this point of time.
Regards,

[VELU_BAPCO]

Velu,

Leave that to the process analyzer vendor. Why would you bother with it?
You need to specify the following:

- Type of fluid and expected (operating) range of measured quality parameters
- Process conditions (P, T)
- Sampling frequency
- Calibration issues (reference methods, repeatability etc.)

The rest should not be the matter of your concern, at least at this point of time.
Regards,

I'm referring to an existing online flash point analyzer which draws samples from a pump's discharge (240 PSIG) and returns the sample to the pump's suction (10 PSIG). The analyzer has in-built pumps to draw the sample and to return it.

The question is what happens if the analyzer sample take off point and sample return point is moved to a different place where the differential pressure is only 5 psi.

Does fast loop sampling calls for high delta P between sample draw and return points?

[JMW]

Any analyser loop requires that the flow through it is representative of the flow in the main process. That may require a suitable sample probe which introduces its own pressure drop.
It is also important that the flow through the analyser satisfies the analyser requirements and that the measurement is timely i.e. that the control response is a minimum.

Fast sample loops may serve more than one analyser and it is important to know if the analyser's pumps are responsible for delivering the fast loop sample flow or simply the flow through the analyser from the fast sample loop. (the analyser may be in paralel with the fast loop).

Thus the important question is if the analyser pump controls the fast loop sample flow or simply the flow through the analyser.

That the sample is taken from the pumps discharge and returned to the pump suction suggests that the analyser pump simply serves the analyser and not the fast loop.

It is important that the variation in process conditions should not cause a variation in the performance of the analyser. For example, an increase in process viscosity could mean that even though the sample loop is between the process pump inlet and outlet, that flow through the analyser loop is minimal under some conditions and the analyser's own pump simply recirculates fluid from the sample loop.

In your case, if you choose to relocate the sample take-off and return away from the current position with substantially lower pressure drop you could indeed affect the fast loop flow rate and make your measurement far less responsive or even representative.

If the fast loop is independently pumped then the take-off point can be optimised for a representative sample and the best control response.

You may find that there are very good reasons to move the take off point to a better location but it may require that you introduce a pump to control the sample flow through the fast loop.

If you have a favourable viscosity condition then a simple centrifugal pump may be all you need. If you have a significant viscosity variation you may need to choose a PD pump. But the important point is that the sample take off point should be chosen to prioritise the control response and the representativeness of the measurement and not simply to exploit any available high pressure drop in the process. The cost benefits of installing a sample pump and choosing an optimum draw off point may be significantly in your favour. For example, by reducing the control response from 3 minutes to 30 seconds you do not affect the accuracy of the measurement but you do affect your ability to control the product quality tolerance.

Velu,

Leave that to the process analyzer vendor. Why would you bother with it?
You need to specify the following:

- Type of fluid and expected (operating) range of measured quality parameters
- Process conditions (P, T)
- Sampling frequency
- Calibration issues (reference methods, repeatability etc.)

The rest should not be the matter of your concern, at least at this point of time.
Regards,

I'm referring to an existing online flash point analyzer which draws samples from a pump's discharge (240 PSIG) and returns the sample to the pump's suction (10 PSIG). The analyzer has in-built pumps to draw the sample and to return it.

The question is what happens if the analyzer sample take off point and sample return point is moved to a different place where the differential pressure is only 5 psi.

Does fast loop sampling calls for high delta P between sample draw and return points?