4 replies to this topic

[Jiten_process]

 sketch.xls   121KB   88 downloads

Dear all,

need your valued comments.

We are dealing with the close loop fluidized bed drying system. This fluidized bed consists of two different section which needs N2 gas as a drying media with different temperature (as shown in sketch). I have attached excel sheet where i have drawn the rough sketch of the system (omitted all instruments and some lines). My question is as per sketch can i provide single centrifugal fan only for both the different section (delta P may be different for both the section), or i have to provide two seperate centrifugal fans. Will the system operation be stabled with single centrifugal fan?

Plz clarify and let me know if further data is required.

Thanks a lot in advance

[Art Montemayor]

Jiten:

I don't know what you mean by "dealing with". I am going to assume you are at the the design or feasibility stage of design and that you are contemplating saving some capital money by investing in only one centrifugal recirculation fan to operate two different fluidized bed dryers. Am I correct?

If so, then you are at the pivotal point where you have to commit to making a decision on the equipment to specifiy for purchase. Your sketch is very educational and instructive because it reveals nearly all we need to know in order to make a reasonable comment on the type and methods of providing the continuous nitrogen medium to both fluidized beds. One single centrifugal fan used for both fluidized bed requirements in this case is a possibility - but it will require very detailed and accurate calculations as to the total head required for the fan as well as some relatively sophisticated instrumentation and surge controls. My personal preference would be to invest in two, separate and specific fans - each dedicated to the specific fluidized bed it services.

My reason for recommending two specific fans for this type of application is that sooner or later, there will be inherent process variations required - either as turndowns or turnups in the capacity of each one of the two fluidized beds. This possible process change could create havoc in the controls of either (or both) of the beds. I don't know the quality or the description of your solid material being dried in the beds, but I would presume that any excess carry-over to the cyclone separator would be a major process excursion that would create many problems. I am presumming that you must have precise and consistent process control and quality of the solid product. If my presumptions are correct, then the need for individually dedicated and controlled fans is a pre-requisite for ensuring good control and consisitent production quality.

I note that the mass flow rate of nitogen to one dryer is twice that of the other. This requires a substantial flow split and causes a totally different static head requirement for each dryer - linked to the ability to split the flow of the main stream. The fan's characteristic performance curve will be heavily taxed to avoid going into a surge condition while maintaining the required total flow for both dryers - and this is at steady state, design conditions. The moment you need to vary the flows to either dryer, the balance of the flows as well as the static heads will change - and affect the performance of the fan.

[Jiten_process]

"I am going to assume you are at the the design or feasibility stage of design and that you are contemplating saving some capital money by investing in only one centrifugal recirculation fan to operate two different fluidized bed dryers. Am I correct?"


you are absolutely correct. We are still at physibility stage and that is why we were evaluating the possibility.

"but it will require very detailed and accurate calculations as to the total head required for the fan as well as some relatively sophisticated instrumentation and surge controls"
Will you please clarify what you exactly referering to surge control with reference to centrifugal fan. we have thought of that we will provide pressure transmittor at the outlet of each steam air heater on the lines of both the fluidised bed section and this will be looped with pressure control valve provided at the inlet of each steam air heater for both the section. But frankly speaking, i am still in dilema.

further you are absolutely correct, during operation we definately have to turndown or turnup as you said and being a centrifugal fan it will follow it's characteristic curve and other fluidised bed sec. may not get suff. pressure. But if i provide two diff. fan then suction side i have to provide a common header only because as per spcs. we have one only one bag filter, one dehumidifier. What is your comment on it? should i split dehumidifier and bag filter too?

Please revert

Again thanks a lot for your valuable guidance, it will really help me a lot in taking decision.

[Art Montemayor]

Jiten:

For a quick and important tutorial on centrifugal fans and their characteristic behavior, please go to: http://www.titan-air...tips/6_2003.pdf

and study well what is being written about the two major types of centrifugal fans - the backward inclined blade and the forward inclined blade. It is very important that you master the characteristics and the reasons the fans behave the way they do.

Your system of recirculated nitrogen is defined by the System Curve. I presume you have already worked on this (at least preliminarily) and will develop a serious and firm System Curve. You must know - as accurately as possible - what will be the expected developed static pressure by the fan(s). When you use one fan and depend on splitting the product discharge flow, you are trying to control a very complex and difficult maneuver.

If you employ a typical, forward blade fan you will undoubtedly get a "dip" in your performance curve to the left of your peak pressure point. You will note that as you progress to the left of the peak pressure point, you reach a section where the corresponding fan capacity related to the static pressure is actually TWO CAPACITIES! This is due to the curve's "dip". The fan can't realistically produce two capacities at the same time in this region - or any region! But the fan doesn't know that fact because it is a stupid machine. It tries to follow its performance curve and as it does, it will go through a very unstable operation - similar to a "surge" operation in a centrifugal compressor. This kind of operation is damaging to any fan. It pulsates and causes vibrations and mechanical damage. You cannot allow a fan to try to operate in this region. It will eventually auto-destroy itself.

Backward inclined blades exhibit a superior performance curve - and they are, as you would expect, more expensive. This is a natural "trade-off" to be expected. However, they also have a tendency to "stall" - although to a lesser degree. Nevertheless, the same rules apply to them as to the forward inclined blade type: you must avoid the region left of the preak pressure point.

Perhaps by now you are beginning to understand why I have been harping on the accuracy of the system calculations and the importance of having top instrumentation installed in such a system. I would certainly recommend you split your systems into "trains", each one having its own fan, fluidized bed, filter, and dehumidifier. That is the best and most certain way to have total, absolute control of the process during any excursion or change in capacity.

I hope I have succeeded in making my point and leading you on to a better understanding of what it is that you are up against. I think you are also aware of how important it was to sketch out your system and spend the time in communicating the real and actual process you are studying. Your efforts paid off in telling us exactly what we needed to know in order to alert you to what we know to be the important points to design around in such systems.

[Jiten_process]

Dear art,

Thanks a lot, that was really informative. You have really helped me in taking a confident decision.

Thanks once again.