8 replies to this topic

[Carlo717]

I need to design both a set of centrifugal fans (inlet and outlet of spray dryer) and a pump for the feed, or at least their power requirements.

The fans needs to pump air (atm 20C) over a heat source heating to 110C and into the dryer which is also operating at atm pressure and then one on the outlet to remove the air. The pump I need will be a centrifugal pump that needs to pump a slurry up about 13m to the top of the dryer.

Can anyone please give me some advice on how to go about this as at the moment I'm completely stuck.

[Art Montemayor]

Carlo:

My comments and recommendations to you here are the same as in your thread regarding steam requirements: You simply are not communicating like an engineer.

You need to furnish ALL the basic data, background (or scope of work), a sketch and your calculations to date. This is best done in one Excel Workbook. That way, all members can easily grasp and understand what it is that you are trying to do and what you are doing wrong (or inefficiently). If you help the Forum to better understand and grasp what you have in front of you, the Forum members will be in a better position to help you.

If you do as I recommend, the ultimate winner will be YOU. The Forum members want to help - but they can't help when they haven't been give any basic information or tools. Please read the Forum Guidelines at the beginning of the Student Forum. These guidelines were put there for a purpose - to assist you in preparing a good, logical and concise query that furnishes the Forum with the basic data required to best help you.

I hope you follow my recommendations because we all want to help you. However, we need your help in this matter.

[Carlo717]

Art Montemayor:

Thank you very much for this valuable information! The information I have regarding pumps and fans is fairly limited.

The fans are situated both at the entrance and exit of the dryer to facilitate flow and prevent negative pressure buildup. The first fan will pump air from atmosphere at 20 C into a vessel also at atmospheric pressure. It is required to pump around 490,000 kg of air per hour (around 151 m³/s .. seems a huge amount! but my mass and energy balances for the spray dryer seem to work out). The second fan will then pump the same volume out at a temperature of 63 C

The pump is a feed pump and needs to pump 21,000 kg/h of a slurry containing 56% water up to a height of around 10 m and into a spray dryer. I'm assuming the pump is my starting point so the suction pressure side of the pump is 0. I have selected a velocity of around 2 m/s to prevent settling and the pipe diameter is around 15 mm

Attached is my spreadsheet containing all calculations for process streams that I have done so far. It's a little messy but important streams/units are boxed off. The energy balance is obtained via the solver section and when the air in and out temp is changed the solver iterates the required air.

Attached Files

•  Workbook1 (version 1) (version 1).xlsx   58.22KB   30 downloads

Edited by Carlo717, 11 March 2012 - 04:55 AM.

[Art Montemayor]

Carlo:

Thank you for furnishing us with a workbook where we can review your work and presentation of the problem and how you are arriving at an engineering solution.

However, for an engineering solution, you also need an engineering PFD – as I originally requested. Without it, the Forum can’t follow your calculations through the process. Of course, you can (in your head) because you know what you seem to be doing or proposing. But WE don’t.

Without a PFD, I can’t point out to you the very obvious mistake you are proposing to do: install centrifugal blowers in series – one before the dryer, and another after the dryer. This is a configuration that is waiting to create havoc. It is extremely difficult to operationally control the operation of two centrifugal fans configured in the manner you propose. If you require a driving force to carry the gaseous stream through the spray dryer, then you either resort to FORCED FLOWor INDUCED FLOW. But you only use ONE centrifugal blower – either at the dryer inlet (forced flow) or at the dryer outlet (induced flow). Depending on the expected, design pressure drop through the spray dryer, you may find that the two types of flow required different blower wheel designs. You should do research on centrifugal blower characteristics and the means / manner of controlling the flow rate through the system.

I used to have a wide and extensive amount of information on centrifugal fan design and operation, but I can’t find it in my library at this time. My library, unfortunately, has multiplied inspite of electronic files. However, there is extensive information on the Internet that you should be researching and studying if you seriously intend to incorporate a centrifugal fan in your process. There are basically two designs of centrifugal fan: the forward curved blade and the backward curved blade. If you have a large gaseous flow rate, then you will probably be forced into a centrifugal fan design instead of a positive displacement fan design – such as a “Roots” type of blower. However, you will have to keep the overall pressure drop (or required fan head) in mind.

For research on Centrifugal Fans on the Internet, for example, you can easily go to:

http://www.cincinnat...ans-blowers.htm
http://en.wikipedia....Centrifugal_fan
http://www.flaktwood...7f-0319f77799ad
http://www.ijaet.org...right-IJAET.pdf

As regards your pump requirements, also be mindful that a slurry pump is one of a special design and should also be researched and studied through the literature and Internet. Your 2 mt/sec superficial velocity for the slurry should be the MINIMUM design velocity. I usually use 4 mt/sec as the maximum slurry velocity and stay within these two boundaries to avoid settling.

You state your are pumping 21,000 kg/h of a slurry containing 56% water, but you fail to tell us WHAT % you are writing about. This is a typical student error. I again advise you to be specific and state if it is volumetric, mass, or molar %.

I hope you continue to receive further comments and advice from our other members on the Forums and that you heed their advice in good faith. Your spread sheet reveals three things immediately:

• The calculations look to be disorganized. There is no explanation of the logic and steps used to arrive at the defined critical answers. There is no mention of the procedure used to generate the results - or use of SOLVER.
• The tabs on the different sheets are not identified. This shows poor and shabby organization of data and calculations. The calculations are meant to be the defined steps used to arrive at a logical, engineering answer. Yet, no importance is given to expressing or pasting numbers that are ridiculously over-emphasized with significant numbers. Your calculations will (or should be) within approximately 15-20% of accuracy, in my opinion. Why do you put down data with eight (8) significant decimal points? This shows dis-interest, or worse, ignorance of what is being calculated. Some instructors would easily throw out such basic data as “not practical or applicable”.
• You failed to include a simple, detailed Process Flow Diagram. Again, as I inferred in my previous post: the quality of the recommendations and responses is directly proportional to the quality of the submitted information.

• Good Luck.

Attached Files

•  Srpay Dryer Workbook Rev2.xlsx   41.43KB   18 downloads

[Carlo717]

Art,

Thank you very much again for correcting my mistakes and helping me learn.

Here is a pfd drawn up using dia of my intended process. http://imageshack.us.../441/latex.png/

Sorry that my spreadsheet is such a mess, I have been learning whilst calculating so I haven't really calculated things in a logical order but instead as I have learnt them.

My reason for adding a fan before and after the system was mainly because the masters k book said to, but because of your reply I will add one before the air heater.

My % is 56% wt of water sorry for that error, I have included it within the pfd.

I didn't mean to give numbers to such a high level of accuracy, I just left it as excel calculated to save time. I would usually only report numbers to the save level of accuracy as my least accurate input.

I hope you can see that I am taking your advice on board and learning from it. My deadline is this Friday and I need to cost up my system and optimise it and then write a 40page report on it so forgive me for being so rushed.

If possible could you please also take a look at my new post about the dryer regarding a mistake I may of made within the energy balance?

Thank you very much for all of your help!

Also I have just read that you once worked with Dupont. Well this project we are doing is for a process manufacturing a TiO2 pigment similar to Duponts using the thermochloride process. We have decided to base outr plant in South Africa and first convert Ilmenite into the rutile feed we require for the Chloride process by using the Austpac ERMS process. My section of the process the is spray dryer and related ancillaries which comes just after the leeching process.

Edited by Carlo717, 11 March 2012 - 03:05 PM.

[Art Montemayor]

Carlo:

Yes, I worked a lot of Dupont Projects during my tenure with them.

I've looked at your submitted sketch and I conclude that you have either not been taught the necessities of accurately and concisely drawing out a PFD, or you have not been practicing this acquired skill with diligence. Your depiction is not of an understandable quality and, frankly, fails the test.

Look at my attached, quick, down & dirty revision 3 PFD of what you should be depicting and you can see the main areas of difference between both.

I am glad you are taking all these comments in stride and with the proper attitude that all this effort and time is directly aimed at your improvement and your key learnings. Hang in there, pay careful attention to the details of communicating correctly and accurately and never stop applying your common sense to an engineering solution.

Good Luck.

Attached Files

•  Srpay Dryer Workbook Rev3.xls   123.5KB   30 downloads

[Carlo717]

Art:

I'm not sure what was wrong with the PFD I drew. We have always been taught to include within the PFD as much information about streams as possible. Although saying that yours is a lot less cluttered and easier to understand. Maybe I should label streams with numbers and include detailed stream information below the PFD within a stream table?

I wish I had discovered this forum earlier in my degree as now I am in my final year with only 3 months to go and feel that I still have so much learning left to do.

With regards to my energy mass and energy balance:

I have balanced the moisture in with the moisture out and obtained the air required and vapour needed to be removed.

When doing my energy balance I have either my product temp or heat losses that can be varied to make it balance whilst keeping the outlet humidity constant. As shown in the included spreadsheet on the final tab... if I have no losses then my product temp ranges from very high to very low depending on air inlet temp ... and if I keep the product temp constant then I have a loss which is fairly low (only 4% on input) and then it drops to requiring extra heat input.

I am unsure why these variations are taking place. I can only assume that if no heat loss takes place then my product acts as a heat sink and has to raise temp but how can it end up at 600C?... and can it really cool down in a dryer? Surely only to the wet bulb temperature if so?

Thanks again and I hope the spread sheet gives more detail.

Attached Files

•  Spray Dryer.xlsx   68.98KB   15 downloads

[Art Montemayor]

Carlo:

I am responding to your last post by quoting you and responding in RED:

“I'm not sure what was wrong with the PFD I drew. We have always been taught to include within the PFD as much information about streams as possible. Although saying that yours is a lot less cluttered and easier to understand. Maybe I should label streams with numbers and include detailed stream information below the PFD within a stream table?”
Look at your PFD version and then look at mine. Don’t pay heed to the “pretty pictures” give importance to how the process is conveyed in a logical manner. For example, you show that the low pressure, hot air is mixing with the slurry feed in-line, prior to being subjected to the Dryer’s sprayer heads. This is NOT TRUE. The slurry is pumped at a relatively high pressure to convert the hydraulic pressure into a fine, expanded spray that can be efficiently mixed with the low-pressure, hot air that is introduced into the bottom section of the dryer in order to achieve counter-current mixing of hot air and cool, fine-particle, slurry fluid that is falling by gravity. The dryer works on the principle of achieving simultaneous heat and mass transfer by carrying it out in the most efficient manner – which is counter current.

Now look at the way the dryer produces the dried product and how it exhausts the humid air that carries off the undesirable water moisture. Your sketch show that the hot, humid air exits together with the dried solid product. THAT IS NOT TRUE. If it were, there would be no drying taking place. You couldn’t use the basis of there being an equilibrium at the dryer’s outlet.
Now look at my sketch. Does it accurately convey the fact that the dryer carries out a phase change and separation in which the solids exit with less water content and that the undesirable water in the feed is carried out by the hot humid air that exits at the top of the dryer and into a bag filter where any entrained fine solid particles are removed and returned? If so, you can then put your calculation basis on my sketch. But your sketch does not complement nor backs up your calculations.

I hope I have succeeded in explaining why a sketch must accurately complement a related calculation.

[Carlo717]

Ah yes, that makes a lot more sense. The reason for me using co-current flow is that I am using an assumption from the masters book that says the residence time of the particle is around that of air. Ie the air enters the top with the particle and carries it down through the dryer. I would of liked to use counter -current flow as I know it's more efficient and seeing as my product isn't heat sensitive it would be very useful But now as I have so much to do I don't have time to change assumptions and use more detailed calculations to get a true residence time.

Are you able to shed any insight on my energy balance predicament?

Thank you again for your wise words.