10 replies to this topic

[J.Kim]

I have a problem with designing lime slaker, which is about ventilation.
Mixing both lime and water to make lime slurry generates dust and vaporized fume with lime suspension.
Unfortunately, current system can't eject the fume properly. (See attachment 1. Current system.pdf, 2. system condition.pdf)
In spite of suggested alternative designs by partner as attached Alternative system (Water-jet scrubber).pdf, we can't apply it because both they designed existing system which has the problem described above and we are not able to verify engineering calculation about new solution.
Therefore I need to verify alternative system with calculating with your help. (See attachment 3. Alternative system (Water-jet scrubber).pdf)
Please advise me, referring to system condition below.

<System condition>

• Installed ventilation system
  
  • Refer to Attachment 1. Current system.pdf
  • Fan : 250mmAq, 30 m3/min (3.7 kW motor)
• System condition (Refer to Attachment 2. System condition)
  
  • Tank
    
  • Liquid : Lime slurry
  • Temp : almost 80 ℃
  • Retention time : 0.5 hr (30min)
  • Lime slurry volume : 3 m3
  • Tank diameter : 2.1 m
  • Tank height : 2.4 m
• Inlet
  
  • Lime : 1,873 kg/hr (2.1 m3/hr)
  • Water for slaking : 7.6 m3/hr (almost 4~5 BarG)
• Outlet
  
  • Limeslurry : 8.2 m3/hr (9,446 kg/hr)

<Attachment>

• Current system.pdf
• System condition.pdf
• Alternative system (Water-jet scrubber).pdf

Could you please review it and tell me your opinion? It would be also very appreciated it you can suggest with better idea or system.

note : engin.j.kim@gmail.com

Attached Files

•  Current system.pdf   137.75KB   115 downloads
•  System condition.pdf   53.88KB   87 downloads
•  Alternative system (Water-jet scrubber).pdf   55.02KB   110 downloads

Edited by J.Kim, 26 April 2010 - 11:56 PM.

[Claudia K.]

Hi,

I can't help you with calculation because we lack of some data: the water- fume contact time needed to remove fume particles, particles size, the ratio water/ fume+air flow. These can be found only experientally. So, with the alternative system, your partner proposed to prolong the time of contact between water and fume. This time increases proportionally with the length of the pipe common for both fluxes. You can expect an improvement in the quality of fume outlet. Some of engineering problems aren't solved by calculation.

If you can make an experiment, find a way to collect fume from the current system in a hose going to a glass absober with a height equal to that of the alternative system (the portion common for water and fume) and perform water absorption, at different water/air+fume ratios.

But, as I said, some of engineering problem can't be solved by calculations.
Good luck,

Claudia K

[kkala]

I have a problem with designing lime slaker, which is about ventilation. Mixing both lime and water to make lime slurry generates dust and vaporized fume with lime suspension. Unfortunately, current system can't eject the fume properly...In spite of suggested alternative designs by partner as attached,... we can't apply it because both they designed existing system which has the problem described above and we are not able to verify engineering calculation about new solution. Therefore I need to verify alternative system with calculating with your help.

"Current system can't eject the fume properly" means fumes coming out of the tank (1), or inefficient fume reduction in the following water spray system (2)? In case (1) more vacuum should be applied on liquid surface in the tank, that is fan flow rate should increase. In case (2) an efficient system to reduce fumes can be recommended by specialized suppliers of gas treating facilities. These may require detail data on gas flow rates and particles (as Claudia K mentions), or recommend a system based on their specific experience on lime slaking tanks, such as yours.
Moreover some comments for the alternative system can be as below.
1. Minimize water inlet with lime, make it zero if possible, so that you can maximize the fresh water passing from the water spray system and ending into the tank.
2. Install water sprays to every area where plugging is expected.
3. Fan (not shown) to be induced draft at "Fume outlet".
4. Inspiration conduit should be wide enough at its connection to the tank roof to result in low gas velocity (about five m/s considered for agitated phosphoric acid reactors).
5. Water sprays (multiple, not one) ΔP is critical, seeing that it may be inefficient (low ΔP) or cause water droplet entrainment with gas (high ΔP). Better to place a hand control valve upstream them, if optimum ΔP is not known.
6. Critical is also the particle size, which cannot be predicted by calculation (to my knowledge). Once (1981) ammonium phosphate fertilizer reactor produced aerosols (supposed of NH4Cl) that needed special treatment (Electrostatic separation, or high ΔP venturi scrubber, etc). If no data can be obtained on particle size, try to find a system already applied to similar installations elsewhere, to minimize risk as much as possible.

Edited by kkala, 02 May 2010 - 09:40 AM.

[J.Kim]

Thank you. Claudia and Kkala.
And I have more questions.
I have read perry's book, jet world, etc.. But they do not clearly explain (may be reason is that my english skill is not good to understand those book) .

My question is :

1) If system needs to suck 1000 kg/hr DAE, does it need to have 1000 kg/hr DAE motive fluid?
2) If ejectors motive fluid is water like venturi scrubber at above system, does it need to have 1 m3/hr motive water (because of density or S.G.)?
3) It is easy to get motive nozzle diameter by critical flow ratio in compressible fluid equation (At dry air, ratio is under 0.53 ). Therefore does ejector venturis throat have diameter which is calculated by critical flow ratio (At dry air, ratio is over 0.53) ?
4) At mixing tank which makes fume, how can I have fumes flow rate? I have calculate it by both liquids reaction heat by temperature and water with reactant flow rate to tank. is it correct solution?

[siretb]

As for calculations, forget about it.You will never know the particule sizes, and so and so.

Why not use a different design ?
During slaking, an Id fan is used to suck the fumes, and keep the tank under slight vacuum.
All the water (including the one needed for extinction) is sprayed (between the tank and the ID fan) into the (vertical) exhaust pipe. it scrubbs the fumes and slakes. It works. No plugging.

[J.Kim]

As for calculations, forget about it.You will never know the particule sizes, and so and so.

Why not use a different design ?
During slaking, an Id fan is used to suck the fumes, and keep the tank under slight vacuum.
All the water (including the one needed for extinction) is sprayed (between the tank and the ID fan) into the (vertical) exhaust pipe. it scrubbs the fumes and slakes. It works. No plugging.

Thanks for your precious answer.

Doesn't it have no plugging??
Some body said that ID fan system has plugging and wear problem is the reason for selecting ejector or scrubber.
And if use alternative system (venturi scrubber), it doesn't need motor as electric equipment and just use injection water for lime slaking.

okay. I'll try to use ID fan system next project.
Forgeting my first question about lime slaking, how about second questions for ejector and venturi scrubber?

Thank you.

Edited by J.Kim, 17 June 2010 - 09:37 PM.

[Qalander (Chem)]

Dear,

I am really envisaging, as if the lime, moisture containing material is being sucked though the Venturi Jet Ejectors& condensers in series vacuum creating system

This may also get clogged/plugged over a certain period of time depending upon particulate(s) content and cling-on tendency of handled material inside venturi's neck (internal orifice area).

Just Check/review if this is not foreseeable for your system.

[J.Kim]

Thank you Qalander.

In case of Venturi ejector using air as motive fluid, possibly make plugging problem in the neck of venturi.
But Venturi scrubber use water as motive fluid.
Therefore, I think, In case of Venturi scrubber, water scrubbing fume with lime suspended power should be clean the neck which plugging predicted.

And..
Could you please answer my Question described in #4 ??

[kkala]

Below is a view to above queries.

1) Question is not clear to me, probably it concerns kg of fume (to be removed from gases) per kg of water injected. I think the ratio should not be necessarily 1. Water is assumed to be recirculating to washing system through a tank, part of it rejected and fresh water added (rejected water from tank conical bottom to avoid plugging).
2) I observe venturi scrubbers for HF in Kohl & Riesenfeld' s "Gas Purification" (Gulf, 1974) and see mass ratio of liquid to gas rate = 0.6 - 0.93 (water as absorbent). It should not be necessarily 1, depending on design.
3) No, venturi scrubbers do not approach critical flow (even though gas flow at neck is high). Gas ΔP is their critical parameter.
4) Your calculation makes a first basis. Check whether air is also required for surface cooling, in addition to the air that will pass from slaker leakages at any case. The latter is necessary for creating sub atmospheric pressure, but it can be estimated by experience (in my opinion), not theoretically.

In brief, you can entrust the design and supply of lime fume system to a company of proven experience, or visit some similar installations and try to "imitate" the working system after having understood it. In both cases type of gas cleaning equipment will be specified with some confidence. Best equipment for use is not yet known (e.g. venturi scrubber may not be the proper solution), a spray tower could probably be sufficient.

Note: Concerning the fan, yes it may need frequent cleaning. But it may not be a problem, especially if lime slaking is batch. Even if it is continuous, you may be able to stop operation for e.g. half an hour to clean it.

[J.Kim]

Thank you Kkala.
And I had test for Ejector and Venturi scrubber.
As you said, Suction pressure and flowrate are very different than my expectation (calculation).
But result of suction pressure and flowrate was sufficiently fine except lime plugging problem because our pilot plant has no lime preparation system.

Question :
How frequently maintenance ID fan? Is it fine to use ID fan at batch lime slaking system which will operate 1 week continuously?

Thank you.

[kkala]

How frequently maintenance ID fan? Is it fine to use ID fan at batch lime slaking system which will operate 1 week continuously?

In my opinion fan maintenance can take place at much longer intervals of time (than a week), if some precautions are taken concerning the layout of the fan and its ducts.
I think best fan location can be upwards lime slaker, but not upright. Piping had better have an inclined part of sufficient length for the entrained solids to precipitate there and washed out to slaker through sprays (during operation) or hoses (when unit stops). Next pipe part to fan can be vertical.
It may not be practical to maintain 5 m/s gas velocity at ducts to fan, but try to avoid excessive velocities favoring solid entrainment.
Most common symptom would be fan casing vibration due to impeller unbalance. If due to deposits, cleaning the impeller would cure the case. If due to corrosion (probably more rear), the impeller should go to maintenance shop for filling and balancing.
Avoid impeller covered with rubber, but fan casing can be covered so. I suppose ducts will be covered with rubber internally, this may be necessary for the part hit by "jet fumes" to avoid erosion.
Above is extrapolation from a phosphoric acid plant, where "eruptions" from agitated gypsum pulp do not occur, but phosphate entrainment does. There were two reactor vessels, the fan of one of them was on the ground and had more maintenance problems. The other fan was located about 10 m higher than reactor ceiling (as described above). Every day it was usual to stop the plant for 15 min to clean the ducts.
Above concerns induced draft fan. Forced draft fan is not judged suitable, since gases with fumes will pass to ambient air from any duct or slaker leakage.

Edited by kkala, 27 July 2010 - 12:52 AM.