Dear Mr Cavin,

Your project can be split into 2 items.

ITEM 1: The pneumatic conveying.

This item can not really be discussed right now, as we lack the necessary information s.a. particle size (distribution), material density, fluidizability, etc.

Eventually, you will have to consult an expert or supplier.

ITEM 2: Cooling the product.

Your idea is, cooling the material with the (conditioned) convey gas.

Let us assume that the convey gas volume is known.

Suppose the residence time is long enough to reach an even mixture temperature before the end of the pipe line.

Then it is rather simple to calculate that equalized temperature.



T(mix) = { Cp * Q(gas) * T(gas) + Cc * Q(mat) * T(mat) }/ {Cp * Q(gas) + Cc * Q(mat) }

or :

T(mix) = { Cp * T(gas) + Cc * mu * T(mat) }/ {Cp + mu * Cc }

in which :

Cp = specific heat-content of gas at constant pressure

Q(gas) = Mass flow of gas

T(gas) = Temperature of gas

Cc = specific heat of material

Q(mat) = Mass flow of material

T(mat) = Temperature of material

mu = loading ratio kg-material/kg gas

Example :

Cp = 0.24 10^3 cal/kg

Q(gas) = 4.54 kg/sec

T(gas) = 150 degr.C

Cc = 0.2 10^3 cal/kg

Q(mat) = 111.1 kg/sec

T(mat) = 382 degr.C



T(mix) ={ 0.24 * 4.54 * 150 + 0.2 * 111.1 * 382 } / { 0.24 * 4.54 + 0.2 * 111.1 } = 371 C

As the material to gas ratio is mostly rather high, the mixture, equalized, temperature is also mostly close to the material temperature.

Furthermore, it is necessary to calculate the cooling through the pie wall and the possible temperature rise by energy losses by friction, while the material is moving through the pipe.

May be, a water cooled conveying pipe would work.

The conveying pipe is then executed as a heat exchanger

(If you require some extra information on a pipeline, considered as a heat exchanger, let me know)

good luck

teus ■

You only have to convey horizontally. After that you are free to cool the product using gravity in any allowable format. Residence time in the gas stream is thereafter a reactor function. Available options include, but are not limited to, Cascading the powder down a spiral liner to increase the residence time, said liner being wrapped around a cool core; inside a cooled jacket: cascading the powder over fluidising trays: gliding the powder down against the upsurging nitrogen, similar to freeze drying.

You will need a cyclone to knock out the conveying gas before the fall process. A doddle. ■

In pneumatic conveying pipelines heat transfer is almost instant for fine powders. This is due to very high surface area and high Reynolds number. It is almost a perfect heat transfer system. Steady state temperature will be achieved within the first few meters then there will only be heat loss to the pipe wall, so the resident time does not matter much. Since the heat capacity of gas is very low as compared to powder and you have to cool the powder by 100 degrees I think you will need a big bore pipe to keep the SLR down. Design for pick up velocities of 20 m/s, if you need an 8” + pipe higher velocities will be needed. ■

Dear Mr Cavin,

According to your description, I tried to figure out the estimated flow ability or fluidizability.

I found 4 indexes:

TD = Tapped Density = 750

BD= Bulk Density = 550

1) Hausnerr Ratio = HR = TD/BD = 750 / 550 = 1.364

If HR < 1.4 then non cohesive material

2) Carr Index = CI = (TD-BD)/TD * 100 = (750-550)/750 * 100 = 26.7

IF 22 < CI < 35 then poor flow

3) Geldart Classification

Group A:

30 – 150 micron

particle density < 1400

then , easily fluidized, large expansion

4) angle of repose

30 degrees then free flowing

45 degrees then cohesive.

Well, eh, I think that your product is well fluidizable and under that condition pneumatically conveyable.

Next step:

Then, for particle size 100 micron and particle density 1000, the floating velocity in ambient air is approx. 1.45 m/sec.

The conveying velocity at atmospheric conditions (air, is approx 71 % nitrogen) is 5 times 1.45 = approx 7.25 m/sec.

Assume a loading ratio of 25 (rather low for this conveying distance)

Air mass = 2000 / 25 = 80 kg/hr or 0.0186 m3/sec at atmospheric conditions (end of pipeline)

Diameter of pipeline is then approx. 2.5 inch

Suppose 1 bar(o) conveying pressure (wild guess)

then v-begin = 7.25 / 2 * (150+273)/273 = 5.64 m/sec

Average velocity = approx 6.43 m/sec

Residence time = approx 30/6.43 = approx. 4.7 seconds

Keeping in mind Mr Matoo's remark of the influence of the residence ime, you now can make your fist estimate.

We are interested in the outcome.

NOTE: This is all, as we say in Holland, done with a “wet finger”

Best regards

Teus

(Checking the wind direction is done by sticking a wet finger in the air. The up wind side of the finger becomes colder, due to evaporation. This is the source of this saying) ■

Dear bvsarma,

Your remark is a valid one.

Thermal equilibrium can be split in 2 steps.

1)

Thermal equilibrium between particles and conveying air.

If the particles are small (sub mm) then the heat exchange between particles and air is probably very quick, although depending on the heat conductivity of the particle material.

For bigger particles, the equilibrium time is longer, resulting in a lower air temperature than calculated with complete heat equilibrium.

(In reality, the heat exchange curve is an e-function that reaches equilibrium asymptotic after an infinite time)

You are right that the reality is more complex than the assumptions might make you think.

2)

Thermal equilibrium between conveying air and environment through the pipe wall.

Here, the equilibrium will be reached after a while, when the heat flow is stabilized after the temperatures are stabilized. The transient phase time is not considered.

Actually, the transient period for the particles is not considered.

In the calculations, the approach of assuming instantaneous thermal equilibrium is used to detect the worst case scenario for condensation of water after pressurizing the air and mixing with a (cold) material.

Which expensive equipment are you referring to?

Have a nice day

Teus ■

Dear anupjacob,

A pneumatic conveying system is not the first choice for drying the material.

In a pneumatic conveying system, the residence times are too low.

In your case a fluidized bed or a rotating dryer is a better solution, because you can control the residence time and the air flow.

Calculating the drying process is not easy, because the surface water (if any) will evaporate quickly, but the material moisture must travel through the particle to the surface and that is a completely different story. ■