72 posts

### seperation efficiency of cyclones

Dear all:

pls relay i want know about cyclones seperation efficency? and cyclones eeficiency how we can calculate it?

**Know the answer to this question? Join the community and register for a free guest account to post a reply.**

79 posts

72 posts

138 posts

### Re: seperation efficiency of cyclones

elwathig,

Formulas for cyclone design are available in many papers and books.

A list has been provided by a poster on a famous engineering forum (1).

You will even see a reference to an excel spreadsheet (2).

Google also returned me a nice page where calculations can be done (3).

I also found a short text which explains clearly a few design formulas (4).

On the same site, the compagny sell an excel sheet for these formulas (5).

If you read german you will find a lot of data in the MVT handbook (6).

The wiki article explains the principles but does not relate the formulas to the cyclone geometry (7).

(1) http://www.eng-tips.com/viewthread.cfm?qid=141255

(2) http://pas.mnsi.net/brochure.htm

(3)http://www.ajdesigner.com/phpcyclone/cyclone_equation_cut_diameter_particle_density.php

(4) http://www.ezzesoft.com/files/CYCLONESver2.pdf

(5) http://www.ezzesoft.com/ezzecyclone.html

(6) http://www.zogg-engineering.ch/MVTTVT/MVT_Handbuch.pdf

(7) http://en.wikipedia.org/wiki/Cyclonic_separation

I tested many cyclone design formulas.

My preference goes to one of the simplest: the popular Dietz formula.

This formula gives the efficiency f(x) for a given (spherical) particle diameter x.

All formulas for f(x) share the same structure:

- they involve a S-shaped function

- they involve a cut-diameter

- the cut diameter depends on the flow

The Dietz formula looks like f(x) = 1 - exp[x²/x0²] .

The Ioza and Leith formula looks like f(x) = 1/(1 + x^b / x0^b) .

The diameter of the particles is denoted by x.

The cut diameter x0 and the exponent b depends on the geometry.

To calculate the actual efficiency of the cyclone, you need to combine the cyclone efficiency curve f(x) with your particle size distribution (PSD) and sum-up what is collected.

There is of course an even simpler model: assuming a perfect cut.

In this case:

f(x) = 0 for x>x0

f(x) = 1 for x<x0

and therefore the efficiency will then be determined by the % of particles smaller than the cut size, which is simply one point of the PSD of your product.

The more sophisticated formulas differ by the slope of f(x) near the cut diameter and by other effects that may have been studied by their author.

I compared many of these formulas and found a large scatter.

The picture below shows the comparison of a few models for the same geometry.

Different formulas do not even agree on the cut diameter, altough the tendencies are always similar.

This is obviously not exact science and one should not rely too heavily on such formulas.

Data sheets from suppliers might be the most precise information available.

Have fun!

Michel