Cyclone Separator Design Question 1

[JD P.E.]

I'd like to evaluate a cyclone separator with a short cone (compared to Perry's 2xDia length recommendation). Does anyone have any guidance on how this performs compared to a longer cone?

 

[davefitz]

There was a great deal of technical papers describing the performance of such separators used in CFB circulating fluidized bed boilers in the time period 1990-2000. Review the ASME published tech papers from the power conferences of that period. Another reference that started it all is "FluidiZation and fluid particle systems" by Frederick Zenz published around 1960 for the oiland gas industry. And of course the handbook for chemical engineering.

"...when logic, and proportion, have fallen, sloppy dead..." Grace Slick

 

[Compositepro]

A cyclone will work with no cone, but then it will create its own cone by depositing powder at its bottom in a cone shape. The purpose of the cone is to transfer separated powder to center exit with as little re-entrainment as possible back into the spiral air flow. A straight walled cyclone can be placed on top of a collection drum, and the spiral flow will continue into the drum.

 

[georgeverghese]

Where in Perry do you find this guideline with this conical section, and which edition is it in ? In the 7th edition, page 26-32, fig 26-18, there is no cone for the liquid section.
Is this cyclonic separator in gas-liquid service or gss - solid service ?
What are your reasons for selecting this type ? Highly fouling service ? High pressure drop acceptable ?

In my previous OpCo design guidelines, there is no cone at the bottom either. Instead, an internal slightly conical shield plate prevents the settled liquid from getting re entrained into the swirling gas stream.

 

[JD P.E.]

In the gas-solid section. I'm not selecting this particular design, I'm evaluating an existing piece of equipment.

 

[georgeverghese]

Cone length influence is discussed on page 17-30 and the the table 17-4.
Cone angle coincident with cone length being 2xDc appears to be 77deg, which is a little better than the min 60deg suggested on page 17-30.
My interpretation of table 17-4 is that the "required cyclone length" is the barrel length plus some part of the conical section.