7 replies to this topic

[ashokg9]

Hi,
I calculated the no. of trays and location of feed tray using fenske equation for an existing distillation column. According to my calculations the feed tray should be located below 90 % of total trays. But in actual the feed tray is located below 67 % of total trays. I used the practical data for the calculations.

1) Why there is difference in actual and calculated values
2) What is the effect of shifting the feed tray location on the compositions of top and bottom products..?

Thanks,

Ashok

[S.AHMAD]

Have you considered tray efficiency? Actual tray and theoretical tray is related by tray efficiency.

[ashokg9]

does the % of trays above feed tray depend on tray efficiency..?

[S.AHMAD]

Yes... tray efficiency for stripping and enriching section may be different.... however you may calculate the overall tray efficiency = No of theoretical tray / No of actual tray x 100. This is for existing unit normally for determining overall performance where distillate and bottom product meet specifications

[siretb]

Hi,
There are three points to keep in mind
1) The fenske equation is a simplified equation, a shortcut. Because the relative volatility is not constant, then deviations from "optimum" and Fenske do exist.
2) As already stated the tray efficiency is not 100%. In an actual column, if it has been optimized, the feed location will be at a location where the ffed compostion is equal to the tray composition.. The tray efficiency has no reason to be constant all over the column. It varies with physical propoertise, liquid and vapor rates..
3) What is the effect of a misplaced feed location. Well some trays will work for little or nothing. Briefly said you do not use their separative power.

You say in your first post that the feed location should be located below 90% of trays, but that the actual feed is higher. If you want to proceed further the next step would be to run a full simulation, not a shortcut to verify whether the feed location is located optimally.

[ashokg9]

Thank you Ahmad and Siretb.

Actually, I have to design a packed column. For that I first calculated no. of trays and using the HETP from the packing manufacturer + 6" - 12" allowance (According to distillation design by Kister*) I calculated actual HETP.

Now, to locate the feed point in packing, (Correct me if I am wrong)
1) I have to calculate tray compositions
(whether i have to consider tray efficiencies here....?)

2) Identify the tray where the composition of tray matches with the composition of the feed
3) Calculate the location(distance) of feed point using HETP

Thanks,

Ashok

Edited by ashokg9, 03 October 2011 - 06:56 AM.

[siretb]

Kister, His name is Henri Z Kister
Packing: Then you use either the HETP(from vendor, for the given separation), or the HTU/NTU. For a preliminary design HETP is likeley to be enough

Yes shortcut calculate the total numdber of stages and feed location(Fenske/Underwood/Gilliland). Provided that the L/V does not change a lot below and above feed (they will change because you have a feed), just check that above the feed the column is not almost dry !), assume the the HEPT is constant, and prorate the bed heigth.
For a packing the HETP or HTU depends on the wetting rate, among others.

More rigorous. Run a rigorous simulation based on the preliminary, Fenske calculation.
Check whether the feed is properly located, if not move it. Use the HETP LRATE and VRATE to size the beds.

Keep in mind that a rate-base approach works very well for packings, but will require a good simulator.

[ashokg9]

Thank you Jedi.