12 replies to this topic

im designing the molecular sieve adsorption bed for drying of gas containing N2/CO2/O2/H2O at 40degc. water content of dry gas at outlet should be less than 1 ppm. here the regeneration gas is split stream of outlet dry gas(around 10% of dry gas). during regeneration, the regeneration gas is cooled to inlet temperature(40 degC), water is removed from gas in knock out drum and then mixing with the inlet stream.
the bed has to be designed for removing the water from the feed gas and saturated water content in regeneration gas. im facing problem while calculating the saturated moisture content in the regeneration gas. from my understanding the regeneration gas wont be saturated during cooling period. then we have to considered the moisture content during only heating hrs or otherwise some particular time during heatingperiod .can anyone help in this issue and my approach is correc or not?.....Thank You inadvance.

[ankur2061]

honeychem,

Your explanation is confusing. Can you furnish a sketch of your system as a excel worksheet and point out where exactly are you facing a problem. It will helps us understand better to help you.

In general the mixed saturated gas (fresh saturated gas + regeneration gas) water content will be the function of the pressure and temperature of the mixed saturated gas entering the molecular sieve bed under steady state equilibrium conditions.

Regards,
Ankur.

[Propacket]

honeychem

What I perceive from your posting is that you want to know whether the regeneration gas is saturated or not. Without providing requisite data, nobody in this forum can tell you that whether or not regeneration gas is saturated. Please supply conditions of gas (temperature and pressure) and gas composition.
If you want to learn how to predict gas saturation, you must consult McKetta chart provided in GPSA Chapter 20-Dehydration. This chart will give you saturated gas water content. In addition to this, following equation will give you a rough idea about saturated gas water content.

W=47,484x Vapor Pressure of Water at specified T and P /Total Pressure

In this, the water content of a gas is assumed to be equal to the vapor pressure of pure water divided by the total pressure of the system. This yields the mole fraction of water in the gas and this is value converted to lb water per MMCF by multiplying by 47,484.

Thanks

Edited by P.Engr, 24 August 2010 - 05:09 AM.

Thank You Ankur

Here, I'm attaching the flow scheme. I want to calulate the amount of desiccant required for the application. I have to calculate the total moisture content in the mixed gas entering into the adsorption bed. From feed conditions, I have calculated the moisture in the feed. How do I calculate the moisture in the regen gas?  Adsorption.xls   52.5KB   176 downloads

[Propacket]

honeychem,

• I have some issues related to your furnished information. There is always considerable pressure drop through absorption as well as regeneration circuit. Your sketch shows that regeneration gas being mixed with feed gas is at same pressure as that of fed gas. This is not possible unless you install a regeneration gas compressor. However, your sketch does not show any compressor.

• Feed gas is not saturated. Feed gas saturation water content is around 1.09 mole% (10,900 ppm).

• Considering 2000 ppm of water in feed gas, water mass flow rate in feed gas is 70.36 Kg/hr.

• Water mass flow rate in regeneration gas after cooling is 70.36 Kg/hr. The regeneration gas is saturated. 30.78 Kg/hr of water will drop in the separator as liquid water. Remaining water (39.58 kg/hr) will be carried with the regeneration gas. Therefore, total water feed to absorption is 70.36+39.58=109.94 Kg/hr which in ppm becomes 2838 ppm.

Hope this will help.

Waiting for Ankur to confirm the above figures.

Thanks

Edited by P.Engr, 25 August 2010 - 06:27 AM.

[ankur2061]

The water content in feed gas as 2000 ppmv is 70.34 kg/h as described by P. Engr.

I am assuming that P. engr. has done the mass balance using some simulator since he has been able to predict the water separation in the separator.

The only question is whether the moisture content in the gas is based on saturation conditions of pressure and temperature. Based strictly on saturation conditions of 5.8 barg and 40 deg C the figure of 2000 ppmv or 70.34 kg/h seems to be low.

Regards,
Ankur.

[Zauberberg]

Ankur is right. At 5.8 barg pressure and 40 degC temperature, saturation water content is ~11,500ppmv. That would be the water content of saturated Regeneration gas at given process conditions.

Assuming that you have 2,000ppmv moisture in the feed gas (well below saturation point), that is equivalent to 1,440kg water adsorbed during 12hrs active cycle. In order to desorb this water during 8hrs heating cycle, you need water "outflow" of 180kg/hr with the regeneration gas. After knocking the water out downstream of the cooler, you remain with 11,500ppmv equilibrium water content in the 6,000 kg/hr Regeneration gas (= 40 kg/hr water), which means that 140kg/hr has been knocked out as liquid in the Regen gas KO drum.

Thank you, this is good news for new visitors.

________________

Thank you Zauberberg.

Your approch is correct. finally we have to design the bed for removing 1440+(40*8)=1760 kgs of water. but water outflow of (1440/8=)180kg/hr in regeneration gas is acceptable? when i discussed with vendor he told that the 80% of water will desord within the first two hours. my doubt is how to predict the water content for remaining hours. if any anybody had experience with desorption bed characteristics and adorption bed dynamics please give some suggestions.

Thank You in advance

Thank you P.Engr

sorry. the pressure drop across the bed is 0.8kg/cm2. The feed saturation is around 12% RH.

[Zauberberg]

There is an intereseting discussion at: http://www.cheresour...n-for-tsa-unit/ including a technical paper on Adsorber design, posted by Art Montemayor. I would suggest you to go through the post.

[Zauberberg]

You don't need to consider water content of Regeneration gas other than the one corresponding to saturated condition. Why? Because this is the highest load you will be seeing on the active bed, at that time when saturated Regen gas is recycled to the feed gas. Also remember that the Regen gas always passes through the KO drum before being recycled to Feed, so it picks up the mositure from the vessel itself where it is in equilibrium with liquid water.

Consider 120 kg/hr water in the feed gas, and 40 kg/hr water with the Recycle gas - for 12hrs adsorption cycle. That's what I would do, for sure. Otherwise I might face the possibility of premature water breakthrough during operation, and probably start polishing my CV and looking for another job.

[Art Montemayor]

PEngr and Zauberberg are right. The regen gas being recycled is considered saturated with water vapor. This regen gas - as depicted in the honeychem schematic - cannot flow as shown. It requires a driving force to return back to where it came from.

Please refer to the revised workbook I am attaching for further information. Most, if not all, of these issues are dealt with and discussed in prior threads on our Forums. All you have to do is use the SEARCH machine to find them. Use the word "Adsorption" as a key.

Attached Files

•  Adsorption Rev1.xls   221.5KB   212 downloads