4 replies to this topic

[jknight291]

I have been charged with creating a plot of Dew Pt. Temperature of Air vs. Time for a truck-mounted air dyer.

 

Basically, air pulled through a desiccant dryer (using activated alumina) by a compressor is used in order to unload anhydrous NaOH off of a truck. We are having some issues with moisture contaminating the caustic in the unloading process and I have been asked to produce a chart of the Dew Pt Temperature (of the dried air at various relative humidity) versus time. I believe the idea is that the desiccant is saturated at some point in the unloading process and the air is not being dried enough before making contact with the caustic.

 

So, my thought process was to assume atmospheric conditions (80ºF and 80% RH) and determine the amount of water vapor pulled into the dryer (at a rate of 450 CFM air). I would then model the adsorption of the water on the alumina so that I could (via material balance) determine the amount of water vapor and air coming out of the dryer (which would then allow me to determine the dew pt temperature).

 

However, I am struggling with this I am having difficulty in developing a dynamic model of water adsorption to the activated alumina. Seems like this would be a property provided by the manufactuer? We are using BASF F-200. I am also going off of pure memory from Mass Transfer / digging around on the Internet as I do not have access to any textbooks currently. I know there is probably a mass transfer model, but I wasn't sure if there was a quick and dirty estimate floating around since I am not trying to actually design any equipment.

 

Basic info: 450 CFM air, 80% RH, 80ºF, 1150lb 3/16" BASF F-200 Activated Aluminum.

 

Any suggestions?

 

Thanks!

 

 

[Art Montemayor]

Joe:

 

This seems to be a very simple and easy project – the drying of the air and not the generation of a dew point curve.  There are various questions I have to ask of you:

1. Are you using a single dryer vessel to dry the air while unloading the anhydrous NaOH off of the tanker-truck?
2. Is the operation batch-wise?  In other words, do you unload and then regenerate the activated alumina to have it ready for the next subsequent unloading?
3. You say “air pulled through a desiccant dryer (using activated alumina) by a compressor”.  Do you mean that the dryer vessel is located on the suction side of the air compressor?  If so, this is not the proper way to dry air.
4. How do you regenerate the spent activated alumina bed?  Do you use dry or moist gas?  What is the temperature of the regenerating gas going into the adsorbent bed?  Where do you introduce the hot regen gas into the bed – is it introduced in the same flow direction as the incoming moist air when the bed is drying?  What is the maximum temperature of the regenerated bed prior to cooling it down?
5. How was the amount of adsorbent (1,150 lb?) used in each bed arrived at?  Can you submit the calculations?

You should be assuming 100% humidity in the air entering the dryer.  That means it is essentially saturated with water moisture.  If I want to dry air, I certainly would use activated alumina.  But I would design the adsorbent bed to retain ALL of the water moisture entering with the incoming air.  My adsorbent bed would be regenerated at a minimum of 450 -500 oF if possible.  The dew point of the air out of the dryer should be approximately -90 oF (about 1.0 ppmv).  Activated alumina can dry your air so well that the attempts to predict any dew point with respect ot time are ludicrous.  I believe you are wasting your time trying to develop dew point curves.  You either have an adequate dryer or you don’t.  Checking the sizing of the dryer will tell you where you stand at the moment and this will alert you to what to do next.  Please refer to the attached Excel workbook and look at the manner that I visualize the dryer should be working.

 

If you are using an air compressor to drive the air, then the smart way to maximize the efficiency of removing the water vapor in the air is to compress the air first, cool the discharge in an aftercooler, and subsequently separate the condensed water in the cool, compressed air in a vapor-liquid separator.  You drain the condensed water collected in the separator with an automatic float drainer and send the cool, compressed air to the top of the adsorber vessel – as shown in the diagram.  Since you have not told us how you are regenerating the adsorbent, I have had to show the regeneration taking place after the unloading by using an air blower and air heater.  The hot, atmospheric air is sent to the top of the bed and exits at the bottom.  This regeneration air is assumed to be saturated with water, so an extra amount of adsorbent has to be used in the dryer bed.

 

If you are a graduate engineer, I can send you typical adsorber dryer calculations that can guide you to size or rate your vessel.

 

 Air Dryer for Unloading.xlsx   14.43KB   226 downloads

 

[jknight291]

Art,

What I should have mentioned is that this system was designed and implemented years and years ago before the plant that I work at had an MOC program, so there is virtually no supporting documentation for this system design (my unit engineer had to count the number of desiccant packages in the trash in order to verify the amount of desiccant used), so I am currently going off of tribal knowledge from my operators.

 

I was also originally misinformed on the reason behind this project. There is NOT an issue with moisture getting into the anhydrous NaOH, our customer just wants a curve of predicted dew pt temperature versus time so that they can write an SOP for offloading the project. They have a standard that they will not offload product once the dew point temperature of the dried air reaches -15ºF and they want to know when they can expect that temperature in regard to time.

I completely agree with you on this being a ludicrous project, but it’s what Sales has requested via the customer.

Now, to answer your questions:

1. I am using a single dryer vessel mounted to a truck (I have attached a schematic)
2. The operation is batchwise – we typical regenerate the alumina at least 24 hours prior to shipment.
3. Yes, the dryer is on the suction side, so we are drying the air prior to the compressor (which is really just a blower and only increasing the pressure by a small amount)
4. For regeneration, we heat air and send it into the top of the dryer (countercurrent to typical air flow). Once the bottom of the dryer reaches 300ºF, we connect a dry air hose to the bottom of the dryer and when the gauge at the top of the dryer reaches 100ºF, the dew point is recorded and the air ports are capped.
5. No, I cannot. This is what my operators “have always been told to do” and have absolutely no supporting documentation. I’m sure someone at some point calculated this.

I would love to (and might) do a rigorous analysis of this system to try and optimize this process, but for now, I am trying to meet a Nov. 5 deadline for a Sales Dept. Action Item by answering their question with what is currently being utilized.

Attached Files

•  drying schematic.PNG   159.74KB   30 downloads

[fseipel]

I would suggest buying an inexpensive Dwyer relative humidity probe and a loop powered indicator (~$500) -- this way you have online dewpoint monitoring at the truck.  If you couple this to an alarm horn or relay you can have the blower shut off on high humidity, addressing any customer/quality concerns.  The pressure will impact dewpoint; if it changes significantly you could use a more sophisticated instrument to compensate, though blower pressure probably doesn't vary all that much -- especially on suction where you presently have dryer.  I'm assuming tanks are configured similarly for similar backpressures.  Also consider a colorimetric humidity indicator that simply changes color, indicating when alumina is spent -- cheaper and fewer components than the electronic device.  Setting trucks up with two beds might be more convenient; when one is expended, switch to the other, and regenerate only that bed when it goes back to depot.  This way you know you always have at least half the capacity unused.  Or alter alumina tank so it includes several vents tied into a common manifold with the humidity sensor/display; opening one of the valves along bed feeding vent and sensor, will show how close you are to breakthrough vs height of bed.  Sintered filter required on each inlet.  You can also close the loop by adding a hose so that tank exhaust from filter goes back into blower inlet; even if you leave a gap (so everything is at atmospheric pressure, e.g. pipe in a larger pipe with gap between the two), most of the air will then be drawn from the already dried exhaust air (air in the solid caustic storage tank headspace will be extremely dry since NaOH will pick up that moisture), whilst preventing over/underpressurization with this scheme.  If the pipes are inside each other with a gap between them, when you first start blower it will draw some ambient air in, once it reaches steady state most air will be drawn from tank exhaust.  Adding an hour counter to the blower may also be useful, i.e. regenerate when dewpoint reaches xx OR run time hours > yy.

Edited by fseipel, 01 November 2013 - 07:46 PM.

[booski]

Hello Art,

 

I am trying to gain the information needed to design a suitable air drier, so you posted above;

"If you are a graduate engineer, I can send you typical adsorber dryer calculations that can guide you to size or rate your vessel."

 

I would love to have some type of formula that I could use for designing an air drier!

Kevin