Forums
Blogs
Articles
File Library
FAQ
Search
FB
Dear all,
I just found this forum looking for some answers to my molsieve-system design problem and it just seems like the perfect place to ask. If my following question has been answered before, I apologize ...and could you tell me where please?
Our system is simple enough, we are removing water from supercritical CO2 using molecular sieve. Once the molsieve is at capacity (here usually around 10% of its weight) we isolate the vessel with the molsieve, depressurize it to ambient pressure and regenerate the molsieve with a stream of air at 250C and 1.4m/s for about 2hrs. So far so easy...
Now we need to of course cool the molsieve again before we can re-pressurize it and put it back online. Since the air here is very humid we would need a proper dryer for the cooling air which is rather expensive (we are a small company). So, since most of the heat energy is in the steel anyway (1200kg steel versus 150kg molsieve) and the vessels are quite tall and skinny (3m x 0.4mOD) I was thinking to jacket the vessels and just cool them from the outside, so cool them without an internal airflow.
My question: Would you expect the molsieve bed to cool down quicker than its heat conductivity suggests (which is pretty much not) due to some internal convection etc.?
Would you think it feasible at all to do it this way or should we just get the dry-air system?
All suggestions very much appreciated!
Cheers,
roland
|
|
Cooling Molecular Sieve Without A(n) (air-) Flow
Started by Roland, Jul 24 2008 01:01 AM
11 replies to this topic
Share this topic:
#2
Art Montemayor
-
- Admin
-
- 5,782 posts
Gold Member
Posted 24 July 2008 - 07:09 AM
roland:
Who has advised you that you can expect selective adsorption to occur in the supercritical state? Have you obtained this knowledge personally or through adsorbent (mol sieve) manufacturers? I have done a lot of adsorbent applications – both design and operation – but never in the supercritical phase. I don't mean to imply that it can't be done; I have always relied on experts to advise me on the actual, expected sorptivity capacity of the adsorbent I plan to work with. I have to presume you have done likewise.
I have operated adsorption dryers on literally tons of CO2 per hour – but always with the CO2 in the gaseous phase. I have also used adsorption to purify liquid NGL, and even liquid Oxygen. I have no information on the supercritical phase. I don't know what you mean by "the molsieve is at capacity (here usually around 10% of its weight)". Do you mean that the mole sieve you are using has a dynamic sorptive capacity for water in supercritical CO2 which is 10 % of the adsorbent weight? I realize this doesn't respond to your specific query, but I am curious as to what you are doing.
Your adsorbent regeneration procedure seems to be conventional – with the exception that you are proposing to use atmospheric air as the regeneration gas. I would not recommend this procedure for the simple reason that you state: the regeneration air is just a source of water for the regenerated bed and will contaminate (take up sorptive capacity) it while cooling it. If you HAVE to use humid air to heat the beds, then you can use another, dry gas to cool them – something such as nitrogen (which you can purchase in relatively small Dewar tanks, in the liquefied state). This is more expensive than the air, but you ensure that the regenerated bed is truly "regenerated" (freed of all water sorbate).
I would not resort to what you call "dry air" system. The air is certainly NOT dry. Additionally, this will be an expensive method since it requires air "dryers" as well – whether refrigeration type or adsorption. I can tell you from personal experience that you will age significantly waiting around for the hot adsorber beds to cool down via natural, internal convection currents and conductive heat transfer through the bed and the vessel walls. This "natural" process will take well into the next day.
You could use one of 2 other options (if you don't want to employ a dry, imported regen gas). The first option involves redesigning your vessels – which is going to cost new capital monies – or employing some of your produced dry, supercritical CO2. I have designed and operated gaseous CO2 adsorption dryers based on using a humid CO2 side stream as the regeneration gas source. I normally recycle it back to the feed gas. To do this, I am in reality sacrificing a portion of the adsorber bed to store the water contained in the regen gas during the cooling cycle. That means, in other words, that I have to oversize my beds as opposed to regenerating with dry gas. Also, I have to regenerate in a CO-CURRENT fashion – as opposed to a counter-current fashion when using dry regen gas. What you would do would be to use the humid air to both heat AND cool the beds – but in a con-current fashion.
The second option is to use a portion of the supercritical product (stored somewhere) by vaporizing it, heating it and recycling it to the beds as regen gas. Simply expanding the supercritical CO2 is not going to do it. You have to vaporize by expansion, heat it and sublime some of the resulting dry ice formed. This can get complicated.
I'll let you digest the above and get back to us – perhaps with a complete description of ALL your basic data and scope of work. I hate working in the dark trying to help other engineers who keep information and basic data to themselves. I can't make serious, practical recommendations when I don't know ALL the basic data and scope.
Who has advised you that you can expect selective adsorption to occur in the supercritical state? Have you obtained this knowledge personally or through adsorbent (mol sieve) manufacturers? I have done a lot of adsorbent applications – both design and operation – but never in the supercritical phase. I don't mean to imply that it can't be done; I have always relied on experts to advise me on the actual, expected sorptivity capacity of the adsorbent I plan to work with. I have to presume you have done likewise.
I have operated adsorption dryers on literally tons of CO2 per hour – but always with the CO2 in the gaseous phase. I have also used adsorption to purify liquid NGL, and even liquid Oxygen. I have no information on the supercritical phase. I don't know what you mean by "the molsieve is at capacity (here usually around 10% of its weight)". Do you mean that the mole sieve you are using has a dynamic sorptive capacity for water in supercritical CO2 which is 10 % of the adsorbent weight? I realize this doesn't respond to your specific query, but I am curious as to what you are doing.
Your adsorbent regeneration procedure seems to be conventional – with the exception that you are proposing to use atmospheric air as the regeneration gas. I would not recommend this procedure for the simple reason that you state: the regeneration air is just a source of water for the regenerated bed and will contaminate (take up sorptive capacity) it while cooling it. If you HAVE to use humid air to heat the beds, then you can use another, dry gas to cool them – something such as nitrogen (which you can purchase in relatively small Dewar tanks, in the liquefied state). This is more expensive than the air, but you ensure that the regenerated bed is truly "regenerated" (freed of all water sorbate).
I would not resort to what you call "dry air" system. The air is certainly NOT dry. Additionally, this will be an expensive method since it requires air "dryers" as well – whether refrigeration type or adsorption. I can tell you from personal experience that you will age significantly waiting around for the hot adsorber beds to cool down via natural, internal convection currents and conductive heat transfer through the bed and the vessel walls. This "natural" process will take well into the next day.
You could use one of 2 other options (if you don't want to employ a dry, imported regen gas). The first option involves redesigning your vessels – which is going to cost new capital monies – or employing some of your produced dry, supercritical CO2. I have designed and operated gaseous CO2 adsorption dryers based on using a humid CO2 side stream as the regeneration gas source. I normally recycle it back to the feed gas. To do this, I am in reality sacrificing a portion of the adsorber bed to store the water contained in the regen gas during the cooling cycle. That means, in other words, that I have to oversize my beds as opposed to regenerating with dry gas. Also, I have to regenerate in a CO-CURRENT fashion – as opposed to a counter-current fashion when using dry regen gas. What you would do would be to use the humid air to both heat AND cool the beds – but in a con-current fashion.
The second option is to use a portion of the supercritical product (stored somewhere) by vaporizing it, heating it and recycling it to the beds as regen gas. Simply expanding the supercritical CO2 is not going to do it. You have to vaporize by expansion, heat it and sublime some of the resulting dry ice formed. This can get complicated.
I'll let you digest the above and get back to us – perhaps with a complete description of ALL your basic data and scope of work. I hate working in the dark trying to help other engineers who keep information and basic data to themselves. I can't make serious, practical recommendations when I don't know ALL the basic data and scope.
#3
Roland
-
- Members
-
- 7 posts
Brand New Member
Posted 25 July 2008 - 02:12 AM
Dear Art,
thank you for your immediate response!
First, more about what we do and know (and not know):
The process and pilot system was designed for us by a company specializing in supercritical CO2 systems. We use the supercritical CO2 as a drying medium in a loop at 200bar and 60C. The wet CO2 is dried with mol sieve. Under these conditions the mol sieve absorbs about 10% of its weight in water (we don't understand why it absorbs only half of the water it would from air under atmospheric pressure). In the pilot system the mol sieve is in a cage and we transfer the cage with the wet mol sieve to a regeneration vessel where it is dried with hot air and then cooled with ~60C air. This only works because the seal between the bottom of the cage and the air inlet in the regeneration vessel does not seal and we are in fact heating the cage with hot air bypassing the cage and removing the water with only a tiny purge flow of air that actually goes through the bed. The same goes for the cooling, the only reason we do not end up with a cage full of dripping wet mol sieve is because most of the air bypasses the bed and cools the cage and only a little bit goes through. The mol sieve is still quite dry after cooling, surprisingly enough (we are weighing the cages before we put them back into the system) - but how dry it really is and how dry the CO2 is in our process we do not know.
This was obviously not what the designer had in mind, but it kind-of-works. We assume that the process is not affected because it looks like it can tolerate slightly wet CO2 as long as the bulk of the water is removed by the mol sieve (which is something we do not know for certain because we have never used properly dry CO2).
Now, we are building a (pilot-)production system. We abandoned the whole transfer-the-cage-and-get-lucky-with-the-way-the-flow-goes design since we could not imagine how to do it on a larger scale. So we need to regenerate the mol sieve bed in place.
Your advice:
Do not take the 'dry' air. OK! We did not like it either!
Mol sieve beds cool 'naturally' only in geologically useful timescales: OK! We need some flow! (or maybe hire younger staff?)
Consider using some of the dry CO2: We are a bit loth to lose too much CO2, so we would rather not - unless there is no other way. We can cool just the mol sieve down to ~120C (from 200C, calculated very roughly averaging density and heat capacity of CO2 over 50-200bar and 200-100C) during the re-pressurization since it comes in at 0-5C - but then we need to cool the steel, too.
Dry N2 for the cooling step: Our problem is that the system is relatively big (at least for us): We are currently only building 10% of the final capacity: 5 mol sieve bed vessels a' 1200kg of steel and 150kg of mol sieve each. Assuming heat capacities of 0.5, 1, 0.55 kj/kgK for stainless steel, air and mol sieve, respectively and a temperature difference of 150C (200C-50C) we would need to dissipate around 28kWh of heat per vessel per cycle. For this we would need at the very very theoretical least 680kg of N2 (not sure what a good estimate for a real amount would be, 2 times? 4 times?). That's quite a bit of N2.
We might take a hybrid approach and use dry N2 for the mol sieve bed while we cool the steel from the outside with air.
One alternative we came up with was to create a closed circuit during the cooling step to circulate the (initially) hot air through the vessel, a heat exchanger and a blower and thus cool the mol sieve down while we are blowing air past the outside of the vessel to cool the steel. Residual water would be ending up back in the mol sieve but if we do this stream con-current with the CO2 flow we should be alright... What do you think about this approach?
Cheers,
roland
thank you for your immediate response!
First, more about what we do and know (and not know):
The process and pilot system was designed for us by a company specializing in supercritical CO2 systems. We use the supercritical CO2 as a drying medium in a loop at 200bar and 60C. The wet CO2 is dried with mol sieve. Under these conditions the mol sieve absorbs about 10% of its weight in water (we don't understand why it absorbs only half of the water it would from air under atmospheric pressure). In the pilot system the mol sieve is in a cage and we transfer the cage with the wet mol sieve to a regeneration vessel where it is dried with hot air and then cooled with ~60C air. This only works because the seal between the bottom of the cage and the air inlet in the regeneration vessel does not seal and we are in fact heating the cage with hot air bypassing the cage and removing the water with only a tiny purge flow of air that actually goes through the bed. The same goes for the cooling, the only reason we do not end up with a cage full of dripping wet mol sieve is because most of the air bypasses the bed and cools the cage and only a little bit goes through. The mol sieve is still quite dry after cooling, surprisingly enough (we are weighing the cages before we put them back into the system) - but how dry it really is and how dry the CO2 is in our process we do not know.
This was obviously not what the designer had in mind, but it kind-of-works. We assume that the process is not affected because it looks like it can tolerate slightly wet CO2 as long as the bulk of the water is removed by the mol sieve (which is something we do not know for certain because we have never used properly dry CO2).
Now, we are building a (pilot-)production system. We abandoned the whole transfer-the-cage-and-get-lucky-with-the-way-the-flow-goes design since we could not imagine how to do it on a larger scale. So we need to regenerate the mol sieve bed in place.
Your advice:
Do not take the 'dry' air. OK! We did not like it either!
Mol sieve beds cool 'naturally' only in geologically useful timescales: OK! We need some flow! (or maybe hire younger staff?)
Consider using some of the dry CO2: We are a bit loth to lose too much CO2, so we would rather not - unless there is no other way. We can cool just the mol sieve down to ~120C (from 200C, calculated very roughly averaging density and heat capacity of CO2 over 50-200bar and 200-100C) during the re-pressurization since it comes in at 0-5C - but then we need to cool the steel, too.
Dry N2 for the cooling step: Our problem is that the system is relatively big (at least for us): We are currently only building 10% of the final capacity: 5 mol sieve bed vessels a' 1200kg of steel and 150kg of mol sieve each. Assuming heat capacities of 0.5, 1, 0.55 kj/kgK for stainless steel, air and mol sieve, respectively and a temperature difference of 150C (200C-50C) we would need to dissipate around 28kWh of heat per vessel per cycle. For this we would need at the very very theoretical least 680kg of N2 (not sure what a good estimate for a real amount would be, 2 times? 4 times?). That's quite a bit of N2.
We might take a hybrid approach and use dry N2 for the mol sieve bed while we cool the steel from the outside with air.
One alternative we came up with was to create a closed circuit during the cooling step to circulate the (initially) hot air through the vessel, a heat exchanger and a blower and thus cool the mol sieve down while we are blowing air past the outside of the vessel to cool the steel. Residual water would be ending up back in the mol sieve but if we do this stream con-current with the CO2 flow we should be alright... What do you think about this approach?
Cheers,
roland
#4
Art Montemayor
-
- Admin
-
- 5,782 posts
Gold Member
Posted 25 July 2008 - 08:59 AM
Roland:
I like your style. You communicate well, albeit in small doses, - but well.
Please refer to the attached Excel Workbook. Even with the absence of basic data and scope, I can't communicate efficiently without use of graphics. I don't want to be mis-interpreted and be the instrument of mistaken ideas or understanding.
I believe what you are proposing is something like I've sketched. If not, then so indicate. The process is simple. The details are reliant on experience and know-how. There is a lot yet to apply with respect to safe and consistent operation. Those are details that can only be worked out when ALL the basic data and scope of work is identified. You are in control of all the details, so that is in your hands.
I believe that what I show is the most practical method of applying an efficient regeneration cycle to what you are operating. The size and type of blower is something that may be critical due to the un-identified pressure drop. The process calculations for the vessels should tell you what was expected by the designer and what you have to overcome.
I hope this helps to convey my thoughts and suggestions on your application.
Drying_SuperCritical_CO2.xls 167.5KB
96 downloads
I like your style. You communicate well, albeit in small doses, - but well.
Please refer to the attached Excel Workbook. Even with the absence of basic data and scope, I can't communicate efficiently without use of graphics. I don't want to be mis-interpreted and be the instrument of mistaken ideas or understanding.
I believe what you are proposing is something like I've sketched. If not, then so indicate. The process is simple. The details are reliant on experience and know-how. There is a lot yet to apply with respect to safe and consistent operation. Those are details that can only be worked out when ALL the basic data and scope of work is identified. You are in control of all the details, so that is in your hands.
I believe that what I show is the most practical method of applying an efficient regeneration cycle to what you are operating. The size and type of blower is something that may be critical due to the un-identified pressure drop. The process calculations for the vessels should tell you what was expected by the designer and what you have to overcome.
I hope this helps to convey my thoughts and suggestions on your application.
Drying_SuperCritical_CO2.xls 167.5KB
96 downloads
#5
joker1
-
- ChE Plus Subscriber
-
- 11 posts
Junior Member
Posted 25 July 2008 - 02:50 PM
Have you tried to circulate the air through a fin-fan or water-cooled heat exchanger ?
#6
pawan
-
- ChE Plus Subscriber
-
- 162 posts
Gold Member
Posted 25 July 2008 - 11:25 PM
Dear Roland
Probably your CO2 SCF supplier must be Thar Technologies. anyway I dont know CO2 Flow & its moisture content but still it seems that adsorber size is very large. Also probably you need to consult good adsorber system supplier who can properly size it. Since its liquid phase adsorbent you need following steps.
1. Adsorption cycle - passing of SCF CO2 through bed.
2. After completion of above step you need to drain entrapped liquid in the bed.
3. Then you need carrier gas which will first remove liquid drops of CO2 in vapor phase & condense them at top. This will be at some higher temperature. OR you can use depressurisation to some extent may be by 10 kg or so....Currently I cannot ascertain these nos as I have to look for them.
4. Then further higher temp of carrier gas to desorb the moisture from bed. Art can help you here based on his vast experience.
5. Then you need cooling.
Generally a closed loop dry nitrogen circuit with blower & heater is used with a condenser after beds to remove moisture for large systems, so that you only need makeup nitrogen.
In this case probably you can also use dry CO2 in gaseous form.
I am attaching slightly modified sketch for using closed loop which may use either N2 or CO2.
Probably your CO2 SCF supplier must be Thar Technologies. anyway I dont know CO2 Flow & its moisture content but still it seems that adsorber size is very large. Also probably you need to consult good adsorber system supplier who can properly size it. Since its liquid phase adsorbent you need following steps.
1. Adsorption cycle - passing of SCF CO2 through bed.
2. After completion of above step you need to drain entrapped liquid in the bed.
3. Then you need carrier gas which will first remove liquid drops of CO2 in vapor phase & condense them at top. This will be at some higher temperature. OR you can use depressurisation to some extent may be by 10 kg or so....Currently I cannot ascertain these nos as I have to look for them.
4. Then further higher temp of carrier gas to desorb the moisture from bed. Art can help you here based on his vast experience.
5. Then you need cooling.
Generally a closed loop dry nitrogen circuit with blower & heater is used with a condenser after beds to remove moisture for large systems, so that you only need makeup nitrogen.
In this case probably you can also use dry CO2 in gaseous form.
I am attaching slightly modified sketch for using closed loop which may use either N2 or CO2.
Attached Files
-
Sieves.xls 37.5KB
76 downloads
#7
Roland
-
- Members
-
- 7 posts
Brand New Member
Posted 28 July 2008 - 04:14 AM
Joker1:
Thanks for the idea. I have not heard of a fin-fan before but we are definitely exploring the idea of a water cooled heat exchanger for a closed loop.
Cheers!
Pawan:
We have pretty much decided on the size of the system, the vessels are on order etc. - this does not mean that we cannot change them anymore but it will be costly (pressure vessels are not cheap).
We planned to completely de-pressurize the system for the regeneration and blow hot air at only slightly elevated pressure through - but we have never thought about leaving the system at high pressure for the regeneration. Interesting idea though, I will have to talk to some heat-exchanger/condenser suppliers to get an idea where the benefits would be and how much a 200bar/250C unit might cost (probably prohibitive).
Thanks!
Art:
thank you, I like your style, too - very sorry about the small doses, though. Believe me, it is not because our system is so top secret or anything, it, embarrassingly, is because we do not know too much about it.
Your drawing looks pretty much like our planned system - but we are planning to have five drying/regeneration vessels since the regeneration times are long (~5h = 2x30min for de- and re-pressurization and 2h each for regeneration and cooling) compared to the drying time (1.5h). Sure, it would be less complicated to have just two vessels - and we would need only a small number of 3", 3000psi, 250C, food grade ball valves at ~8k USD each. Unfortunately it appears that pressure vessels get exponentially more expensive with increasing size and it looks (a lot) cheaper to have a bunch of small ones instead.
Another difference would be that we will have to have another air/N2 line whether in a closed loop or not to cool the molecular sieve without wetting it again.
The options I see there are:
1- Jacket the vessel, cool through the jacket and isolate the vessel during cooling and hope that the CO2 system can cope with the extra heat.
2- N2 or closed air loop to cool the molecular sieve (and the entire vessel).
3- Jacket the vessel, cool through the jacket and send an N2 stream or closed loop air flow through to cool just the molecular sieve
I am very much in favor of the 3rd option, mainly because it was my idea, no, seriously, I like the jacketed vessel because it will give us options when we realize during commissioning that our assumptions were not quite right...
What would you think?
You also spotted one of our biggest unknowns, the pressure drop over the molecular sieve (4A, 2-5mm diameter) bed. The (recently unavailable) designers had said 0.5bar over 3m (290mm diameter) of molecular sieve at an air speed of 1.2m/s, the molecular sieve suppliers calculated 0.05bar (in a huge vessel, 100tons of stainless, 3m diameter and a bed height of 3m as well) and we are measuring 0.15 bar over 2m (150mm diameter) at 1.4m/s in our pilot plant after we stuffed a few meters of rubber hose (seal) between the cage with the molecular sieve and the inner wall of the regeneration vessel. Is there a thumb rule for the expected pressure drop for molecular sieve beds?
From what I have read I got the impression that the flow through the molecular sieve bed is not that important as long as the bed is heated properly and the flow does not fall below ~0.2m/s. Only because the molecular sieve is usually heated with the hot air higher air speeds are used.
Is that assumption correct?
Do you think it makes any sense at all to jacket a vessel to aid regeneration during the heating and cooling cycles?
Sorry, Art, so many questions and so little we really know... would be great if you would have a bit more advice.
Thank you!
Cheers,
roland
Thanks for the idea. I have not heard of a fin-fan before but we are definitely exploring the idea of a water cooled heat exchanger for a closed loop.
Cheers!
Pawan:
We have pretty much decided on the size of the system, the vessels are on order etc. - this does not mean that we cannot change them anymore but it will be costly (pressure vessels are not cheap).
We planned to completely de-pressurize the system for the regeneration and blow hot air at only slightly elevated pressure through - but we have never thought about leaving the system at high pressure for the regeneration. Interesting idea though, I will have to talk to some heat-exchanger/condenser suppliers to get an idea where the benefits would be and how much a 200bar/250C unit might cost (probably prohibitive).
Thanks!
Art:
thank you, I like your style, too - very sorry about the small doses, though. Believe me, it is not because our system is so top secret or anything, it, embarrassingly, is because we do not know too much about it.
Your drawing looks pretty much like our planned system - but we are planning to have five drying/regeneration vessels since the regeneration times are long (~5h = 2x30min for de- and re-pressurization and 2h each for regeneration and cooling) compared to the drying time (1.5h). Sure, it would be less complicated to have just two vessels - and we would need only a small number of 3", 3000psi, 250C, food grade ball valves at ~8k USD each. Unfortunately it appears that pressure vessels get exponentially more expensive with increasing size and it looks (a lot) cheaper to have a bunch of small ones instead.
Another difference would be that we will have to have another air/N2 line whether in a closed loop or not to cool the molecular sieve without wetting it again.
The options I see there are:
1- Jacket the vessel, cool through the jacket and isolate the vessel during cooling and hope that the CO2 system can cope with the extra heat.
2- N2 or closed air loop to cool the molecular sieve (and the entire vessel).
3- Jacket the vessel, cool through the jacket and send an N2 stream or closed loop air flow through to cool just the molecular sieve
I am very much in favor of the 3rd option, mainly because it was my idea, no, seriously, I like the jacketed vessel because it will give us options when we realize during commissioning that our assumptions were not quite right...
What would you think?
You also spotted one of our biggest unknowns, the pressure drop over the molecular sieve (4A, 2-5mm diameter) bed. The (recently unavailable) designers had said 0.5bar over 3m (290mm diameter) of molecular sieve at an air speed of 1.2m/s, the molecular sieve suppliers calculated 0.05bar (in a huge vessel, 100tons of stainless, 3m diameter and a bed height of 3m as well) and we are measuring 0.15 bar over 2m (150mm diameter) at 1.4m/s in our pilot plant after we stuffed a few meters of rubber hose (seal) between the cage with the molecular sieve and the inner wall of the regeneration vessel. Is there a thumb rule for the expected pressure drop for molecular sieve beds?
From what I have read I got the impression that the flow through the molecular sieve bed is not that important as long as the bed is heated properly and the flow does not fall below ~0.2m/s. Only because the molecular sieve is usually heated with the hot air higher air speeds are used.
Is that assumption correct?
Do you think it makes any sense at all to jacket a vessel to aid regeneration during the heating and cooling cycles?
Sorry, Art, so many questions and so little we really know... would be great if you would have a bit more advice.
Thank you!
Cheers,
roland
#8
Mehrdad
-
- Members
-
- 57 posts
Gold Member
Posted 28 July 2008 - 06:25 AM
hi roland
i hope you will find a good solution for your problem.
i don't advice another gas except air for cooling the molsive.
you have a regional problem depend on weather condition that make problem in your plant and the effect of humidity increases in warm monthes.
i suggest that you add more molsive package and use a recirculating system for cooling the cages and a fin cooler for cooling recirculated air or
use an air dryer based on air cooling system (dehumidification system by cooling the air)that it need a chiller and an aircooler system .
in this way you not only eliminate the the air humidity,but also reducing the time of cooling the cages.
regard
i hope you will find a good solution for your problem.
i don't advice another gas except air for cooling the molsive.
you have a regional problem depend on weather condition that make problem in your plant and the effect of humidity increases in warm monthes.
i suggest that you add more molsive package and use a recirculating system for cooling the cages and a fin cooler for cooling recirculated air or
use an air dryer based on air cooling system (dehumidification system by cooling the air)that it need a chiller and an aircooler system .
in this way you not only eliminate the the air humidity,but also reducing the time of cooling the cages.
regard
#9
Roland
-
- Members
-
- 7 posts
Brand New Member
Posted 05 August 2008 - 10:29 PM
Hi Mehrdad,
thanks for the suggestion! We are definitely looking into a separate loop to re-circulate air through a cooler during the cooling stage.
Our biggest problem though is that we have these enormous vessels (1200kg of stainless per 150kg of MS). No matter which way we calculate it we will need a huge flow of hot air to simply heat it up and then a similarly big flow to cool it down again... Just the amount of energy we waste this way is huge - not to mention the capital investment for the blowers, heaters etc.
We will still do it this way if we can't come up with anything better...
What we are currently investigating is trying to immerse a (pressure and temp proof) microwave emitter in the MS to heat up and evaporate just the water and then remove the steam with a bit of a scavenger flow of air over the MS. I have found a couple of patents/papers regarding this matter and we are talking to a microwave manufacturer - and it seems, in principle, feasible. Only question is if we can make it work for our slightly demanding application.
If this would work it would obviously solve lots of problems, less energy would be needed, shorter cooling time etc. Am I missing anything?
Does anybody have any experience in using microwaves to regenerate MS?
Cheers!
thanks for the suggestion! We are definitely looking into a separate loop to re-circulate air through a cooler during the cooling stage.
Our biggest problem though is that we have these enormous vessels (1200kg of stainless per 150kg of MS). No matter which way we calculate it we will need a huge flow of hot air to simply heat it up and then a similarly big flow to cool it down again... Just the amount of energy we waste this way is huge - not to mention the capital investment for the blowers, heaters etc.
We will still do it this way if we can't come up with anything better...
What we are currently investigating is trying to immerse a (pressure and temp proof) microwave emitter in the MS to heat up and evaporate just the water and then remove the steam with a bit of a scavenger flow of air over the MS. I have found a couple of patents/papers regarding this matter and we are talking to a microwave manufacturer - and it seems, in principle, feasible. Only question is if we can make it work for our slightly demanding application.
If this would work it would obviously solve lots of problems, less energy would be needed, shorter cooling time etc. Am I missing anything?
Does anybody have any experience in using microwaves to regenerate MS?
Cheers!
#10
Mehrdad
-
- Members
-
- 57 posts
Gold Member
Posted 06 August 2008 - 01:35 AM
hi
about MW and ability for regeneration of MS and mechanism of action i can't accept the proposal according to MW effect mechanism and apparatus.
but everything is possible and science is in progress !
i suggest that you will change the system specialy reducing the mass of cages and bulid cages structure by the pipe because:
1-reducing the mass of cages
2-heating and cooling by water flow through the pipes structure
by using an alternate cooling-heating program, you can save a lot of energy.
i guess the price of this changes be less than other ways.
tell me if you need more detail.
cheers
about MW and ability for regeneration of MS and mechanism of action i can't accept the proposal according to MW effect mechanism and apparatus.
but everything is possible and science is in progress !
i suggest that you will change the system specialy reducing the mass of cages and bulid cages structure by the pipe because:
1-reducing the mass of cages
2-heating and cooling by water flow through the pipes structure
by using an alternate cooling-heating program, you can save a lot of energy.
i guess the price of this changes be less than other ways.
tell me if you need more detail.
cheers
#11
Roland
-
- Members
-
- 7 posts
Brand New Member
Posted 11 August 2008 - 03:11 AM
mehrdad,
we hope that the microwave system will work - but we probably won't be able to get a system in time for our plant. We are working with a supplier though to see if we can develop a system for the later stages of our plant.
In the meantime we are focusing on reducing the amount of energy lost by heating the steel. We will probably insert a ptfe (or similar) liner into the vessel for insulation. Sure, it is not a good insulator but it survives the temperatures, pressures and solvents (supercriticial CO2) in our process and it definitely insulates a bit. Would you know a better alternative? Do yuo have experience with coating steel with an insulating material?
We can't really do much about the vessels, they need to withstand 200bar and are hence a bit fat.
Cheers!
roland
we hope that the microwave system will work - but we probably won't be able to get a system in time for our plant. We are working with a supplier though to see if we can develop a system for the later stages of our plant.
In the meantime we are focusing on reducing the amount of energy lost by heating the steel. We will probably insert a ptfe (or similar) liner into the vessel for insulation. Sure, it is not a good insulator but it survives the temperatures, pressures and solvents (supercriticial CO2) in our process and it definitely insulates a bit. Would you know a better alternative? Do yuo have experience with coating steel with an insulating material?
We can't really do much about the vessels, they need to withstand 200bar and are hence a bit fat.
Cheers!
roland
#12
Mehrdad
-
- Members
-
- 57 posts
Gold Member
Posted 12 August 2008 - 04:22 AM
Hi Roland
insulating of cages only decrease the rate of heat transfer and performance of insulation depend on the time of heating and insulation character.
if the time of heating be long the cages at last will be hot as heating medium temperatur and now for cooling the MS you have a new problem because insulation don't let the cages body to transfer the heat and being cool.
bye
insulating of cages only decrease the rate of heat transfer and performance of insulation depend on the time of heating and insulation character.
if the time of heating be long the cages at last will be hot as heating medium temperatur and now for cooling the MS you have a new problem because insulation don't let the cages body to transfer the heat and being cool.
bye
Similar Topics
Heat Exchanger Steam FlowStarted by Guest_aliebrahem17_* , 25 Nov 2024 |
|
|
||
Flow Through Normally No Flow LineStarted by Guest_iippure_* , 08 Apr 2025 |
|
|
||
Flare Header Reverse FlowStarted by Guest_Ahmadhamzahperta_* , 04 Apr 2025 |
|
|
||
Dynamic Simulation After Feed Flow ReductionStarted by Guest_Kakashi-01_* , 20 Mar 2025 |
|
|
||
Smr Reformer Flue Gas FlowStarted by Guest_kaidlut_* , 14 Feb 2025 |
|
|
