Jump to content

  • Log in with Facebook Log In with Google      Sign In   
  • Create Account


ChExpress Blog - 9/17/14

Read the latest news from the chemical industry in Christa's blog.

Featured Articles

Check out the latest featured articles.

File Library

Check out the latest downloads available in the File Library.

New Article

Product Viscosity vs. Shear

Featured File

Air Vessel Sizing for Water Hammer Prevention

New Blog Entry

Flare Dispersion FAQs- posted in Ankur's blog

Vacuum Condenser Configuration


This topic has been archived. This means that you cannot reply to this topic.
28 replies to this topic
Share this topic:

#1 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 31 May 2010 - 02:09 PM

Maximum vapor flow rate handled by main condenser is 23,500 lbs/hr of acetone. At high rates, ( 26,000 lb/hr) I need another small condenser ( this rate is limited by flooding in columns with high efficiency trays, superfrac) to handle this rate.

Option 1: Installing vent condenser on vacuum line to pump requires more piping and hence pressure drop.

Option 2: Parallel condenser requires vapor flow meter and control valves on chilled water and vapor side( pressure drop and cost)

Option 3: To install condenser on pressure equalization line. This seems to be better as I have to upsize only one nozzle on main condenser and ofcourse piping too.

I have attached drawing in my previous post. Please help me with your advice/ recommendations on above options.

Attached Files



#2 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 4,841 posts

Posted 01 June 2010 - 02:03 AM


Chemroopa:

I haven’t seen your previous post. However, for a 10.6% capacity increase I don’t believe you have that many options. This is a relatively small increase in capacity expected in any heat exchanger. I would expect for you to have it inherently in your existing total condenser. Nevertheless, I have to give you credit for having tried it already and finding out that you are at the limit of your total condenser’s capacity.

If that is the case and you really have to have condensing capacity for the additional vapor, then I would resort to installing a small total condenser to take the additional load – very much as I’ve quickly sketched on the attached workbook. Please excuse the poor quality of my sketch since I had to slap it together very quickly. I am currently preparing to return home to Texas after a year’s stay on this project here in Spain and things are in a rush.

For a small, 10% increase, I would not be hard-pressed to build my own total condenser on plant site, in the maintenance shack. This is a small unit and shouldn’t be a big expense. I would not install a vapor flow meter or control valves on the chilled water and much less on the vapor side – they simply are not needed. I would merely employ both condensers running 100% of the time or simply put a block valve on the second condenser’s vapor line if I had to isolate one from the other (which I don’t believe is a requirement.

I hope this helps.

Attached Files



#3 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 01 June 2010 - 09:29 AM

Thanks Art.

I live in Pasadena,TX and you have a safe flight!

Over the years, we squeezed this condenser to its limit. In order to configure according to your option, I need to fabricate additional nozzles on reflux drum. Option 3 incurs less cost but I am not sure if it will effective or not.

Option 3: excess/uncondensed vapor from 4" vent line on main condenser will be condensed through secondary condenser. Vent line from secondary condenser will tie in to existing 4" nozzle on reflux drum ( one being used for 4" vent line from main condenser). Liquid from secondary condenser will tie in to main condenser liquid line prior to seal leg.

Appreciate your comments on this option.

Thanks a lot!!

#4 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 01 June 2010 - 10:34 AM

Art,

I have attached the drawing with tab name "option 3"

Thank You

Attached Files



#5 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 02 June 2010 - 07:37 AM

Please advice me on above request. Though Art suggested one option, I would like to know pitfalls of other option I mentioned in earlier post.

Thank You

#6 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 05 June 2010 - 12:14 PM

Art,

If I have a parallel condenser and no flow control valve and meter on this vapor line, doesn't it overload this condenser? For 3000 lb/hr, my line size would be 4". I am worried if it overloads small condenser.

Small condenser will have chilled water and main condenser is using cooling tower water.

Thank You

#7 pawan

pawan

    Gold Member

  • ChE Plus Subscriber
  • 162 posts

Posted 06 June 2010 - 09:08 AM

I agree with your analysis & will consider the Option-3 as best because If you put second condenser parallel it will not be able to take total vapor load & hence reflux drum will again be full of vapors requiring huge load on vaccum pump and loss of material also.

Further you are right in terms of nozzle requirement etc.

If you are planning to condense remaining vapors using chilled brine/water it will be beneficial in terms of reducing loss & improving vaccum also in the column.

Option-1 will also result in vapor losses due to situation in pump suction & hence will not work effectively. It is also not desirable to have reciever full of vapors whereas in Option-3 your reciever will also have minimum vapors due to subcooled liquid from secondary condenser if you plan it that way.

Option-2 putting a parallel condenser will require extra piping or control valve as you rightly mentioned. However the sketch provided by Art will also have following problems.

1. The system design is layout dependent.
2. The size of secondary condenser is small so it will release more vapors to reciever & therefore losses & vaccum problem.


So in my opinion you are 100% right in your analysis.

#8 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 511 posts

Posted 06 June 2010 - 11:17 AM

Art Montemayor's design with the small parallel condenser (Option 2 in chemroopa's list) is simple, elegant, robust and self controlling. No extra control valves or control circuits are required. The secret to the design is the U-leg seals between the condensers and the Reflux Drum. In a situation like this it is important to remember that it is the condenser that creates the vacuum - the vacuum pump is merely there to get rid of incondensibles. Each condenser in the parallel scheme will "suck" as much vapor as it can, according to its heat transfer capability. If you design for an additional capacity of 20% and you provide a second condenser with 20% of the area of the fist then it will simply pull in and condense 20% as much as the first would. It is not necessary in this scheme to provide the second condenser with colder water than the first, but the design is robust enough that even if you did that it would still control itself perfectly.

The condensed liquid flows through the U-leg seals, preventing the passing of vapors to the Reflux Drum. These legs must be sized to prevent syphoning the contents of the U-leg into the Reflux Drum, or the balancing vent line (shown as a finer line in Art's drawing) must be linked to the top of the inverted U-legs. I have added this detail to Art's sketch and attached it below.

Putting the second condenser in series with the existing unit will be much harder (but not impossible) to control as any change in the performance of either will immediately impact on the other.

Attached Files



#9 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 06 June 2010 - 06:44 PM

Katmar,

I completely agree with you if it were total condensers. Being partial condensers, uncondensed vapors will escape through pressure equalization line and increase load on vacuum pumps.

Thank You
Roopa

#10 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 10 June 2010 - 07:34 AM

Art and Katmar,

I ran a simulation on main condenser and found out that main codenser efficiency will drop due to excess uncondensed vapors in option 3.

I decided to go with option 2 that Katmar and Art agreed with. Now that I have pressure equalization line for both condensers, I don't need seal legs in condensate line.

Is this true statement?

Thank You
Roopa A&M 2007

#11 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 511 posts

Posted 10 June 2010 - 10:35 AM

Roopa, you have not explained why these are partial condensers and not total condensers. In his reply and sketch Art referred to them as total condensers, and I certainly took them to be total condensers from your original description. It is only recently that you informed us that they are partial condensers. What are the uncondensed vapors and where do they go?

No matter which option you select you must install separate seal legs for each condenser. Do not combine the liquid outlets from the condensers into one seal leg the way you showed on your sketch. You do not want vapors to bypass from either condenser to the reflux drum, or from one condenser to the other. Also take note of the anti-syphon measures for the seal legs that I discussed earlier and included in my sketch.

#12 Art Montemayor

Art Montemayor

    Gold Member

  • Admin
  • 4,841 posts

Posted 10 June 2010 - 02:19 PM


Roopa:

There simply is no substitute for peer checking.

I am now back from discovering the marvels and the wonders of the Alhambra with my granddaughter. We have seen and touched the wonders of the Moorish art, culture, and engineering accomplishments while in Granada and gone on to Nerja on the Mediterranean Costa del Sol. After some of the best Paella and Rioja wine that Spain can offer, I can now spend some minutes on the Forum before continuing on to Houston in another week.

When you have had the good fortune to participate in some of the real big projects, you can appreciate good, experienced and unequaled peer checking by such as Katmar's. Thank God for such veterans like him. He caught my goof in not detailing the anti-syphon lines on the “goose neck drains” I showed on the condensate drains out of each condenser. – And by the way, there has never been any prior mention of a “PARTIAL” condenser in this thread. The way you show the vacuum pump piped up to the receiver leaves one with only the idea that you are dealing with a TOTAL condenser. That is why I had to assume the same conclusion. I don’t see the manner in which you can have a partial condenser in your sketch. If, indeed, you are dealing with a partial condenser than the recommendation takes on a much more complex list of alternative options. Please confirm that you are indeed dealing with a partial condenser application and clearly show us with a detailed, clear sketch. If you are condensing the entire overhead product to a saturated state, then your original sketch is valid and our original recommendations are valid also.

Pay particular and careful attention to a recognized Unit Operations expert like Katmar. What he is telling you is 110% accurate and full of engineering smarts. This is definitely “good poop” as we called it in 1960 at TAMU. You must install separate seal legs for each condenser. Otherwise, you will install a lot of trouble and fluid mechanics nightmares. Definitely do not combine the liquid outlets from the condensers into one seal leg the way you showed on your sketch.

A vacuum pump only expels the non-condensables existing in (or leaking into) the overheads vapor system. The condensers literally “condense” everything else. This is typically what a total condenser is piped up as.


#13 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 511 posts

Posted 10 June 2010 - 04:25 PM

Welcome back Art - I had a feeling that you must have been traveling somewhere. I got very jealous when I read of your latest exploits! Well, at least here in South Africa we have the Soccer World Cup starting tomorrow and the whole country is partying like never before.

I did not regard your lack of detailing the seal legs as a "goof". I know that if you were detailing a design you would would have included that detail, but in a conceptual sketch nobody would normally include that detail. It was only because I supected that area to be where roopa was having trouble with his understanding that I added the additional detail.

#14 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 11 June 2010 - 11:44 AM

I apologize for confusion on condenser type. It is TOTAL CONDENSER with vacuum pump expelling non condensible air (leaked air).

Unfortunately, I have no other engineer(s) to peer check my design with them and managers with no chemical engineering background. I heavily rely on gained theoritical knowledge in A&M, text books and experts who are willing to share their experience and knowledge in forums.

Pardon me for bad sketch but I put in all information in the attached drawing including seal leg height.

Seal legs from two condensers will be connected to the reflux drum separately. Appreciate your comments and time.

Attached Files


Edited by chemroopa, 11 June 2010 - 11:46 AM.


#15 narendrasony

narendrasony

    Gold Member

  • Members
  • 87 posts

Posted 18 June 2010 - 02:28 PM

Dear Katmar /Art ,
Liquid outlets from both the condensors joining into one seal leg- How the vapors will bypass from one condensor to the other ? Though common seal leg seems to be awkward and should be avoided, but is it such a serious problem? Requesting you for some more insight.

Regards
Narendra

#16 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 511 posts

Posted 19 June 2010 - 06:30 AM

Narendra, There is a good chance that combining the liquid outlets into one seal leg will work perfectly well, but there is also a chance that it will not work at all. It is similar to the recommendation you will find in every steam trap vendor's literature that each coil should have its own steam trap. Two coils into one trap might work, but it might also cause severe problems. It is "good engineering practice" and it solves problems before they occur.

The liquid drains from the condensers should be designed for self venting flow. This means that there could well be a continuous vapor phase all the way down to the seal leg. If the drains are combined it is possible to have a continuous vapor phase from one condenser all the way to the other. If the pressures in the two condensers at the drain points are slightly different because of condenser or piping geometry you could induce back flow from one to the other. There are lots of "ifs" here and that is why I said it could be OK to combine the two drains. But for the cost of that little extra bit of pipe I would buy the peace of mind it gives. Also, for maintenance purposes it would be better (but not essential) to have them separate to allow one to operate without the other.

So I agree with you - probably not a serious problem, but so easy to avoid that it is not worth taking a chance.

#17 narendrasony

narendrasony

    Gold Member

  • Members
  • 87 posts

Posted 20 June 2010 - 01:11 AM

Dear Katmar,
Thanks once again for your wise advice. Single seal leg is better to be avoided, but there was temptation to go ahead without a new nozzle on the reflux drum.

I expected that flow distribution may not be as per our expectations due to piping geometry. For example, instead of desired 80%:20% (Existing : New) it may be initally say 90%:10%, then the later condensor may foul faster than the first one and progrssively the flow through the first will increase, but there would be forward flow of vapors towards both the condensor. I think this situation can be avoided with a butterfly valve on vapor feed lines to the condensors.
I'm trying to visualize how the vapors would back-flow from one condensor outlet to the second condensor with single seal-leg, I always expect a forward vapor flow to both condensors. Please help me understand this phenomenon more clearly.

Regards and thanks
Narendra

#18 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 511 posts

Posted 20 June 2010 - 01:46 AM

I am getting very confused over whose problem we are addressing? Or do we have 2 (or 3) people with identical problems? We had Chemroopa, who seems to have changed to Mahesh, and we have Narendra too. Anyway, let me stick to the understandable engineering.

To Narendra: If it is a real problem to put another nozzle on the reflux drum and the existing nozzle is big enough to avoid a syphon by using one of the methods I discussed earlier, then with careful design you could compromise by combining the liquid drains after the individual seals.

Another way to avoid syphons, and which is very widely used, is to run the piping for the liquid drain from the condenser down to a height which is lower than the reflux drum and then up again so that it discharges into the top of the reflux drum. As long as the height of the discharge is above the bottom of the seal it cannot syphon the seal dry. This arrangement may lend itself to using a single nozzle on the reflux drum better than the other methods I spoke of before.

Regarding the question of vapor back flow: It would be a small flow, and it would be for exactly the reasons you have given for the mismatch in flow to your two condensers, i.e. the geometry of the piping and condensers.

#19 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 21 June 2010 - 10:55 AM

Katmar,

Original post is from me (Mahesh). I changed my username though.

I have additional spare nozzle on reflux drum and I too incline with having two seal legs for condensate drains and connecting to different nozzles. As per the sketch I provided earlier, we are fabricating piping with 3' seal height (distance between exchanger drain nozzle and top of reflux drum is 10') . I will install a butterfly valve on vapor line to smaller condenser(for tweaking flow rates if required).

We will install this in our August turnaround( I am also revamping acetone - water column for high rates with superfrac trays and modifying feed and reflux nozzle for better flow distribution and efficiency). Keep you guys updated on the outcome.

Thank You
Mahesh

#20 pawan

pawan

    Gold Member

  • ChE Plus Subscriber
  • 162 posts

Posted 22 June 2010 - 12:19 AM

Till Now we have a very good discussion & analysis of this problem. The suggestions from Art & Katmar are definitely simple & fully workable but there is a difference in my logic & analysis from process engineering point of view rather than mechanically seeing it in terms of implementation & possibilities.

All the options mentioned by Mahesh are workable in theory as well as in practice. The difference is best solution. Let us go one by one; we all agree that original option-1 is not considered a good one by all. So it’s a debate between option-2 & option-3.

1. Layout
Now in the suggested option-2, though it is simple, one thing is directly emerging from the discussion of Katmar & Narendra that it is layout dependent as I mentioned earlier. So engineering should take care of this part otherwise it may or may not work properly. (Experience plays a role). This can also be examined form the sketch attached by Mahesh (Post #14).

In the sketch, the vapor lines are not symmetrical at the inlet of both condensers. Now the pulling forces acting are the evacuation in the lines due to condensation inside the exchanger which is pulling the vapor load & therefore, extra pressure drop required towards smaller condenser will result in undesirable distribution of vapors. Since vapor line from first condenser to the reflux drum is easily accessible to vacuum pump & is having larger driving force it will result into losses.

These are practical issues which are actually seen & measured in process plants. In one of such cases we could reduce the losses by 1% during audit.

Conclusion – Possible but layout dependent, Need higher engineering skills & analysis of actual geometrical configuration.

2. Control

Control becomes more problematic when the column is operating at significantly lower capacity than design. In this case, the distribution becomes much more uneven and may result in rapid fouling of smaller one than expected as there is no control. Katmar & Art both are right that even in this situation we do not need any control if they are symmetric.

In fact, the butterfly valve for “tweaking flow rate” planned by Mahesh may be more detrimental than a system without any valve in the vapor inlet.

Conclusion – Control philosophy again requires high level of engineering with all possible scenarios to be factored in.

3. Equalization lines

The equalization line shown in the sketch (Post #14) will result into the losses of vapor to vacuum pump especially through main condenser. It is right that vacuum pump is meant for removing only non condensable but when access is available it can pull any vapor / gas.

The problem with the main condenser is that it has very short travel distance between inlet of vapors & outlet of equalization line. So if we do some more rigorous analysis of the exchanger some part of the vapors will easily escape to vacuum pump through equalization line via drum. This is due to difference in residence time distribution of different molecules inside the exchanger.

Whatever robust be the design of any exchanger, these practical differences are going to occur unless we go into more depth of transport phenomenon happening inside any process / equipment.

Just For example & to explain my point of view, there is no change in the design philosophy of a 1:2 pass exchanger if you reverse the inlet & outlet of tube passes keeping shell inlet at fixed location except correction in “F” factor which does not account for change in approach to equilibrium. Whereas I have improved performance of many oil coolers of gas compressors (where it is most likely incorrect I do not know why?) by at least 4 C just by reversing the inlet & outlet connections of cooling water. Please note that I could not prove it by theory – It was only by logic and it worked every time.

These losses may be more significant or may be seen best when you simply increase the vacuum pump flow (hypothetically possible) or your column is running at lower capacity, the losses of acetone will increase. Mahesh can confirm this practically by considering a mass balance around the distillation setup at two different pressures.

Overall impact – The design do not have any limitation, it is possible but process wise practical limitations are the issues.

I am looking forward for valuable suggestions / comments from Art, Katmar & others on my thoughts.

#21 katmar

katmar

    Gold Member

  • ChE Plus Subscriber
  • 511 posts

Posted 22 June 2010 - 06:24 AM

Pawan, you are absolutely correct that even the best process design can be made useless by bad mechanical implementation. This is true of Options 2 & 3. However, given the ability of Mahesh and Narendra to properly engineer the mechanical implementation I still think Option 2 (parallel condensers) is the better one.

The two condensers are different sizes, so there is no way that they can be made truly symmetrical. But if the common piping, and the piping to each individual condenser, is properly sized the Option 2 will be stable under self control - even with varying loads.

If you put the two condensers in series, and you still want the column to operate at the original -12.5"Hg then you have to deduct the extra 0.35 psi (0.71"Hg) pressure drop of the second condenser from the pressure at the vacuum pump and now you need to reach -16.5"Hg at the pump. This may require a new pump.

The vent on Mahesh's main condenser (sketch from post #14) is a bit unusual in that it is in the middle of the body, but he has the vapour inlets at the two ends so in effect his main condenser consists of two condensers in one body, and the vent is a common vent at the end of each of the two condensers.

Nobody has mentioned any tendency to fouling. As far as I know acetone is a non-fouling application.

One of the joys of engineering is that there are always multiple workable solutions to any particular problem. I have no doubt that a properly engineered version of Option 3 could be made to work, but I still prefer Option 2 for ease of design and control. If you do use Option 3 then the separate drain seals are no longer just "nice to have" - they become absolutely essential because the drains are at different pressures.

#22 pawan

pawan

    Gold Member

  • ChE Plus Subscriber
  • 162 posts

Posted 22 June 2010 - 09:50 PM

Agreed Katmar
Just tried to put some more process flavor into this discussion.

#23 narendrasony

narendrasony

    Gold Member

  • Members
  • 87 posts

Posted 23 June 2010 - 02:01 PM

Attached File  Secondary_Vacuum_Condenser_Rev3.xls   37KB   65 downloadsDear Katmar/ Mahesh,
I apologize for jumping in between and creating unnecessary confusion. This is one such problem which every process engineer has to encounter some times, so couldn't resist. The condensor I referred in my first post was the original condensor of Mahesh but I did not mention it clearly. We also came across similar situation few years before, but the capacity increment was ~50% and finally to make things simple, old condensor was replaced with a new one with 50% added capacity.
I've attached the sketch as per your last suggestion. Anti-siphon vents are marked as removed.
Mahesh, I hope you have already considred the option of increasing the cooling water flow by changing CW line and nozzle sizes.

Regards

Edited by narendrasony, 23 June 2010 - 02:02 PM.


#24 Mahesh@A&M

Mahesh@A&M

    Gold Member

  • Members
  • 78 posts

Posted 24 June 2010 - 03:04 PM

Thanks Guys!!

Pawan,

1. Layout

Layout is not symmetric and even if it is symmetric less pressure drop in smaller condenser will distribute uneven flows. This is the reason why I will have a ball valve ( pressure drop through globe valve is 0.6 psi which can effect vacuum pump performance)to adjust for optimum performance. This will also help us to take smaller condenser out of service at low capacity runs.

Katmar & Art did not mention about piping symmetry in their previous post. " Condenser will create vacuum and suck according to their heat transfer capability".

I don't understand why equalization lines will lead to more vapor losses. Main condenser is BJ21M type shell. Lets say if 80% percent of vapor is condensed by main condenser and 20% by small one. If I increase vapor rate, main condenser pressure drop will increase and flow diverts to less pressure side (small condenser which still has additional room to condense).

Narendra,

Limiting factor is shell side heat transfer coefficient and I don't have chilled water capacity to switch from cooling water to chilled water and it is not economical.

Will the liquid flow through anti syphon vent line since its height below the level of condenser drain nozzle? Please comment.

I will post my sketch with dimensions in couple of days. Thanks again for putting up such a great discussion.

Thank You

Edited by Mahesh@A&M, 24 June 2010 - 07:44 PM.


#25 pawan

pawan

    Gold Member

  • ChE Plus Subscriber
  • 162 posts

Posted 25 June 2010 - 04:44 AM

Mahesh

1. Layout

Layout is not symmetric & this is a problem area which will be further problematic if you put a valve in the feed line to secondary condenser. Becasue your feed is nearer to first one so you need more pressure to push vapors to secondary but in actual phenomenon it happens thru condensation as explained by Katmar. (There are two things in life always - Similarly flow happens in two ways - one is by push - other is by pull) Here in this case, you can not push the flow, so it has to happen thru pull. This is what Katmar mentioned in his previous mail.

So for the desired flow rates of 80 - 20 you need to create lower pressure in secondary than primary such that flow becomes 80-20, but this is not possible in this system as more pressure drop exist in the system on secondary side. So there is a mismatch in pull & system requirement. (This is similar to pump curve & system curve - Here pump curve is pull by condensation capacity - system curve is drop in upstream of secondary) So similar to pumps the final equilibrium will be somewhere around say 90 -10 (this is an arbitrary no - it will depend on actual capability of primary) This is one scenario. Now if you go further deep in the analysis following is another possibility.

If let us say you do not have capacity of primary as more than 80% by any chance then your system will raise the upstream pressure slightly such that the system equalizes as explained. In order to maintain the same vacuum you need to pull more vapors thru vacuum system. This is one of the problem & reason of losses.

If you need to analyse it more properly & in a more visible manner - just think of an imaginary situation - if your secondary is kept 100 meters away from first what will happen - then you will be able to see the minor difference on large scale.

Regarding losses due to equalization line.

The actual phenomenon is that condenser is filled with vapors now it doesn't get condensed all at once - It requires sometime to condense each inlet molecule one by one. Meanwhile if your pump is pulling, there are again two forces on each inlet molecule - One is suction created by condensation & second is a physical pull from pump. If there is slight (slight means slight at molecular level) time gap the molecule will be picked up by pump not by condenser so resulting into losses. This is how equalization line is not my favorite.

Your concept if vapors are more 80% will be condensed in first & rest is diverted to secondary is not - (I am sorry to mention this, but need to do so to specify the concept & fundamentals along with their gravity) - based on your wish. Process do not follow our wish. Why do you think it will go to secondary where the Pull force is less than Pull force by vacuum pump.

Also remember that the normal flow to vacuum pump is thru reflux drum vapor space & thru equalization lines from your column not thru seals. So that is the normal path for gases to travel to vacuum pump & hence any vapor along with it can easily be escaping rather than going thru condensation route.

In fact, in my personal opinion, equalization lines are not required. Let everything go thru condenser to reflux drum & drum should be connected to vacuum pump. For occasional removal of non-condensibles from the exchanger you can have small vent which will be there in any case. So all the vapors have to cross condenser all the time. So in case of any disturbance from downstream, the impact is reduced while in your case you may loose half of the material by the time you notice any change.

Will the liquid flow through anti syphon vent line since its height below the level of condenser drain nozzle? Please comment. - No if syphon lines are properly sized, Yes if they are not.




Similar Topics