20 replies to this topic

[daraj]

Hi,  a general question to experienced  professionals there. Often while doing conceptual  cost estimates for a proposed future design or plant we rely on process simulations(such as aspen or HYSYS) for coming up with a preliminary design and the using the applications to arrive at a preliminary design which is then used as input in costing tools(such as Aspen economic estimator) to arrive at a preliminary estimate.

 

My question is the final pressure in say, a tray column, in a simulation represents operating pressure I believe. Often to estimate costs costing tools ask for parametsr like design pressure of column and such. So Iam wondering how to use simulation information and convert it into a design parameter for cost estimation. not limited to columns. Are there any quick thumb rules?

 

for eample if I have a converged simulation that shows that i need a pressure of 10 bar for a tray column n my process, is that pressure the operating prssure? then what would be the ideal design pressure for that column? because the design pressure is what is needed to arrive at the fabrication and other costs.

 

Likewise for reflux drums, surge vessels i can calculate flow rates from simulation. But what should be the design volume, if I assume a residence time or holdup time of say 5 misn or 10 mins? how much empty space needs to  be addedto arrive at final design volume?

these are the type of questions i have.

 

The design parameters for temperature, pressure etc. that you seen in an actual process flow diagram for an exisitjng plant, how far are  they from the simulation values? That is, if I were trying to simulate an exiisting plant i would be using operating pressure/temp for those columns and not what is listed in the PFD right?

 

Please help

[xavio]

hi daraj,

 

This link should help you:

http://people.clarks...ign/heurist.pdf

 

Skip to last three pages to answer your particular questions.

Bear in mind that this is just a guideline, not law.

 

Good luck!

 

xavio

[daraj]

xavio, thanks. Sometimes these heuristics seem to vary based on source. is there any other authentic engineering source for arriving at design paramteres? There has to be some engineering calculations involved to arrive at the final design pressure or volume, right? It has to be based on science at the end of the day, cannot be arbitrary. So I would like to know  abit about the knowledge behind these heuristics and also look for an authentic source.(preferably onlien accessible)  

[daraj]

also what Iam more interested is in, is to know the approach from simulation to design, rather than directly getting suggestions for design paramters. I keep hearing operating pressure/temp, then maximum operating pressure /temp and then design pressure/temp. The differences between all these three and how to get the oher two when the first is known (iam assuming operating press/temp is what I get out of simulation)

[paulhorth]

Daraj,

I believe you should consider Hysys as a heat and material balance tool ONLY. It should not be used to design or size process equipment - you have already explained some reasons for this. That is something you have to do yourself, as a process engineer, using well established design and sizing techniques which are readily available in the literature, both in the public domain and proprietary (e.g. Shell). As for cost estimation - well, Hysys won't tell you the materials of construction, which has a major effect on costs.

To develop design pressures, I recommend you proceed thus:

• Draw a flowsheet of your system. You can print out a Hysys flowsheet but it's better to draw one with pencil and paper so you know where all the streams go, and why.
• Mark on  the flowsheet the operating pressures and temperatures from Hysys
• Then stop and think - where are the natural breaks in system pressure? Equipment should be grouped together for design pressure. Mark the design pressure breaks on the diagram. Design pressure is often taken as 10% above the highest operating pressure in a system, or say 1.5 bar above, whichever is greater, with a minimum of 3.5 barg (50 psig) but companies will each have their own standard guidelines.
• Remember that for each connection between different design pressure systems you will have an overpressure relief case. It's good design to try to eliminate relief cases where practical.

And this is merely the beginning of the process design journey.

 

Paul

[marchem]

the document suggested by xavio is a good starting point

but many books and manuals discuss these topics,

for example for gas processing GPSA is a good resource,

but there are really many (google can help)

also a few simulators do include utilities for design optimization,

 

http://en.wikipedia....cess_simulators

 

these may be useful for preliminary design

[daraj]

Hi Paul, thanks. Lets assuem my deasign pressure = 110% max operating pressure and I know the material as well. The question is to how to arrive at maximum operating pressure for a column guiven the operating pressure from aspen plus? Iam doing only a conceptual design/estimation exercise here, not a detailed engineering/design.So any acceptable(but based on solid grounding) rules of thumb are OK but I want to understand the science behind hese rules also so that i dont blindly apply them to arrive at the design pressure

[paulhorth]

Daraj,

 

There is no science, it's engineering judgment.

For your column, you should consider how you intend to control the pressure. What can cause the pressure to deviate from "normal" and how fast can that happen? How would your control system respond? Factors to consider include:

• Does the column have a partial or total condenser?
• How much (and why) can the condensing utility vary in temperature or flow?
• If there is a vapour outlet, where does it flow to?
• Does the feed vary in composition?
• Is the source of feed at a significantly higher pressure?
• etc............

You then decide on a maximum operating pressure, at or above which you would set a high pressure shutdown (PSHH), and the design pressure would then be a further margin above that, as i have already explained. Your margins should be sufficient to avoid shutdowns or relief valves opening because of operating fluctuations. That is for you to judge, guided by past practice and company standards.

 

Paul

[Bobby Strain]

I wouldn't worry much about the design values. You won't get even close to a reasonable cost estimate from where you are starting.

 

Bobby

[xavio]

hi daraj,

 

I can't agree more with paulhorth, it is all about engineering judgment, why do you insist that there must be scientific base?

Heurisctics, guidelines, and design criteria are all based on "good" engineering judgment which comes from accumulated "bad" experiences from the past.

Some of the recommendations are based on safety consideration, some others are based on economic consideration, still some others are simply based on the need to standardize design procedure.

 

I think you should spend some time reading design handbooks ('authentic source'), the reason behind each guideline is usually explained in brief.

 

 

Thanks.

 

xavio

[daraj]

Hio Paul and xavio. If the rules are coming out of condensed experience then I think it shouldnt vary from oen  source to another. I have a problem when one source says keep design pressure 25% above say operating pressure and another source says keep it atleast 10% above and so on. what is the authentic source or two that I can use for most equipment? Is" chemical process equipment  - selection and design" by Penny and Cooper a good book?

 

Bobby, Iam not sure what you mean. for getting a conceptual cost estimae of a plant /plant section(+/- 40% say) a simulation is a good starting point for proviiding design inputs. In fact some of the times if you just input operating pressure/temperature there are tools such as aspen economics/capital cost estimator etc. that automatically calculates the design values for you and gives you a reasonable estimate within the error bars I mentioned. Of course material of construction has to be proper and that can be guessed as well based on the properties of the process fluids and working conditions. Most companies at an early stage in design, do these type of estimates with a preliminary simulation-based design and estimates are within +/- 50% most of the time. But I just did not want to rely on the software blindly and wanted to knmow for my own sake how to arrive at these design conditions. Meaning,getting an insight of an engineer who does FEED is always good to make informed decisions even at the 1st stage-conceptual level. So that you arent too much off with your assumptions. That is my aim. My objectve is not to get a really accurate estimate, that will happen much later at the detailed engineering stage and I dont do that

[daraj]

also one more quedtion I have. I have often found textbooks or other sources caution that while the assumption of liquid and vapor being in equilibrium is highly idealized and does not happen in industrial cases(which is understandable because of mass transfer issues) they say that itis however reasonable to assume that vapor and liquid leaving a tray are at same temperature, based on industrial columns. I dont understand this. If equilibrium in mass transfer is not achieved in a single tray how is it OK to assume that temperature is same for liquid and vapor leaving the same tray? are'nt these two interrelated? if  a tray shows departure from equlibrium then temp. of vapor and liquid are not same. Are they implying that this assumption of same temperature to simplify calculations does not affect overall results as much, mathematically? Pl explain

[marchem]

consider the many different processes, design codes, safety codes etc. etc.

there are not simple rules to define operating and design values as these depend from many factors,

a simulator is just a tool which allows you to estimate mass & energy balance but this is just the beginning...

for mass transfer consider a good book as for example Separation Process Principles by Seader/Henley/Roper,

nowadays most simulators do include these models, for a free software see

http://www.chemsep.com

Edited by marchem, 29 April 2014 - 05:11 AM.

[xavio]

daraj,

 

It is normal that one guideline differs from another, it is why we call it guideline and not law.

Guidelines are written by people coming from various background (industries), hence with different experiences and viewpoints.

That's why big engineering companies usually issue design criteria or manuals, they wish to "standardize" the guidelines to avoid confusion.

Even so, design manuals are still not law, they may differ from project to project, place to place, time to time.

Which one is the most "authentic", judge it by yourself.

Actually you can also define your own "guidelines", use them as you see fit.

 

chemical process equipment  - selection and design" by Penny and Cooper a good book? --> it is a good book

The authors are Couper/Penney/Fair/Walas, try to respect these great gurus by writing their names correctly.

Another book has been suggested by marchem, still another book is Applied Process Design by Ludwig.

I believe there are still many many other good books.

 

 

also one more quedtion I have. I have often found textbooks or other sources caution that while the assumption of liquid and vapor being in equilibrium is highly idealized and does not happen in industrial cases(which is understandable because of mass transfer issues) they say that itis however reasonable to assume that vapor and liquid leaving a tray are at same temperature, based on industrial columns. I dont understand this. If equilibrium in mass transfer is not achieved in a single tray how is it OK to assume that temperature is same for liquid and vapor leaving the same tray? are'nt these two interrelated? if  a tray shows departure from equlibrium then temp. of vapor and liquid are not same. Are they implying that this assumption of same temperature to simplify calculations does not affect overall results as much, mathematically? Pl explain

 

I suggest you to look closer at the ideal equilibrium stage equation.

Compare it to equilibrium equation when stage efficiency is introduced (e.g., Murphree).

The calculated stage temperature is the same regardless the composition is in equilibrium or not.

In reality, vapor and liquid temp at one stage may differ slightly due to continuous mass transfer taking place, instead of discrete equilibrium point as assumed.

 

Good luck.

 

xavio

[daraj]

"

I suggest you to look closer at the ideal equilibrium stage equation.

Compare it to equilibrium equation when stage efficiency is introduced (e.g., Murphree).

The calculated stage temperature is the same regardless the composition is in equilibrium or not."

 

Hi xavio and marchem thanks.

Xavio,Iam trying to understand this at a physical level, as to why the temperatures should be the same when they are not in equlibrium.

Mathematically, temperature might not vary all that much. But Iam trying to look at it from a physical/thermodynamic point of view

 

also physically what is meant by stage temperature? Meaning, a stage at given point in time, has both liquid and vapor phases. So if we were to hypothetically measure stage temperature ourselves in a column, where would we stick the thermometer? irrespective of whereever the thermomneteris in a tray will it show the same reading?

[daraj]

In fact i have same question with "reboiler stage  temperature" as well. Most simulation software report the temperature of the last stage(which is usually the reboiler" and return a value for stage temperature after the calculations for column are converged. My question, how ti interpret this reboiler temperature? where is this temperature measured? Physically, the process fluid enyers and is continuously heated and partially vaporizes and leaves the reboiler. so there is a temperature profile across the tube (or shell). so is this reboiler stage temperature something like an averaged out temperature? or is it the entry/exit temperature?

[marchem]

you'll find the answers in books discussing modeling of distillation columns,

for example Separation Process Principles (see previous post)

[xavio]

daraj,

 

We usually stick thermometer to the liquid phase of a stage, then we assume that vapor phase temperature is the same.

As I've told you, in reality vapor and liquid temp at one stage may differ slightly due to continuous mass transfer taking place.

Temperature on the same tray will also differ locally due to imperfect mixing or maldistribution.

But, will you care much about that difference?

 

It seems that you have so much curiousity about distillation system and are eager to learn.

Follow marchem's advice and you shall find.

In addition, distillation books by Henry Kister should be in you reading list too.

 

Good luck.

 

xavio

[daraj]

marchem and xavio, thanks for your inputs and suggestions

[Padmakar Katre]

Hello,

Economic feasibility (+/- cost estimates) and basic engineering is mainly done by expertise such as licensors/detail engg contractors. They have the design database for these units; what mainly changes from one to another project is the intended unit capacity, product specs, feed quality, utility specifications and metrological data. These inputs are enough to arrive at +/- 30% cost estimates. Operating conditions of unit are nearly similar but may change based on feed & product specifications and other constraints. They have the simulation models to generate HMB and broad equipment specifications such as column diameter/height/material/tray count or exchanger type/area/material or pump capacity/model etc. Based on this information cost estimates can be done as day in-day out they interact with vendors and have the latest information on pricing and all. These organizations have there offices across the globe and have required vendor database w.r.t. capabilities.

Certain guidelines differ slightly among different companies but direction remains same. This is mainly based on the past experience, future capacity increase if any and lots of such factors during conceptualization stage.

Hope it helps.

[daraj]

Thanks Padmakar