9 replies to this topic

[John885]

Guys,

 

I'm working on a gas pressure reduction re-configuration plant. The client has an outlet gas temperature problem and has asked for a solution. What I've done so far, I've simulated JT Valve in HYSYS to see what the outlet temperature would be based on the gas composition.

There are 2 heaters which heat gas upto 52 degree C before entering the regulator. Gas temperature reduces about 35 degree C after regulating and should be around 17 but in the worst case it is -4 deg C.

 

The distance between heaters outlet and regulator is about 40m, and the pipe has no insulation on it. I think the problem is because of heat loss along the pipe length to regulator.

I tried to simulate the heat loss in HYSYS pipe segment but couldn't do that because I haven't done it before. The data which I have are:

Gas Temp.: 52 deg C

Pipe Lehgth: 40m

Ambient Temp.: 3deg C

Gas Pressure: 10200 kPa

Gas flow rate: 5295 scmh

Gas density: 0.75 kg/m3

Gas compressibility: 0.9980

Gas viscosity: 10.8 x 10^(-6)

Cp= 2.075 kJ/kg

Pipe material = mild steel

Pipe size: 8"  OD=8.63" (219.1mm)    ID=7.98" (202.7mm)       Sch 40, wall thickness: 0.32" (8.18mm)

 

Could anyboy help telling me how to calculate the gas temperature at the end of 40m pipe (pipe outlet)?

 

Or telling how to simulate in HYSYS.

 

Thank you guys.

Edited by John885, 22 September 2014 - 10:56 PM.

[Ajay S. Satpute]

John,

 

Heat loss (along the pipe length) calculations using HYSYS: 

 

1. Select a pipe segment.

2. Provide inlet, outlet material stream and energy stream

3. Provide inlet stream T, P, flow rate and composition. Inlet stream shall now converge.

4. Double click on pipe segment. Go to Rating/sizing. Provide length, OD, ID.

5. Then go to Heat Transfer in the same tab. Specify by "Estimate HTC". Provide Ambient temp., ambient medium, medium velocity, provide insulation details (if it is relevant). Check following tabs; "Include pipe wall, Include Inner HTC, Include Insulation (if it is relevant) and Include Outer HTC".

6. It should now converge.

 

I'm using Ver. 8.6. But same procedure should be applied for other versions.

 

Let me know if you could solve this.

 

Regards.

 

Ajay S. Satpute

[PingPong]

I am sorry to say, but if one needs a simulator to calculate heat loss over a piece of pipe one should not be in this business.

 

Moreover: what is the point of using Hysys if one has to enter the outer HTC oneself ?

 

A gas flowrate of only 5295 scmh seems very low to me for an 8 inch pipe: that is only about 50 actual m³/h at 102 bar, giving a pipe velocity of less than 0.5 m/s.

I don't believe that small flowrate. Check your numbers.

 

How reliable are the temperature measurements, both at heater outlet and pressure reduction outlet? I suggest you try to measure heater outlet, valve inlet and a valve outlet temperatures in another way (contact thermometer against pipe with a lot of insulation around it, or an IR thermometer, or ....).

Edited by PingPong, 23 September 2014 - 04:39 AM.

[Ajay S. Satpute]

PingPong,

 

My guess is that you have never used Hysys for piping heat loss calculations. If you use it more often, then you would appreciate the same and perhaps you would not ask the question regarding HTC. 

 

I have nothing against hand calculations (and am very good at them) . But if software tools are available, then why not use it?

 

 

 

Regards.

 

Ajay S. Satpute

[latexman]

"Suum cuique" - Cicero

[PingPong]

Ajay, my comments were adressed to John885, not to you.

 

Unless the gas velocity in the pipe is really only 0.5 m/s (which I can't believe), the overall HTC of an uninsulated pipe will be mainly determined by the outside HTC, which depends very much on convection heat loss, and on the radiation heat loss of the pipewall (which depends very much on its emissivity). If it rains the pipe heat loss will obviously be much higher, but I don't think even the Holy Hysys can calculate that.

 

If the flowrate in this 40 meters of 8 inch pipe is really only 5295 scmh, then the heatloss, when it is windy, could have a noticable impact on the temperature.

If however that flowrate is a typing error, and it is really a factor 10 (or 100) higher, then the impact on temperature will be negligable.

 

Moreover: why try to calculate a temperature drop in an existing sytem, if one can simply measure that temperature drop using an IR thermometer ?

 

 

I see many topics on this forum in which inexperienced engineers want to know which button in Hysys they should click to do something they have absolutely no knowledge of. Hysys seems to be the preferred tool for the clueless. Problem with that is that when they get a completely unrealistic outcome (because they made some input mistake) they do not notice that because they have absolutely no clue what a realistic outcome should be. By doing a quick estimate by hand. before doing the simulation, such mistakes could be avoided before ending up with the client.

[John885]

Thanks Ajay for your help. I'm sure those people which talk like PingPong have never used a software to resolve or simulate an issue. I'm totally ignoring such people and I'm sure they don't have enough knowledge to use computer engineering software. If you read his/her comments carefully you would realise that he/she doesn't have any idea of the industrial projects at all. Doesn't realise that there might be conditions in which you don't have access to the site to measure the temperature or other things. 

Data which I gave in my post are read from SCADA system monitoring the plant in another country miles away. I'm sure pingpong doesn't even imagine that it's possible to control a plant remotely.

 

Anyway, I appreciate your advice regarding HYSYS. 

[PingPong]

Through the past decades I have used many different simulators for many process designs, which all work fine.  I probably was already doing simulations when you were not even born.

 

In the mean time you have completely ignored my comments regarding your obviously wrong numbers.  You can imagine what I would do if I were your client.

[Ajay S. Satpute]

John,

 

PingPong is a very respectable member of this forum and he has always tried to help others (including myself) with his expertise. I cannot endorse your comments at all.

 

Nevertheless, please read through his comments, where he emphasizes on hand calculations importance. In my opinion. one should use simulators, if possible, but must know how to solve the sum with hand calculations. Hysys also uses same set of equations, you would use in hand calculations.

 

FYI, Hysys cannot do radiation heat transfer calculations.

 

 

Regards.

 

Ajay S. Satpute

[Art Montemayor]

John885:

 

The topic of this thread is the resulting temperature from a Joule-Thomson gas expansion and I believe PingPong has given you a typical, alternative method of estimating and calculating the answer – as opposed to using a simulation program.  As Latexman has summed it up, “Suum cuique” – or, "to each his own".

 

Not everyone has a simulation program at his disposal – especially Hysys.  But even if they did, that doesn’t make a simulation program the answer to every problem.  Your comment that some members “don't have enough knowledge to use computer engineering software” is neither a rational nor a fair one.  As an old chemical engineer and one who is on the verge of “hanging it up”, I can relate to what PingPong points out and fully endorse what he states: reliance on a simulation program for your answers to a specific problem carries with it the responsibility of knowing what the algorithm used is and whether it is the correct one or not.  All the stupid computer does is crunch numbers.  It can’t think or logic out a problem.  I can’t generate the respect for a computer that I have for the human brain.  One is the product of the other and therefore doesn’t deserve the recognition.

 

Old, experienced, or veteran engineers are not necessarily ignorant of today’s so-called technology.  Hell, some of us were the pioneers in developing the use of computers and simulators and helped develop the programs you have today.  I’m probably older than PingPong but I remember perfectly well taking the first Fortran course in 1958 for the IBM 360 (yes, with punched cards).  Yes, I wrote and understand Fortran and machine language – having used both of them.  I also have written Visual Basic programs and have operated most high level simulation programs and HTRI.  That doesn’t make me smarter or better.  I’ve led process teams that fulfilled my need for process simulation.  I don’t do simulation no more than I do AutoCad design; like all engineering managers or leads, I leave that up to others who don’t have the credentials and experience to lead yet.  My main concern has been the safe and profitable application of the simulation results – not the simulation calculations themselves.  That is why I feel I understand what PingPong has tried to expound in showing another way to look at resolving the problem.

 

I hope our members take into consideration the effort and steps taken to try to help each other in our Forums and don’t assume that some members are ignorant or unaware of current engineering tools being used today.  The tools will never supersede the logic and ingenuity employed by an engineer.  After all, it is the engineer who invents, selects, and uses the tools.  Never the other way around.