Tek-Tips is the largest IT community on the Internet today!
Members share and learn making Tek-Tips Forums the best source of peer-reviewed technical information on the Internet!
-
Congratulations MintJulep on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!
Articles for Hife-Pipe Jacket Heat Transfer 1
- Thread starter RJB32482
- Start date
- Thread starter
- #2
RJB, I do not have any articles about what you are asking about, but the major differences between the three types of jackets are 1) Surface Area and 2) Turbulence within the jacket and 3) the amount of "dead flow" within each of the three types of systems.
While Dimple and conventional (which I believe you mean tank inside of tank construction), you have maximized your heating surface area. You have better turbulence with dimple obviously and less of a chance of dead flow.
With Half-Pipe jackets, you have no chance of dead flow areas, but you limit your heating surface area because you need space between the welds. This is a very crude way to jacket a tank and I would not recommend it to anyone.
So, of the three, Dimple jacketing is the best method in my opinion.
If you are looking for another type of jacket for a tank, I would suggest looking into ControTrace. It is an external heating jacket that has a lot of engineering behind it.
Hope this helps.
While Dimple and conventional (which I believe you mean tank inside of tank construction), you have maximized your heating surface area. You have better turbulence with dimple obviously and less of a chance of dead flow.
With Half-Pipe jackets, you have no chance of dead flow areas, but you limit your heating surface area because you need space between the welds. This is a very crude way to jacket a tank and I would not recommend it to anyone.
So, of the three, Dimple jacketing is the best method in my opinion.
If you are looking for another type of jacket for a tank, I would suggest looking into ControTrace. It is an external heating jacket that has a lot of engineering behind it.
Hope this helps.
Chemical Processing ( did have an article that you could download but I don't see it there now.
Some Lessons in Reactor Design
Batch reactions need proper heat transfer, and that takes good reactors
By Eugene J. Sak, P.E. Pfaudler Inc., Rochester, N.Y.
January 01, 2000
Also, take a look at this thread...
thread391-71471
Some Lessons in Reactor Design
Batch reactions need proper heat transfer, and that takes good reactors
By Eugene J. Sak, P.E. Pfaudler Inc., Rochester, N.Y.
January 01, 2000
Also, take a look at this thread...
thread391-71471
The thread referenced by EGT01 will give you all you need to know about the heat transfer coefficients, but I would like to comment on a practical point.
A very important difference between the different types of jacketing methods (but which was not listed by perdog) is that they have very different abilities to withstand pressure external to the shell. A half pipe jacket will take a relatively high pressure, whereas an anular jacket will require a very thick shell to withstand the external pressure because of the larger unsupported areas.
I recently saw a tank destroyed because the operator left the jacket's cooling water return valve closed and just the pressure of the cooling water buckled the shell beyond repair. You can also have problems if the jacket is isolated and then a hot liquid is added to the tank. This can cause expansion or phase change in the jacket, leading to massive pressures. Always install relief valves if this is a risk.
I have used half pipe (and even angle irons) as external coils quite successfully, but with low heat transfer coefficients and high pressure drops. But they can be broken down into several parallel circuits.
I have also used dimpled jackets on low pressure refigerated water duties where we were simply maintaining the temperature (no actual cooling). The dimples aid turbulence, and the welds in each dimple give extra stability to the shell for the external pressure.
A very important difference between the different types of jacketing methods (but which was not listed by perdog) is that they have very different abilities to withstand pressure external to the shell. A half pipe jacket will take a relatively high pressure, whereas an anular jacket will require a very thick shell to withstand the external pressure because of the larger unsupported areas.
I recently saw a tank destroyed because the operator left the jacket's cooling water return valve closed and just the pressure of the cooling water buckled the shell beyond repair. You can also have problems if the jacket is isolated and then a hot liquid is added to the tank. This can cause expansion or phase change in the jacket, leading to massive pressures. Always install relief valves if this is a risk.
I have used half pipe (and even angle irons) as external coils quite successfully, but with low heat transfer coefficients and high pressure drops. But they can be broken down into several parallel circuits.
I have also used dimpled jackets on low pressure refigerated water duties where we were simply maintaining the temperature (no actual cooling). The dimples aid turbulence, and the welds in each dimple give extra stability to the shell for the external pressure.
-
1
- #6
I would like to offer some support for the half-pipe jacket from the process point of view when cooling water from once through cooling water or non-tempered cooling tower loops are used.
Dimple jackets are prone to plug over the years because I have yet to see anyone diligently maintain their cooling water quality at an acceptable level year after year after year.
Likewise the traditionally poor flow distribution in the conventional jacket gives uneven heat transfer from day one and in the long run tends to collect solids in the stagnant areas. Again, poor cooling water quality does it in in the long run.
Because the flow path of the half-pipe jacket is defined by the half-pipe and vessel boundaries, it is inherently better in the long run, especially when maintaining cooling water quality takes a back seat to other pressing issues.
I have personally thrown away about a dozen dimple jacket reactors and one or two conventional jacket reactors due to poor heat transfer after about 20 years service. I have never thrown away a half-pipe jacket reactor even after longer periods of time. Yes, they are tougher to fabricate. A full penetration weld is a MUST on the jacket to vessel welds. I only accept proof of a fabricators method only after a destructive examination of a mock-up using precisely the same welders, materials and procedures.
Yes, the gap between half-pipes is not wetted on the cooling water side, but it is not totally ineffective. Through conduction and the fin effect of the two half-pipes beside the gap, I have found it to be 60-80% as effective as the area wetted on both sides.
Just my $0.02 worth.
Good luck,
Latexman
Dimple jackets are prone to plug over the years because I have yet to see anyone diligently maintain their cooling water quality at an acceptable level year after year after year.
Likewise the traditionally poor flow distribution in the conventional jacket gives uneven heat transfer from day one and in the long run tends to collect solids in the stagnant areas. Again, poor cooling water quality does it in in the long run.
Because the flow path of the half-pipe jacket is defined by the half-pipe and vessel boundaries, it is inherently better in the long run, especially when maintaining cooling water quality takes a back seat to other pressing issues.
I have personally thrown away about a dozen dimple jacket reactors and one or two conventional jacket reactors due to poor heat transfer after about 20 years service. I have never thrown away a half-pipe jacket reactor even after longer periods of time. Yes, they are tougher to fabricate. A full penetration weld is a MUST on the jacket to vessel welds. I only accept proof of a fabricators method only after a destructive examination of a mock-up using precisely the same welders, materials and procedures.
Yes, the gap between half-pipes is not wetted on the cooling water side, but it is not totally ineffective. Through conduction and the fin effect of the two half-pipes beside the gap, I have found it to be 60-80% as effective as the area wetted on both sides.
Just my $0.02 worth.
Good luck,
Latexman
mechprocess
Mechanical
I have read some threads that underestimate the jacket side heat transfer coefficient for half pipe
I am designing a jacket using 3 inch 180° half pipe.
I calculated a jacket film coefficient of 639 btu/hr-ft^2-°F for the following:
Cooling medium: 25% Propylene Glycol,
Tank: 19 ft ID
Half Pipe Length: 567 ft
On center: 7 inches
40% straight side
45 °F inlet temp.
Flow rate: 79 gpm,
ReDH: 28,410
DP 30 psi
I used the Sieder and Tate equation.
I downloaded the program DeltT and confirmed the results. However, the effective surface area calculated by the program was almost double what I calculated since the program takes credit for the area between coils. I only took credit for the contact area of the half pipe.
I use Kerns Equations that are published in Perry’s (11-18) latest edition to calculate the tank temperature vs time. I say this because some have posted measured “h” values for half pipe in the 70 range. This is much too low. I would say that a measured value is only good if the tank temperature has reached steady state and this number is based on the cooling water inlet and outlet temperatures. I don’t see how you can measure any overall U value during temperature transients unless you do what I did one time and that is extrapolate the solution of the transient temperature equation assuming you know the effective surface area and neglect heat loss other than from the jacket.
I am designing a jacket using 3 inch 180° half pipe.
I calculated a jacket film coefficient of 639 btu/hr-ft^2-°F for the following:
Cooling medium: 25% Propylene Glycol,
Tank: 19 ft ID
Half Pipe Length: 567 ft
On center: 7 inches
40% straight side
45 °F inlet temp.
Flow rate: 79 gpm,
ReDH: 28,410
DP 30 psi
I used the Sieder and Tate equation.
I downloaded the program DeltT and confirmed the results. However, the effective surface area calculated by the program was almost double what I calculated since the program takes credit for the area between coils. I only took credit for the contact area of the half pipe.
I use Kerns Equations that are published in Perry’s (11-18) latest edition to calculate the tank temperature vs time. I say this because some have posted measured “h” values for half pipe in the 70 range. This is much too low. I would say that a measured value is only good if the tank temperature has reached steady state and this number is based on the cooling water inlet and outlet temperatures. I don’t see how you can measure any overall U value during temperature transients unless you do what I did one time and that is extrapolate the solution of the transient temperature equation assuming you know the effective surface area and neglect heat loss other than from the jacket.
- Status
Similar threads
- Question
- Locked
- Question
- Locked
- Question
