Continue to Site

Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

  • Congratulations cowski on being selected by the Eng-Tips community for having the most helpful posts in the forums last week. Way to Go!

Calculation for steam tracing requirement 1

Status
Not open for further replies.

ddkm

Chemical
Nov 9, 2005
94
We intend to install steam tracing on a pump for the purposes of maintaining a certain temperature, which enables the fluid in the pump to remain in liquid form.

What are the considerations for doing the heat transfer calculations for the required steam tracing?

My view is as follows:
- Volume of the fluid sitting in the pump when not in operation. This can be estimated from the pump drawings
- Surface area on which the tracing is being lined. Assuming this is on one face of the pump impeller casing area, the area can be estimated as the circular cross section (on the pump side) and tubular area (on the steam tracing side)
- Heat transfer coefficients for the pump material and for the steam tracing material (eg: copper)
- Ignore fouling or deposits inside the pump and inside the copper tubes

What else?

---engineering your life---
 
Replies continue below

Recommended for you

ddkm:

I seriously doubt whether your pump has so big a surface area that you have the option to consider “optimizing” the amount of available surface area to steam trace.

On the contrary I believe that, like in most similar situations with the same type of scope, you will probably not have enough surface area as you would like to have. This is what happens in real-life engineering and your challenge is not to calculate the amount of steam tracing required, but rather to employ your ingenuity in getting the physical tubing, heat transfer cement, tubing joints, condensate trap(s), and condensate drains installed with as much effective heat transfer as you can manage under the circumstances.

The factors you mention have little or nothing to do with effectively maintaining the pump casing hot when you consider the practical aspects of what you are proposing. The amount of liquid in the casing is static so you have an unsteady state heat transfer case. The time that it takes to heat the liquid may not be important as just keeping it hot. The remaining other factors (casing face surface area, heat transfer coefficients) should also be of little concern for what you are trying to do. You are probably looking at a surface area of approximately 1.0 ft2 or less. Am I right? Besides, you haven’t even considered the most important item in steam tracing: the heat transfer cement. The successful application of the cement is the real secret in expecting to transfer any heat via conduction (which is the heart of the steam tracing process). Your required steam trap will probably be ½” size – and oversized at that.

I strongly believe you should steer away from the theoretical heat transfer calculations you are proposing and simply get down to the practicality of how you can maximize the effective heat transfer through the pump casing while still allowing for successful operation and maintenance of the unit. That, in my opinion is your strongest challenge. Trying to bend and double-back a 3/8” or ½” stainless steel tubing (complete with Swagelock disconnect unions) while keeping the entire flexible assembly in positive, surface contact with the casing is a tough enough job without imposing a theoretically derived heat transfer “answer” to the physical application. The guys in the maintenance shop will be watching your every move and design. If you don’t give them the flexibility and the maintenance capability they need and deserve, you will return after one weekend to find all your heat tracing tubing, steam trap, and insulation all ripped out and in a near-by dumpster. That, I can almost guarantee you.

Please don't take this advice as pessimistic; after 46 years I can assure you it is realistic.

 
To add to Montemayor comments:
If you need to keep the liquid warm, the pump should have his internal steam channels or steam jacket. This you will find probably on positive displacement (screw types) pumps.
On centrifugal pumps, just steam trace and insulate the suction and discharge lines. Spirax-Sarco and Armstrong have a wealth of information on steam tracing.
If the maintenance guys have to change out the pump or do PM and your construction is in the way...
 
Montemayor,

Thanks as always for your valuable input. I couldn't put a finger on what my problem with this whole "calculation" thing was, until your comments above.

For the surface area available to the pump casing, I've made a rough estimate, which comes out to about 0.85ft2, so you're quite close.

I will focus on your suggestion to:
maximize the effective heat transfer through the pump casing while still allowing for successful operation and maintenance of the unit.

Having said that, maybe you could elaborate more on the following:

- you mentioned about heat transfer cement, which I'm not familiar with. We don't use this at all in our applications; is this critical for all steam tracing installations? What function does it serve (basic principle)?

- would insulating the casing be necessary to ensure that the heating is contained?

- would it be necessary to run double-tracings on the pump casing, i.e. two layers, but a single flow (and single trap)



---engineering your life---
 
svanels, it's a centrifugal pump. You're saying that suction lines and discharge lines on the pump are more important to trace? I will consider those in addition to the pump casing itself.

Also, reference to what Montemayor said about the heat transfer cement, what is your view on maintenance issues? i.e. how will the maintenance guys deal with the tracing and cement, everytime they have to open up the pump? I imagine that the tracing and cement need not be destroyed, as long as the installation was "perfect" and not interfering with the key sections (example: flanges, bolt/nuts areas, etc). Right?

---engineering your life---
 
In steam tracing you attach a small pipe 1/2" or 3/4" to a larger diameter. The cement is necessary to isolate the tracer and line from the environment and minimize energy losses and more important to prevent someone from getting burned. The cement can come in premolded sleeves or you can use isolation whool reinforced with wire mesh. The whole assembly is covered with insulation (aluminium) sheets to protect the insulation from the environment (rain, erosion, dust etc..).
 
svanels:
Thanks for following up and contributing to this thread. Your Thermon reference would have been exactly what I would have posted. I have been on an extended week-end, 4th of July family reunion and had distanced myself from the forums.

ddkm:
The main function of the Thermon cement is to furnish a postitive conduction medium for the expected heat flow. Without it, you are at the mercy of the air space around the steam tracing tubing - which has a heat transfer coefficient next to nothing. Without the cement you can expect only token heat tracing. The Thermon company is the place to go to for advice, free software, and all the pointers and supplies you will require for steam tracing. They are the acknowledged experts in the field - both steam and electric. I think you can appreciate the detailed, isodrawing of the pump tracing furnished in the bulletin.

Good Luck!
 
Actually I helped myself also, I was searching on google with steam tracer installation and encountered Thermon, definitely worth a second look, since I am a maintenance professional in an oil refinery. [shadeshappy]
 
Status
Not open for further replies.

Part and Inventory Search

Sponsor