Temperature of pipe

[sameg]

Hey everyone. Looking for some help with a calculation I am having trouble wrapping my head around.

I have a steam line that is NPS 36, 22.22mm W.T. pipe with 100mm of mineral wool insulation.

The pipe will be cold prior to start up and we would like to use a hot air compressor to warm up the steel before filling it with steam to prevent massive amounts of condensate dropping out.

I need to find an equation to figure out the temperature of the pipe at any given time, after the dry air has been running through it.

The air compressor's run at 150psi, 300C and 1500scfm

Any help would be greatly appreciated.

 

[prex]

You must figure out the heat transfer coefficient between the hot gas and the pipe wall: any heat transfer book will give you that.
Then, assuming of course a thin wall pipe, your equation is
T_g-T_w=(T_g-T_wi)e^-[α]t
[α]=hA/cM
h=heat transfer coeff.
A=area of pipe exposed to hot gas
c=specific heat of steel per unit mass
M=mass of steel to be heated

prex

http://www.xcalcs.com

: Online engineering calculations

http://www.megamag.it

: Magnetic brakes and launchers for fun rides

http://www.levitans.com

: Air bearing pads

 

[TD2K]

Do you want to be able to estimate the temperature of the pipe at any location along its length AND at any time after you start injecting hot air into it?

Prex has given you the basic equation to estimate heat input into the steel pipe. The weight of the steel pipe and its heat capacity gives you the temperature rise in the pipe. Another factor is the heat transfer from the pipe to the insulation which is another heat sink and finally, heat loss to the ambient air (which likely isn't that significant for part at least of the warm up). Repeat down the length of pipe complicated by heat transfer along the pipe itself.

It's anything but a simple calculation. Unfortunately, I don't have a calculation to just direct you to.

 

[sameg]

TD2K - yes, that was the original intent but it looks like it is going to be too complicated without some form of simulation software.

Perhaps an easier approach would be to have a set distance and try to find out how long it takes the pipeline to get to a certain temperature?

The pipeline is 3.7km and it should reach 100C. Multiple air compressors can be used if needed.

 

[chicopee]

Just remarking about the air compressor(s) capable of delivering compressed air at 300 dC. Of all air compressors that I have been involved with, I have never seen one provide that type of performance. Don,t forget that the 1500scfm is its free air delivery and not one 150 psig.

 

[vpl]

Normally, what I have seen done in power plants is that a valve is cracked allowing a small amount of steam to enter the piping and bring it up to temperature and pressure. This is done slowly, with condensate drains open to get all the condensate out of the line.

I can't imagine the size of an air compressor needed to dry out and bring up to temperature and pressure the the main steam lines at a 1000 megawatt nuclear plant! Of course the original poster didn't comment on what "cold" or "hot" meant -- other than it involved steam. He also didn't mention how long his lines were. A lot of difference if it's a lab set up versus a power plant.

And, of course, this was a one-time poster who hasn't checked back since he got "THE EQUATION" and is happily calculating away.

Patricia Lougheed

******

Please see FAQ731-376: Eng-Tips.com Forum Policies for tips on how to make the best use of the Eng-Tips Forums.

 

[berkshire]

VPL,
He said his pipeline was 3.7 km long.
B.E.

The good engineer does not need to memorize every formula; he just needs to know where he can find them when he needs them. Old professor

 

[hacksaw]

what goofy question, why would you want compressed air with all the oils and particulates in your steam line, just cleaning the lines is difficult enough-that's my tip

 

[rmw]

I'm with hacksaw on this one. I have worked with combustion turbines that have lots of hot air coming off of the compressor so I have a feel for the kind of HP required to get this job done. To flow that kind of air through that length of pipe long enough to warm the entire mass of everything involved over whole length of that pipeline means that the OP is long on air compressors and short on better ideas.

If it is so critical not to drop out any condensate, why not just heat trace the pipe with steam or electric (like a MI type HT) and maintain the pipe wall at a temperature where steam entering won't condense.

I think it is safe to say that every steam line I have ever worked with - and many had some mighty expensive turbine equipment at the other end of them - had means to deal with condensate collection during start up.

rmw

 

[sameg]

Hey guys,

I was away on vacation with no minimal access to internet and didnt really want to deal with work haha.

Thanks for all your responses.

Just for some background on this project. The client has used this technique before with great results. On a previous line they had a bunch of trucks waiting to collect all the condensate which they didn't end up needing.

What they want now is some math to back it all up. They would like us to tell them where the best injection points are along the line and how long it will take.

 

[zekeman]

"Tg-Tw=(Tg-Twi)e-αt
α=hA/cM"

This problem comes up frequently and is seldom properly addressed.

The equation above is not valid for a long pipe since its central assumption is that the gas temperature remains constant throughout its journey through the pipe, not the case for this problem. Ignoring insulation and heat transfer to ambient are less important assumptions.

The actual system is described by partial differential equatins, which don't have closed form solutions . Best you can do is solve numerically or find a program for this.

A spread sheet solution can also be developed which gives decent results.I used one some years ago, but can't seem to dig it up now.

 

[Hurricanes]

Zekeman,
I've run into this problem before and none of my references go indepth enough to advance beyond the basics. Could you direct me to some links or texts of the actual PDEs involved?

 

[hacksaw]

lets see 1500 scfm with the pipe initially at atmospheric conditions...the air velocity is 1.25 m/s...with a 3.7km line, it takes 49 minutes for the air to reach the far end...no wonder the trucks left...


no pde's for this one, plenty of pipeline simulators out there, experience dictates a waste of time to even tie up resources to preheat in this way, but if mandated to do it anyway, you measure the discharge temp, but the temperature won't increase very much given the heat loss in such a poorly insulated pipe

 

[zekeman]

"hacksaw (Mechanical)
21 Apr 12 7:43
lets see 1500 scfm with the pipe initially at atmospheric conditions...the air velocity is 1.25 m/s...with a 3.7km line, it takes 49 minutes for the air to reach the far end...no wonder the trucks left.."

Where did you get v=1.25m/s ?

I don't see how. You need the whole energy equation over the prescribed length , quite a complicated mess.

 

[IRstuff]

hmm, I got about 1.19 m/s, but close enough, at least for the initial velocity. Since the air will cool as it moves along the pipe, the calculated time is the absolute fastest anything could actually happen.

TTFN
faq731-376

 

[racookpe1978]

No.

You're using the electric power to drive an air compressor (which will use part of its (expensive) electric power to heat the air at the compressor head -> goes to compressor tank (loses heat from tank walls and transfer piping) -> goes to 3.7 km x 30 inch dia pipe, where the (formerly hot) compressed air loses its (very expensive) heat energy by conduction into the pipe walls.

Then, to get more (formerly hot) air into the very long and only-a-little-bit hotter pipe, you have to vent that (very expensive compressed but now cold) air into the atmosphere.

To heat the pipe a little bit more, you need to compressor more (cold) air into (hot and compressed air) which then flows into a cold and uncompressed pipe which then heats up the pipe a little bit more.

Would it not be better (more economical) to attach trace heat strips ON the pipe directly, and use ALL of that original electric energy to heat the ONLY the pipe, and not waste energy compressing air then bleeding it off? Control the series of heat strips by region, so as each section of pipe heats up to near operational temps, you turn off/turn down the electricity in that region to reduce the heat rate to that of thermal losses through the insulation.

No condensate. Simple controls. Less power. Cheaper to build: Compare heat strips to buying a huge air compressor.