Eng-Tips is the largest engineering community on the Internet

Intelligent Work Forums for Engineering Professionals

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

Where in a pipe should I measure temperature?

Status
Not open for further replies.

shackney

Mechanical
Jun 27, 2003
10
0
0
CA

I need to measure the temperature of a liquid in a pipe. Assume turbulent flow, temperature slightly elevated over ambient.

Is there an accepted standard for the insertion depth into the pipe of a thermometer bulb (or other temp sensor)? What's left of my common sense tells me that any reading near the centre of the pipe will tell me the temp of the bulk of the fluid. But I no longer trust common sense, and there's probably other factors at work here (like maybe the restriction added by the thermowell??

Thanks in advance.

shackney
 
Replies continue below

Recommended for you

Temperature element must be small (relative to the pipe inner diameter) to avoid any flow resrictions. It also must have minimum length to insure the measuring tip is not effected by the pipe outside temperature. Assume you are talking about small pipe diameters. The solusions:
1. Find (or create) along the measured line an Elbow. Replace it to a " T " connection. Insert an RTD at the new outlet of the Elbow so it penetrate into the pipe, at the middle of flow with much larger length then a radial installation can get. Normally Temp. element comes wit Thermo well which inlarge the measuring diameter. If the pipe is small (less then an Inch) consider using Temp. Element without a well.
 
I would suggest a well for temperature sensors, they fail. Temperature wells should be installed with flow. Heat dispersion compound inside the well is a must. With turbulent flow, well placement should be anywhere it does not restrict flow.

Building Automation Professional
 
Conduction error is minimal in most liquid flows (turbulent) for L/D of the sensor/well > 2.5 or so. It is not always necessary to place the sensor at the center of the process unless other factors are involved.

If the liquid is near ambient but you need to measure it accurately use an rtd or a thermister. thermocouples start having noise/compensation problems if the millivolts at the sesor are too small.

you did not specify the response time requirement.



 
For turbulent flow a ROT is to place the thermowell(TW) in the middle third of the pipe.

Depends on your application.
If the pipe is large enough to accomodate the TW then it can be placed at the point desired. If the pipe is too small to accept the TW then one may have to insert it into a Tee or expand the pipe as long as one maintains turbulent flow in the expanded section.
As for the restriction: if it was in a pump suction I would investigae further but normally there is no problem. If the flow area is reduced significantly then I would find an equivalent restriction and model it that way. If a Tee has to be used then one may have a significant reduction in the flow area requiring further investigation of the hydraulics.
 
Which is your aplications? Which liquid, which temperature?What do you need this temperature for? If the temperature is small (100-200 F) you should use RTD, but if you temperature is high uses termocuple. To do some measures, and if you can use other sensor in near point, it can be helpful. And then use stadistic theory to calculate error.

 
Status
Not open for further replies.
Back
Top