from the Forum department...
Thermowell Length for best accuracy
Sensor technologies. topic
1 out of 1 members thought this post was helpful...
Posted by GAFOOR on 11 January, 2013 - 7:31 am
We have 42'' pipeline salesgas, thickness is 25.4 mm., velocity max 12.2 m/sec.What is the Thermowell immersion length required for best accuracy? Is there any length limitation for gas application?

Is there any velocity collars required?

Pls help me

1 out of 1 members thought this post was helpful...
Posted by Roy Matson on 11 January, 2013 - 11:28 am
I don't pretend to be a flow expert.

For liquids I normally insert the thermowell just a couple of inches. with turbulent flow the temperature profile is even, I think with gas I would go about 4 inches.

A lot of guys like to get right in the center of the pipe but if you are not careful the vortex shedding around the probe will put too much stress on the well. I know Rosemount will calculate the likelihood of that happening if you ask.

Bear in mind also that the probe insertion length may effect the accuracy of your flow measurement device.

Posted by gafoor on 12 January, 2013 - 1:29 am
it is for temperature application. Pipe size is 42'' and thickness is 25.4 mm.

since pipe dia is 42'' what will be the insertion length for best accuracy? 300 mm is ok???

for gas application "u" length is 1/3 of the pipe size is required???
pls help.

2 out of 2 members thought this post was helpful...
Posted by Asok Kumar Hait on 13 January, 2013 - 7:08 am
The error in the temperature measurement happens in such cases due to a well know phenomena called "conduction error". One side of the thermowell is in the hot fluid (inside the pipe) and the other side (outside the pipe)is at cold atmosphere and therefore, a temperature gradient is established when an equilibrium stage is reached due to heat conduction and radiation from the thermowell. Due to this temperature gradient, the tip of the thermowell will not achieve the same temperature of the fluid (or near to that) unless the thermowell immersion length inside the fluid is sufficiently immersed. A thermodynamic calculation can be done to find out exactly how much should be the thermowell immersion length taking care of many factors.

However, in Industry we normally don't do this calculation. We use a thumb rule whic is - if the thermowell immersion length is 4 times the thermowell tip dia then the temperature mesurement will be sufficiently accurate.

A typical thermowell tip dia is 18 mm and therefore, in my opinion 4 x18 mm i.e. 72 mm inside the pipe will be sufficient to measure the correct temperature. Since your pipe dia is big and there is sufficient room to penetrate more you can increase the thermowell length more to take care of the temperature gradient if the pipeline is not insulated. Normally we use a maximum length of 200 mm in the flowing fluid. But you have to ask your temperature sensor supplier to check the wake frequency calculation so that this immersion length of 200 mm is passed in the stress and wake frequency calculation.

To summarise, in your application, if the pipe nozzle stand out is 200 mm, you need to consider a thermowell with total insertion length:

200 mm + 72 mm = 272 mm minimum
200 mm + 200 mm = 400 mm maximum if allowed by stress & frequency calculation.

Hope this helps.

Posted by Mitch Johnson on 14 January, 2013 - 7:16 am
The longer and skinnier the well, the better for response time and accuracy. The shorter and fatter it is, the stronger it is.

The rule of thumb is to get to the middle third of the pipe, but before accepting that as your immersion length you need to confirm the design strength per ASME PTC 19.3TW-2010. After all it doesn't help that your well is long and skinny if it breaks or the installed sensor fails due to vibration.

You may want to check out at SwiftyCalc

SwiftyCalc is a free software that will provide you with a quick max insertion length and wake frequency calculation result in seconds applying the ASME 19.3TW calculation.

Posted by Mitch Johnson on 14 January, 2013 - 7:27 am
Here is a link to an article that you may find to be helpful: -the-asme-standard

Posted by gafoor on 14 January, 2013 - 10:25 am
Thanks to every body....

Posted by Roy Matson on 14 January, 2013 - 1:49 pm
Does this 1/3 include the nozzle, if so it wouldn't make sense. It would be ok on your large pipe but not on smaller ones
42"/3 = 14"

a typical flanged nozzle has a projection of 6" so the probe would be in the pipe 14-6 = 8"

for a 12" pipe the U would be
12/3 = 4" and the probe wouldn't reach the pipe.

I like Asok Kumar Haits simple formula.

As for the temperature gradient, usually the pipeline is insulated so the nozzle should be very close to the process temperature and there won't be much gradient.

I try to standardize on just a few different "U" lengths so the probe for a 12" pipe would be the same "U" length as the probe for your 42"


Posted by Mitch Johnson on 15 January, 2013 - 4:57 am
> Does this 1/3 include the nozzle,

> I try to standardize on just a few different "U" lengths so the probe for a
> 12" pipe would be the same "U" length as the probe for your 42"

Good practice. You will want to confirm that the design passes the ASME calculation for all conditions at a standard length

Posted by Ulf on 30 January, 2013 - 6:56 am
Here is new online tool to calculate your thermowell according to ASME eneral-Use-Sensors/Thermowell-Calculations/Pages/preliminary.aspx

moderator's note: please make sure you delete any spaces in the long url when you cut and paste this in your browser.

Posted by d- on 1 February, 2013 - 9:35 am
you need to specify what is accurate enough
along with the temperature, pressure, and whether of not the pipe is insulated, including the base of the thermowell.

traditional thermowell length assignments tend to be very risky both from a mechanical integrity point of view, but also in senor reliability and measurement accuracy.

There are special thermowells suited to low pressure, low velocity gasses, that get past the longer is better scenario, but they are typically quite short.

Posted by Mitch on 19 September, 2013 - 2:26 am

I need to size thermowells for pipe from DN65 to DN700 (air application, velocity around 30m/s, pressure below between 0bar g and 1bar g).

Thermowell data type:
Shank profile: straight
mounting: by threaded 1/2

Problem is my thermowell geometry is out of scope of ASME PTC 19.3TW-2010. support diameter is identical to tip diameter (9mm). Support diameter mini is 15.8mm in ASME code.


Since velocity looks high (i think?) I must check that a thermowell length of ~1/3 1/2 of the pipe fits well.

Posted by d- on 22 September, 2013 - 1:25 pm
you can always give it a try,

while it may be outside the scope of the current 19.3 dimensionally, the engineering principles involved can be used to insure that you have a safe design.

Your use of this site is subject to the terms and conditions set forth under Legal Notices and the Privacy Policy. Please read those terms and conditions carefully. Subject to the rights expressly reserved to others under Legal Notices, the content of this site and the compilation thereof is © 1999-2014 Nerds in Control, LLC. All rights reserved.

Users of this site are benefiting from open source technologies, including PHP, MySQL and Apache. Be happy.

A copy of the universe is not what is required of art; one of the
damned things is ample.
-- Rebecca West
Advertise here
Servo, stepping motor control, analog & web HMI in one system!
Time to incorporate data handling, web HMI and motion in one system!
Servo, steppers, analog, digital & web HMI - Fully Integrated!
View free setup and multi-vendor EtherCAT demo videos online is the largest Automation community on the web. Learn how to advertise here now...
our advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive