Today is...
Monday, June 25, 2018
Welcome to, the global online
community of automation professionals.
Featured Video...
Featured Video
EtherCAT with CTC’s master lets your multivendor network play well together...
Our Advertisers
Help keep our servers running...
Patronize our advertisers!
Visit our Post Archive
thermowell insertion length in vessel
Can anyone guide me about standard of thermowell's insertion length?
By fereshteh on 28 April, 2007 - 5:51 pm

Hi! Everyone,
Can anyone guide me about standard of thermowell's insertion length? we want to install one tempreture transmitter in side mount of vessel. can you give me some information about thermowell's insertion length on vessel.


By Tapas K.Bhattacharyya on 12 May, 2007 - 5:49 pm

Thermowells are generally of 5 Types from the point of view of it's construction.

5.Closed end Tube.
and Two tyes from the point of view of it's Process connection.

Thermowell are generally fitted to tanks as side mounted where it's tip can reach the area of temp jone meant for measurement.

Also for a tank usual process connection is flanged.usually 1 1/2" ANSI 300 # flage a used.

"U" insertion length mau be as low as 25 MM to 300 MM.

but a resonant frequency check is advised by calculation,if fluid velocity inside tank is available.

also one should take care about material of thermowell as fluid in contact may have corrosive that case Nace MR-75-01 give guideline for correct selection.

There is a well developed standard avaialable with PIP(Process Industry Prctice).one may consult that too.

the subject of thermowell is well discussed in Instrumenation design field,and almost every design engineer has his experience of design and use.

If you want more and indepth discussion,may be very exotic and challenging application.
Please feel free to contact me.

By fereshteh on 19 May, 2007 - 3:02 pm

Thanks a lot for your reply.

I want to know if ther are any standard about insertion length ?or it is based on experiance?

best regards

By Tapas K.Bhattacharyya on 30 December, 2007 - 3:52 pm

Insertion Length should be sufficient to measure the bulk temperature of the fluid without conduction effect.

The tip of the Thermowell should reach unstagnated or flowing fluid but not exceeding velocity limits for gas and liquid.


Good themowell design is always checked for resonance damage as per ASME PTC 19.3

please feel free to contacte-mail:

Dear Sir,

We want to know which drilling machine and CNC lathe machine is needed for drilling and making bore of kind of thermowell. And if possible inform me of standard numbers for making thermowell and designing it. Please let me have your kind reply by email. Thanks.

By Tapas K.Bhattacharyya on 30 December, 2007 - 3:48 pm

Any standard Gun-drilling OR Broaching Machine can do the job.

The Material to drill is generally SS316.
the job size is to drill a fine bore of 0.260" to 0.385" with tolerence of better than 0.01".

The minimum lenth is generally 146 mm to 150mm.

Bar stock with smooth outside diameter and OD insertable to a stanadard 1 " or 1 1/2" flange or M55 thredolet/wedolet with hole dia of 1/2" is the drilled component with built in inside tip centering pin for special design and sprinf loading element base,and a cricumferential weld with good SS 316 intregity is the Requirement.

There are simple CNC programs downloadable from PC which may be further intregatable to Regular Design softwere for Instrumentation design such as INtools or COMOS. and even futher Intrgatable with Web service For state of the art Project management on global scale manufacturing.

By Anonymous on 5 August, 2007 - 6:55 pm

i need to discuss on thermowell design for cyrogenic applications

can you give me your mail ID ?


By Tapas K.Bhattacharyya on 3 January, 2008 - 1:40 am

Dear Sudharshanan,

The Teperature below ambient are meant for Cryogenic application. Metal behavior is characterized somewhat by crystalline structure. The yield and tensile stregths of metals that crystallize in the body centered cubic from IRON, MOLYBDENUM, VANADIUM & CHROMIUM, and depend greatly on temperature. These metals loss in ductility in a narrow temperature region below room temperature.

The tensile strength of metals with face centered cubic structures - ALUMINIUM, COPPER, NICKEL & AUSTENITIC STAILESS STEEL is more temperature dependent than their yield strength, and the metals often increase in ductility as temperature decreases.

Tranformation occuring in compositions that are normally stable at room temperature, but metastable at cryogenic temperatures, can greatly alter their behaviour.

For example, the combination of gross plastic deformation and cryogenic temperature can cause a normally ductile and tough stainless steel, e.g. 301, 302, 304, 321, to pertially transform to body centre cubic structure, resulting in an impairment of ductility and toughness and failure under stress. Whereas a fully stable stainless steel 310 cannot be transformed at cryogenic temperatures. The 300 series steels offer a good combination of toughness and weldability for service to the lowest temperature but inadequate for lightweight structures.

Precipitation hardening A286 Stainless steel has higher strength when cold worked before aging.

The only alloy steel recommended for cryogenic service is 9% nickel steel.

It is satisfactory for service down to -195 degree C and is used for transport and storage of cryogens because of its low cost and ease of fabrication.

Other alloy steels for service in the cryogenic ranges are the Steels A201 and T-1 and can suffice to - 45 degree C.

Nickel Steels with 2.25% Ni can suffice to -59 deg C. and Nickel steels with 3.5% Ni to -101 Deg C.

Some other low teperature Steels are A203 grade D, E, A543 grade 1, 2.



For more Physics and metallurgical discussion like fermi enegies of metals, fermi temperatures, and velocities, quantum physics like Einstein-Debye Specific Heats, etc., you can have your study further for knowledge's sake.

But for Engineering point of view, the selection of material so far Thermowell is concerned, above information is hoped to be sufficient.

tapas @ daresd. com

By Anonymous on 3 May, 2008 - 3:30 am

From precise connection point of view there are three types of thermowells available as per my knowledge:
1. flanged
2. threaded
3. welded

From construction point of view there are two types:
1. tapered
2. straight

I have not come across anything about the other types Mr. Tapas mentioned.

By Anonymous on 3 May, 2008 - 3:26 am

The immersion length should be calculated to ensure that the sensing element has sufficient immersion into the process medium to give accurate measurement. To ensure optimum results and freedom from outside influences, thermowell tip should be located at a point which equates with the lower portion of the centre third of the pipe line.

For tanks, the depth of immersion obviously varies considerably but, as a rule of thumb, 300 mm from outer wall will give ideal results. In high velocity applications, wake frequency effects must also be considered.

By Sureshkumar on 18 January, 2009 - 10:58 am

Somebody posted below mentioned for thermowell insertion length in pipe line:

>The immersion length should be calculated to ensure that the sensing element has sufficient immersion into the process medium to give accurate measurement. To ensure optimum results and freedom from outside influences, thermowell tip should be located at a point which equates with the lower portion of the centre third of the pipe line.>

Please clarify what is mean by lower portion of the centre third of pipe line.

Thermocouples and most RTDs have a short active length. Bi-metal thermometers about 1-1/2". Unless the tank contents are very viscous an insertion of 4" (100mm) should be more than adequate. I know some designers will specify a "U" length to the center of pipe be it 4" or 24", I think that's wrong.

I do some very high pressure applications where we strap the sensor on the outside of pipe under the insulation, there seems to be very little difference in the reading although there is a slight delay.


By dave bartran on 15 February, 2009 - 7:49 pm

That is my experience, some 90% of the thermowell errors I've encountered are un-related to insertion length.

Amazing that the 1/3 rule persists.

You can use this spreadsheet to calculate the acceptable lengths:</b.

Material Selection for different purposes:

We've been using flanged Thermowell, Van-Stone type to be precise as we use SS316 Thermowells with Carbon Steel Flange connection due to concerns on SS316 suffering from chloride stress cracking for temp over around 65'C. Best to use One Piece Inconel or above.

For the U-length, as long as the Frequency Ratio Unsatisfactory (r => 0.8) it is acceptable. Normally the longer the U length the higher the ratio will be (not good).

You need to make sure you have sufficient length for the insertion, so for pipe mounted you have to check the pipe wall thickness, and normally the insertion is around 1/3 depth, then from the pipe surface to the thermowell flange there's a Standoff distance you need to set, normally we use 150mm, add everything up and you got your U-Length.

Minimum pipe diameter for thermowell should be DN100(mm), any pipe smaller than that you need to extend it to a bigger diameter.

For Tank/Vessel just check the wall thickness and asign the depth you want the thermowell to reach and use around 150mm standoff.

By Saeed Beheshti Maal on 9 June, 2009 - 11:44 am

TW insertion length in horizontal vessels shall be 457 mm while not exceeding half of the vessel diameter.

Saeed Beheshti Maal
beheshtimaal [at]