Explosion Proof Thermowell

[etownsend]

I work for a company that manufactures explosion proof electrical equipment and enclosures, and have recently been asked to include a temperature sensor with transmitter to measure the temperature inside the enclosure, with the transmitter mounted outside the enclosure (in an IP66 head). Selecting a sensor and transmitter (intrinsically safe) has not been a problem, however finding a thermowell which is certified (ATEX) to maintain the explosion proof certification of our enclosure has not been successful.

I have talked to numerous suppliers and manufacturers, as well as gone through numerous internet searches, and although there are explosion proof thermowells available they are designed to provide protection from an explosion in the transmitter head (inside the thermowell out), and not from the measurement area in (outside the thermowell in) - as I am looking to do.

I have also spoken to contacts at hazardous area certification agencies and they are not aware of any product offhand, however they are checking with people they know.

It is imperative that the thermowell be certified for use as we intend to use it as it will be certified as part of our overall enclosure which will be put under scrutiny by our certifying agency. There is an option of buying a non-certified one and certifying it ourselves, but I do not want to go down that path if I do not have to.

Not sure if this is the best form for this question, but thought I would post it here as it is related to temperature measurement equipment.

Any help appreciated.

E. Townsend.

 

[JimCasey]

Explosionproof instruments are certified by filling them (internally) with a flammable mixture of air and acetylene.(NEMA 7, 9; or Class 1 Div 1 groups A,B, C, D. Where the housing may not be rated for all 4 groups) When the mixture is detonated inside the housing, the housing must not allow the flame front to propogate to the outside where it could trigger additional combustion. Xproof enclosures have heavy section, the covers are threaded with at least 7 threads engaged or they have a wide flange surface to make the flame path long enough for the flame to be quenched by the heat absorption of the metal.

There is, to my knowledge, no standard regarding withstanding an EXTERNAL explosion....That could just as likely fracture the pipe on which the thermowell is attached. Some instrumentation sold for military use must withstand shock-and vibrration testing to show that it will remain useful after the vehicle in which it is used sustains combat damage.

 

[djs]

Hello;
If you have an intrinsically safe sensor and tramnsmitter, then you do not need to use a thermowell. Mount the sensor and transmitter internal to the equipment and bring out the signal wires.

 

[etownsend]

The issue I am having is that I am trying to measure the temperature inside a certified flameproof enclosure, which requires that any entry to that enclosure must also be certified as a flameproof component.

I had previously looked at mounting all of the components inside the flameproof enclosure, as you have suggested, but in order to have intrinsically safe circuits mounted inside, the enclosure must be designed and certified Ex d[ib] (both flameproof and an intrinsically safe apparatus), which would encompasses a large design change to the internals of the enclosure (there are many spacing and metallic barrier requirements between IS and non-IS circuits), which would make the certification and design costs unreasonable. It was for that reason that I was looking for an Ex d certified thermowell which would keep all of the IS instrumentation outside of the flameproof enclosure.

Glad to know that my thought process is not out of whack, that others are thinking along the same lines, thanks for the post.

E. Townsend.

 

[Ashereng]

E.

I am sure you have considered this, but what is the difference between measuring the enclosure surface tempeature vs. measuring the temperature inside the enclosure through the surface of a thermowell?

Is it a difference of the thermowell thickness vs. the enclosure wall thickness?

"Do not worry about your problems with mathematics, I assure you mine are far greater."
Albert Einstein
Have you read FAQ731-376 to make the best use of Eng-Tips Forums?

 

[danw2]

You're trying to measure the temperature inside an EXP housing with a thermocouple.

Any wiring to that EXP housing has to go through seal-offs.

Why not run the thermocouple through one of the standard conduit access ports and have the T/C extension wire sealed off, like any normal signal wire.

The entry of the T/C wire is through the existing conduit ports, and conduit installed per code should cover this situation, should it not?

Then you could run the thermocouple extension wire to an IS barrier in your panel (which is outside the EXP housing), to ensure that the thermocouple won't be a path for hazardous voltage/current back into the EXP housing.

Dan

 

[etownsend]

In the end we are deciding to include a thermowell as part of our enclosure when we go to have it tested and certified, thereby certifying a thermowell ourselves.

It will not add to the cost of our certification, however it will be a one time "field certification", and if we need to do this again in the future for another enclosure we will once again need to do the explosion and pressure testing to field certify the enclosure including the thermowell.

Just as a reference for those wondering, our enclosure is being certified flameproof to IEC 60079-1 (IEC 60079-0 general requirements also applies), and any intrinsically safe components are IEC 60079-11. If any intrinsically safe circuits are run near other circuits in an enclosure protected by another type of explosive atmosphere protection (Ex d, Ex p, Ex e) than it must also meet all of the requirements for Ex i, which usually requires a lot more design, documentation, and certification testing which would significantly add cost if all you needed it for was temperature measurement. (I do like your thought of the barrier DanW2 - but at the moment am going for the simplest route with the lower costs).

Thanks.

 

[BrianR]

How about using an inductive coupling through the wall of the enclosure with the sensor contained within the enclosure. There is the added complexity of having to modulate / demodulate a high frequency magnetic field but this is well understood technology. I am not sure what coupling you will get through a solid metal wall as there will be a lot of attenuation, but it depends on what metal it is constructed from.