13 replies to this topic

[JoeWong]

Dear all,

Emissivity factor is one of the factor affecting flare radiation. However, there are so many correlations available in the literature to determine this factor.

In recent refinery project, i have encountered a difficult situation.

The fluid to flare tip contains about 10% C1-C3, 30% C4-C7, 35% C2-C4 Olefins, 10% H2, 12% H2O & 3% BTEX. The average MW is about 50. (all in mole%).

Few reputable vendors come back with different emissivity factor (0.13, 0.25, 0.3). The impact of the factor is the flare stack height is 120m (with 0.3) compare to 60m (with 0.13).

Any idea how shall we approach this situation.
regards,

JoeWong

[pleckner]

Ask each vendor how they came up with their particular values. Perhaps you can investigate after being given the answer to this question. You can tell them about the other values you received and try to get them to come up with an explanation as to why the differences are so large.

Best thing might be to take a sample of the mixture and have it tested.

[STEVENLEE]

It appears from the data that the emissvity factor is dependent on the height of the flare tip. I am guessing that the emissivity factors are measured near ground level rather than near the combustion tip becasue of the potential of exposing a person to radiation in the form of heat. I am new to flare systems and would like know what you find out.

Thanks.

[JoeWong]

Phil,

Thanks for your ideas.
Infact, questions asked and the response from all vendors are :

a) Vendor quoted LOW emissivity factor said :
"We have XX years of experience in design and supply flare tip for similar capacity. Our figure is experience and test figure.". In addition, they provide the lab test results for similar application.
"Other vendors design is OLD design, overly conservative......with our new design....."


Vendor quoted HIGH emissivity factor said :
"We have XX years of experience in design and supply flare tip for similar capacity. Our figure is experience and test figure.". In addition, they provide the lab test results for similar application.

"Other vendors are too optimistic. They know that you will not test and prove it once it's installed"

They also have IDENTICAL and COMMON asnwer :
Our estimation is CONFIDENTIAL and shall not be DISCLOSED.



"Best thing might be to take a sample of the mixture and have it tested."
Per my experience, almost NO one project we have tested it. Mainly due to $$ issue...

Any idea how to handle above situation ?


regards,

JoeWong

[pleckner]

Any of these vendors happen to be John Zink? They are the granddaddy of flare design and I have a lot of confidence in them.

[JoeWong]

Phil,

One of the vendor is JZ. They are using LOW emissivity factor. No doubt, JZ is one of the most reputable vendor in Flare. We always request JZ for proposal (almost all jobs).

In this particular case, we have more vendors (equally good reputation) using HIGH emissivity factor.

Secondly we have compared other project with low MW (about 25-28), the emissivity by JZ is more-or-less similar to this case where the MW is close to 50.

As you know, MW is one of the factor affecting emissivity. However, JZ has considered MW has almost no impact on the emissivity. This caused some difficulties for me to accept it.

regards,
JoeWong

[joerd]

There is a lot of discussion about the proper value of emissivity. In your case, probably having hydrogen in there makes a big difference too. I have a book by Selma Guigard, Heat Radiation From Flares, May 2000, which evaluates a lot of methods. There seems to be no good-for-all method available (like you are experiencing). I quote from the executive summary:

This report briefly reviews and summarizes theoretical and observational relationships for determining the fraction of heat radiated from flares in proximity of a flame. Nine articles are reported in which the fraction of heat radiated in proximity of a flame is determined by theoretically-derived relationships. Two articles are reported in which the fraction of heat radiated in proximity of a flame is determined by empirically-derived relationships. A matrix summarising which parameters have been used to determine the fraction of heat radiated for each of these relationships is shown below. The applicability of these relationships to the general case is limited. The theoretical or empirical conditions for which many of these relationships are based upon are situation-specific. In addition, limited information was provided in many instances on numerous parameters that are known to influence flare heat radiation losses (e.g. stack exit velocity, crosswind velocity, aerodynamics of the flame, etc.).

So it's really up to the vendors and their credibility to supply you with the right flare, I'm afraid.

[JoeWong]

joerd,
Thanks for your response.

...So it's really up to the vendors and their credibility to supply you with the right flare, I'm afraid.

I am stun with this statement but this is a FACT with my limited knowledge and experience in Flare.

I have the book that you mentioned and thoroughtly gone through this book. I also have worked out the emissivity with almost all model in this book. Most results tend to give me HIGH emissivity. This not realy inline with JZ's prediction (NOTE). No doubt JZ is the grand-daddy" in flare but above has put me in a difficult situation.

I believe some of us have experienced same situation in the past and may be experiencing it near future.
I am almost certain that there are many flare specialist reading this forum. I am sincerely invite you to participate in this discussion for the good future of everyone involve in Flare.



NOTE : This post is PURELY for TECHNICAL discussion. There's NO intention to against any party included JZ.


regards,

JoeWong

[process equipment]

Dear all,

I'm also facing this problem : the emmisivity factor of H2O and CO2 from fluegas at atmospheric pressure in a burner ?.

Should I calculate each fraction or as mixtures ?

Cheers

[Joyy]

Various methods and reference to estimate F factors are discussed in the attachment.

Conclusion on Fraction of Heat Radiated, F:
- There is no direct relation of F with mol. wt. Even higher mol. wt. gases can have lower F eg. Propylene and propane
- Use F = 0.3 in API 521 equation unless using simulation software. Use the height of flare stack calculated into the datasheet with vendor to verify mark.
- If client insists on a vendor providing lower stack height due to the lower F of its flare tip model, ask the vendor to guarantee a radiation test at site using radiation intensity meters at emergency or design loads.

Attached Files

•  Fraction_of_Heat_Radiated__F_.doc   415KB   322 downloads

[JoeWong]

Joyy,
Thanks for your great inputs. Appreciate.

Do you mind to advise the vendor 1, 2 &3 ? You may drop me a personal note if you feel unconfortable to release this information in public.

Sometime really doubt if guarantee will works and how this can be proved for a life plant.

[Mehrdad]

hi

you can try:

http://www.flaretip.com



may it's be useful

[Joyy]

Hi,

In continuation to the various methods discussed in my attachment in the above post, check out this new model to predict flame emissivity / fraction of heat radiated

See the article “Making the flare safe” by David Shore (Flaregas Corporation) published in J. Loss Prev. Process Ind. Vol. 9 No. 6. pp. 363-381, 1996

You can download it from here,
www.geocities.com/flareman_xs/Systems/makingtheflaresafe.pdf

Refer to 365, a new model to predict flame emissivity / fraction of heat radiated is proposed. I have attached the extract of the model here.

Hope you find it useful.

Best Regards,
Joyy

Attached Files

•  Emissivity_Model_by_David_Shore.doc   88KB   182 downloads

[JoeWong]

Joyy,

I have read through this document sometime ago... Thanks for your information.

I don't know how other engineers managing this issue, personally i am still struggling with it... by taking ALARP principle but not giving up on conservatism...

David is active in Eng-Tips forum. Really hope that he can provide some inputs and advice...


Note :
To add the complexity... many vendors proposed to use emissivity factor of 0.1-0.12 for sonic tip. I have expereicne JZ used 0.08 in one of my previous project. Nevertheless, there was research (old paper) shown that minimum emissivity factor of 0.17...