7 replies to this topic

[Lowflo]

There is an apparent conflict between OSHA 1910.106 and API 2000. The API 2000 F factors allows a reduction of heat input below the F values allowable in OSHA 1910.106. It seems inconceivable that API would publish a design procedure that is in conflict with US regulations. If true then most all of the emergency vent devices in the US are be in violation of OSHA.

I'm thinking there's a simple explanation for this apparent conflict, but after searching for the answer I'm stumped. Can anyone shed some light on this?

Here's the issue…….

•The heat input equation is the same (Q = 21,000A^(0.82)) in OSHA and API 2000. That's the NFPA-30 equation that was adopted by OSHA, and therefore used in API 2000.

•The basis for determining exposed surface area is the same in OSHA and API 2000. It's based on NFPA-30 rules.

•But, the environmental factors in API 2000 are different. OSHA uses the NFPA30 F factors, consistent with it's use of the NFPA heat equation and surface area. The API 2000 factors can be much lower, resulting in a much lower heat input from the fire….. That's the apparent OSHA violation.

OSHA 1910.106 Flammable and combustible liquids
1910.106 b.2.v.f

The required airflow rate of subdivision © or (e) of this subdivision may be multiplied by the appropriate factor listed in the following schedule when protection is provided as indicated. Only one factor may be used for any one tank.
0.5 for drainage in accordance with subdivision viib of this subparagraph for tanks over 200 square feet of wetted area.
0.3 for approved water spray.
0.3 for approved insulation.
0.15 for approved water spray with approved insulation.

NFPA-30 (2008 ed) 22.7.3.5

(1) A reduction factor of 0.5 shall be allowed for tanks with wetted area greater than 200 ft2 (19m2) that are provided with drainage that meets the requirements of 22.11.1.
(2) A reduction factor of 0.3 shall be allowed for tanks that are protected with a water spray system that meets the requirements of NFPA 15, Standard for Water Spray Fixed
Systems for Fire Protection, and that are provided with drainage that meets the requirements of 22.11.1.
(3) A reduction factor of 0.3 shall be allowed for tanks that are protected with an automatically actuated water spray system that meets the requirements of NFPA 15, Standard
for Water Spray Fixed Systems for Fire Protection.
(4) A reduction factor of 0.3 shall be allowed for tanks protected with insulation that meets the requirements of 22.7.3.8.
(5) A reduction factor of 0.15 shall be allowed for tanks that are protected with a water spray system that meets the requirements of NFPA 15, Standard for Water Spray Fixed Systems for Fire Protection, and that have insulation that meets the requirements of 22.7.3.8.

API 2000
See Table 4 in attached file.

[joerd]

It is the other way around, NFPA adopted the API 2000 basis and gave it its own twist. Most of your concern is addressed in Appendix B, from which I quote:

This basis for emergency venting was adopted by the
National Fire Protection Association (NFPA) and was successfully
used for many years. As far as can be determined,
except for some containers of unusually small capacities, no
case has been recorded in which a tank failed from overpressure
because of insufficient emergency venting capacity when
vented in accordance with this basis.

Because of some concerns expressed about the differences
in various methods for determining fire case venting requirements,
and a desire to standardize on one method, the subcommittee
surveyed approximately 100 companies from
1993 to 1996. This survey indicated that there was no detectable
difference in the level of safety provided by using the Þre
sizing methods found in this document, API RP 520, API
RP 521, NFPA documents, or other commonly used Þre case
venting calculation methods. The subcommittee abandoned
efforts to standardize the industry on one method for determining
fire case venting requirements in 1996.

I concede that these quotes don't really answer your question, and of course a higher F factor is never wrong, but it looks like the industry has been using API 2000 for many years without incidents resulting from that basis.

[Lowflo]

joerd,

I'm not questioning whether or not the guidance in API 2000 is safe. The point is whether or not it's a legally acceptable design practice for US installations. When this was first brought to my attention, I thought there must be a simple explanation.....there's no way this default global standard for sizing emergency relief valves could possibly be in violation of US OSHA regulations. But, if there is an explanation I can't find it. I'm hoping that someone will point out something that I've overlooked that explains why it is legally acceptable to use API-2000 in the US. Otherwise, those of us in the US are violating the law by applying API-2000. Again, that sounds crazy! Is it true?

For you folks in the refinery industry, has this been brought up during an OSHA NEP audit? Is OSHA and/or API aware of this conflict?

I sure hope there's a simple explanation.

Thanks,
lowflo

[latexman]

Lowflo,

Do the notes that were excluded from your attachment shed any light on the subject?

[Lowflo]

Latexman - See the attached file. This is API-2000 Table 4 (Environmental F factors) with the notes. Unfortunately, the notes don't offer an answer to this puzzle.

[Lowflo]

Well, let me try that again...the attachment wasn't in the previous post.

Attached Files

•  Table 4 with notes.doc   96.5KB   39 downloads

[latexman]

Yep, no help.

[Lowflo]

Can anyone explain why the API-2000 fire heat input calculations is not in violation of US OSHA reg's?

Have any of you been through an NEP audit? My linkhttp://www.osha.gov/pls/oshaweb/owadisp.show_document?p_table=DIRECTIVES&p_id=3589

B. Relief Systems

3. Does the employer's process safety information (PSI) include the codes and standards they used in the design of relief systems?
Yes No N/A

If no, possible violations include: 119(d)(3)(i)(F) – the employer did not include in its PSI, the codes and standards they used in the design of relief systems.

Compliance Guidance: This is a fundamental management system requirement for ALL equipment in a PSM-covered process. The employer must specify the design they employ for their covered equipment.

I'm guessing that OSHA is completely unaware of this conflict. If one listed API-2000 as the basis for an emergency relief vent sizing, that would be viewed as acceptable from a RAGAGEP perspective (which it really is) but the OSHA inspector wouldn't realize that that's not acceptable per OSHA 1910.106 (1910.106 b.2.v.f).