EAC Basis and Vapor Cloud Dispersion Software Impact

[bobgorno]

Does anyone have any good or bad experiences using vapor cloud dispersion modeling as part of a basis for establishing the extents of a hazardous location around a compressor station blowdown? API RP500 seems to encourage their use, with sound judgment. Fig 14 refelcts minimum distances, and 8.2.3.1.1 and 7.2.3 seem to point you to dipersion modeling. Issue is EPA and API seem to have concluded dispersion modeling for large releases tends to overpredict the the size of the vapor cloud. If process engineers had their way, I should be classifying the whole county. Case in point, LTA capor cloud, station blowdown, 96% methane, under the right ambient conditions, will have a cloud almost 10X what the most conservative scenario in API RP500 Appendix D, Fig D-4 calls for. Granted these scenarios only last for a 10 minute event, but we do not classify based on a quantitative approach in RP500.

:rant:

 

[rbalex]

The problem with blowdown is, although infrequent, it is still a "normal" activitiy, which in the NEC world makes a substantial part of the release envelope Division 1 as Figure 14 recognizes. Appendix D generally tries to ignore Division 1 since it is really an IEC adaptation.

That said, vapor cloud dispersion can be downright scary. When I was on the API RP 500 Task Group, one of our members ran a series of well over 200 scenarios. Some of them revealed envelopes for heavier than air vapors of over 500 yards from the source. It took some highly unusual, but nevertheless possible, conditions. Lighter than air vapors were considerably more forgiving.

From a "judgment call" perspective I would consider Section 10.6.6 and Figure 49 as a basis.

 

[bobgorno]

BobA,
So you seem to be comfortable (as I am), in using API as a basis instead of taking dispersion software as an overriding basis? What was the overall feeling on the API Task Force regarding diseprsion modeling doomsday scenarios? Was there any serious considerations? Will the new RP500 (which I hear is out in draft form) consider extending the radii of blowdowns to align with dispersion modeling?

I was leaning to to a combination of Fig. 14 and Table D-4 to establish a greater extent for D2 (30-foot R and 50-feet above) while stretching the D1 to 15-feet. My compressor station is transmission, not production, so my tendancy is to also consider Figure 104, with added D1. My rationale is nothing more than increasing over the minimum recommendations of API. Obviously on an elevated blowdown stack there will be no impact on electrical installations near the stack. Establishing a 200-300-foot radius or vapor plume path @ 100-feet above grade as recommended by the process engineer would impact roads (yeah I know, not electrical), possibly roadway lighting, maybe even a Taco Bell (tongue in cheek here), but most importantly overhead HV transmission lines > 150 feet away and nearly 100-feet high.

Thanks for the input. Seems I'm not completely off my rocker.

BobG

 

[sandsnow]

What type of site are you guys talking about? I'm just wondering if it is something I might run into.

Thanks

 

[rbalex]

BobA,
So you seem to be comfortable (as I am), in using API as a basis instead of taking dispersion software as an overriding basis? What was the overall feeling on the API Task Force regarding diseprsion modeling doomsday scenarios? Was there any serious considerations? Will the new RP500 (which I hear is out in draft form) consider extending the radii of blowdowns to align with dispersion modeling?

I was leaning to to a combination of Fig. 14 and Table D-4 to establish a greater extent for D2 (30-foot R and 50-feet above) while stretching the D1 to 15-feet. My compressor station is transmission, not production, so my tendancy is to also consider Figure 104, with added D1. My rationale is nothing more than increasing over the minimum recommendations of API. Obviously on an elevated blowdown stack there will be no impact on electrical installations near the stack. Establishing a 200-300-foot radius or vapor plume path @ 100-feet above grade as recommended by the process engineer would impact roads (yeah I know, not electrical), possibly roadway lighting, maybe even a Taco Bell (tongue in cheek here), but most importantly overhead HV transmission lines > 150 feet away and nearly 100-feet high.

Thanks for the input. Seems I'm not completely off my rocker.

BobG

I'm no longer on the task group and haven't been since 2003. When the we looked at the dispersion modeling, we quitely acknowledged its existance - and basically ignored it. There was no way anyone would buy into the imposed costs for such unlikely conditions. I don't know the makeup of the group any longer, many have retired. I don't know the content of the next revison. Of course, there's always a contingent that wants absolute analytical results (i.e., no responsibility for their own judgment - "BUT, But, but; that's what the computer said.") In such cases, the process engineers are correct - the whole world is at least Division 2.

Personally, I wouldn't do much more than I said before. The activity is so infrequent and the dispersion for LTA so rapid it doesn't make much sense (to me) to be more conservative - and scare Taco Bell to boot.

 

[rbalex]

What type of site are you guys talking about? I'm just wondering if it is something I might run into.

Thanks

Larry,

A "blowdown" is basically a cleaning process for petrochem lines. Depending on the materials involved, it typically uses compressed nitrogen, sometimes steam, or (on very rare occasions) compressed air. Depending on the product, after the first blowdown, a line may never be cleaned again unless there is a process upset that demands it. Other lines are routinely cleaned during facility ?turnarounds.?

Residual (and ignitable) hydrocarbons are virtually always present for secondary blowdowns. The residual quantity, pressure and material affect the hazard envelope.

Bob

 

[bobgorno]

For clarification, the "blowdown" I am referring to in my post is not the "purging or blowdown" of gas lines to clean them. Rather it is the situation where a gas compression station is shutdown under emergency or other non-frequent conditions and where the inlet and outlet to the station are blocked by fail-closed shutdown valves and the "trapped gas" in the compressors and station piping are vented to atmosphere in order to bring about a safe and static condition.

 

[rbalex]

For clarification, the "blowdown" I am referring to in my post is not the "purging or blowdown" of gas lines to clean them. Rather it is the situation where a gas compression station is shutdown under emergency or other non-frequent conditions and where the inlet and outlet to the station are blocked by fail-closed shutdown valves and the "trapped gas" in the compressors and station piping are vented to atmosphere in order to bring about a safe and static condition.

I forgot the "transportation" application of the term. In that case, I'd use Figure 104 only and not look back. The release velocity is usually so high, dispersion is almost instant.

You may also want to review the American Gas Association's XF0227. Technically, it's a "utility" document and isn't ANSI, but the laws of physics are no different and it has served that industry well for years. RP 500 and NFPA 497 are only recommended practices anyway.

 

[bobgorno]

Bingo Bob,

You went right where I wanted. AGA XFO 277, Figure 1 with RP500 Fig 104 as backup support. I had AGA Figure 1 documented in my report to my PM.

I can wave that flag to my project manager as a confirmed. :thumbsup:

Thanks,
Bob

 

[sandsnow]

Thanks, Bob and Bob

 

[rbalex]

Bingo Bob,

You went right where I wanted. AGA XFO 277, Figure 1 with RP500 Fig 104 as backup support. I had AGA Figure 1 documented in my report to my PM.

I can wave that flag to my project manager as a confirmed. :thumbsup:

Thanks,
Bob

Welcome and I would support it.

 

[rbalex]

Thanks, Bob and Bob

You're welcome