3 ms·

Let's not forget about the time they blew up San Bruno: https://en.wikipedia.org/wiki/San_Bruno_pipeline_explosion https://en.wikipedia.org/wiki/San_Bruno_pipe

Let's not forget about the time they blew up San Bruno: https://en.wikipedia.org/wiki/San_Bruno_pipeline_explosion https://en.wikipedia.org/wiki/San_Bruno_pipeline_explosion

x0x0 8y ago
The pipe was installed in 1956, before modern testing methods such as xrays. It worked for over 50 years. Per PG&E, they have 6,700 miles of gas distribution pipelines [1] like the ones in San Bruno. They have another 40k miles of other gas pipelines. Building perfectly safe systems in the face of enormous lengths of pipe -- ie there will be manufacturing defects -- and over 50 year service lives seems like a very difficult problem to address. Also, 2/3 of their gas pipelines were built before 1970 > A Chronicle investigation pointed out in October that two-thirds of PG&E's pipelines were built before 1970. That was a key year, in which regulations requiring better welding techniques were enacted. > Also, the vast majority of PG&E's old pipelines can't be inspected with the most modern technique - a device called a smart pig, which is bristling with sensors and is run through the interior of a pipe. PG&E has also been unable to find complete documentation confirming the safety of maximum pressure levels for about one-fourth of its transmission pipelines in and around urban areas. [2] Per the pdf [3], page 18, PG&E spent $834m capex and $4.9m opex over 4 years replacing 186 mi of pipeline. This averages out to $4.4m/mi, and much of it doesn't seem to have been in urban areas, which massively increases expense and public disapproval. So you can see the magnitude of the cost to replace the (2/3) of 6700 mi of pipeline built before 1970. Roughly (2/3) x 6700 x 4.4e6 = $19.6B. None of this means PG&E is spectacularly well run. But the $100m they apparently were supposed to spend on pipe replacements is barely a drop in the bucket of doing a system-wide pipe replacement that we perhaps ought to be doing. [1] https://www.pge.com/en_US/safety/how-the-system-works/natural-gas-system-overview/natural-gas-system-overview.page https://www.pge.com/en_US/safety/how-the-system-works/natura... [2] https://www.sfgate.com/news/article/PG-E-under-spent-on-pipeline-replacement-program-2371476.php https://www.sfgate.com/news/article/PG-E-under-spent-on-pipe... [3] http://www.cpuc.ca.gov/WorkArea/DownloadAsset.aspx?id=4594 http://www.cpuc.ca.gov/WorkArea/DownloadAsset.aspx?id=4594
- iso1337 8y ago
  Great details! Makes you think that we should really be getting rid of gas pipelines and making electric-only houses. Maybe you wouldn’t even need to get rid of the pipeline during decommissioning? Just turn off the source and drain the network. Unfortunately, gas is so cheap right now and also many prefer cooking with gas.

Let's not forget about the time they blew up San Bruno: https://en.wikipedia.org/wiki/San_Bruno_pipeline_explosion https://en.wikipedia.org/wiki/San_Bruno_pipeline_explosion

x0x0 8y ago

The pipe was installed in 1956, before modern testing methods such as xrays. It worked for over 50 years. Per PG&E, they have 6,700 miles of gas distribution pipelines [1] like the ones in San Bruno. They have another 40k miles of other gas pipelines. Building perfectly safe systems in the face of enormous lengths of pipe -- ie there will be manufacturing defects -- and over 50 year service lives seems like a very difficult problem to address. Also, 2/3 of their gas pipelines were built before 1970 > A Chronicle investigation pointed out in October that two-thirds of PG&E's pipelines were built before 1970. That was a key year, in which regulations requiring better welding techniques were enacted. > Also, the vast majority of PG&E's old pipelines can't be inspected with the most modern technique - a device called a smart pig, which is bristling with sensors and is run through the interior of a pipe. PG&E has also been unable to find complete documentation confirming the safety of maximum pressure levels for about one-fourth of its transmission pipelines in and around urban areas. [2] Per the pdf [3], page 18, PG&E spent $834m capex and $4.9m opex over 4 years replacing 186 mi of pipeline. This averages out to $4.4m/mi, and much of it doesn't seem to have been in urban areas, which massively increases expense and public disapproval. So you can see the magnitude of the cost to replace the (2/3) of 6700 mi of pipeline built before 1970. Roughly (2/3) x 6700 x 4.4e6 = $19.6B. None of this means PG&E is spectacularly well run. But the $100m they apparently were supposed to spend on pipe replacements is barely a drop in the bucket of doing a system-wide pipe replacement that we perhaps ought to be doing. [1] https://www.pge.com/en_US/safety/how-the-system-works/natural-gas-system-overview/natural-gas-system-overview.page https://www.pge.com/en_US/safety/how-the-system-works/natura... [2] https://www.sfgate.com/news/article/PG-E-under-spent-on-pipeline-replacement-program-2371476.php https://www.sfgate.com/news/article/PG-E-under-spent-on-pipe... [3] http://www.cpuc.ca.gov/WorkArea/DownloadAsset.aspx?id=4594 http://www.cpuc.ca.gov/WorkArea/DownloadAsset.aspx?id=4594

iso1337 8y ago
Great details! Makes you think that we should really be getting rid of gas pipelines and making electric-only houses. Maybe you wouldn’t even need to get rid of the pipeline during decommissioning? Just turn off the source and drain the network. Unfortunately, gas is so cheap right now and also many prefer cooking with gas.