Robots inspect nuclear waste canisters for damage at San Onofre
A clunky robotic device resembling a tiny car — equipped with a 3-D camera — has been crawling up and down canisters loaded with nuclear waste at San Onofre’s concrete bunkers, trying to quantify the depth and breadth of scratches inflicted on their way into the vaults.
An inspection system for the Holtec Hi-Storm UMAX wasn’t supposed to be ready for years, but necessity is often the mother of invention. The Nuclear Regulatory Commission refuses to allow Southern California Edison to move nuclear waste from wet to dry storage until questions about the seriousness of scratches on the canisters is addressed, said Scott Morris, the NRC’s Region IV administrator.
“Edison took dozens of corrective actions. They did a really good job with most of it — but not all of it,” Morris said in a sit-down with the Southern California News Group on Thursday, March 28.
For example, new load sensors — which will trigger an alarm if there’s a sudden loss of weight as canisters are being lowered into the bunker — hadn’t been adequately tested, said Linda Howell, deputy director of the NRC’s Division of Nuclear Materials Safety.
But the big issue for the NRC right now is the scratches, officials said.
Some scratching allowed
The Holtec UMAX’s final safety analysis report says, ” ‘There will not be scratches on the canisters.’ Period. End of story, ” Morris said.
But the license issued to Edison, which allows it to use the Hotec UMAX, says the system must be in compliance with American Society of Mechanical Engineers’ specifications — which do allow a certain amount of scratching.
“Clearly, there’s an inconsistency between these two documents that has to be rectified,” Morris said. “Edison has to resolve this. But you can’t just say, ‘It’s OK to have scratches.’ There has to be a technical basis for it. You have to have an analysis that says why it is OK. Give me the calculations. Give me the data.”
At first, Edison and Holtec used calculations to try to prove the scratches were minor. The NRC wasn’t buying.
Then they tried a controlled experiment involving a stress test on the metal of the canister shells — but this suggested their original calculations contained errors.
“How do you know what you have scratched?” Howell asked. “You have to actually look. Actually look at an actual canister, instead of at these surrogates.”
90 percent of canisters inspected
And so, last week, Edison started actually looking.
Using technology now employed to examine jet turbine blades for cracking, the robot crawled into the storage vaults and was able to examine 90 to 95 percent of canister exteriors, Howell said.
It inspected three of the 29 canisters already loaded into the dry storage bunker: the one involved in the infamous near-drop last Aug. 3, the one that had alignment troubles during loading on July 22, and another, random canister.
The 3-D camera system can see scratches down to one-thousandth of an inch, and has been able to document scratches that are about 26/1000 of an inch on canisters, Morris and Howell said. If those are the deepest scratches, they are well within tolerance, they said.
Scratches are no more than the thickness of a credit card, said Edison’s Tom Palmisano at the San Onofre Community Engagement Panel meeting Thursday, March 28. The oxide layer on the exterior of the canisters reforms quickly, he said, so there’s no risk from corrosion in the short or long term.
The NRC expects Edison to submit its data shortly. The NRC then will decide if Edison’s work and analysis are satisfactory — and if it wants more canisters to be inspected before allowing Edison to resume moving nuclear waste from the wet pools where it now cools, and into the dry storage system.
There have been 29 canisters loaded into the Holtec Hi Storm UMAX system so far, and 43 more to go.
Contractor Holtec began transferring waste from wet to dry storage last year, but hiccups quickly tripped up progress.
Workers were preparing a canister for loading in February 2018 when they discovered a loose, stainless-steel bolt inside, about 4 inches long. An investigation revealed that Holtec had altered the canister design without permission from the NRC.
On July 22, workers had difficulties centering and aligning a canister as it was being lowered into a vault. The canister did not get stuck, though, and the incident did not get entered into the plant’s “corrective action program” — a system designed to catch and learn from mishaps.
But on Aug. 3, there was a more serious misalignment problem. A 50-ton canister filled with nuclear waste got stuck on a shield ring near the top of the 18-foot-deep vault where it was to be entombed. Workers didn’t realize that the slings supporting the canister’s massive weight had gone slack. It hung there, unsupported, for close to an hour, in danger of dropping.
Edison failed to realize the seriousness of the event, which happened on a Friday, and didn’t notify the NRC until the following Monday, the commission said.
Transfers screeched to a halt as the NRC and Edison tried to unravel what happened, and how to stop it from happening again. In November, the NRC laid blame squarely at Edison’s feet, saying it “fell asleep at the switch” and concluded that the near-drop was the result of inadequate training, oversight and supervision.
Since then, Edison has adopted many new checks and balances that will prevent the errors of the past from repeating themselves, officials said.
Cameras will watch as the canisters descend into the storage vaults. Alarms will go off if there’s a sudden, significant change in the weight supported by the canister-lowering machinery. The number of people at the storage pad observing the loading of each canister will increase. Workers at all levels have been more rigorously trained and personnel changes have been made at the top and down the chain of command.