Hundreds of drunk-driving convictions in San Francisco could be thrown out because of questions about how police handled the device that measures blood alcohol levels, say prosecutors and the public defender's office, the San Francisco Chronicle reports. The problem could affect as many as 1,000 convictions, said Public Defender Jeff Adachi, whose office is working with District Attorney George Gascón's prosecutors to identify guilty verdicts that may be in jeopardy.
At issue is how the Police Department conducted accuracy checks for preliminary alcohol screening devices, which officers use to determine whether a drunken-driving suspect's blood alcohol level is above 0.08 percent, the legal limit for intoxication. Suspects exhale into the device to produce a reading. The manufacturer of the Alco-Sensor IV devices says police should conduct accuracy checks every 10 days, or after 150 tests, by using the device to measure the alcohol level in a gas canister. The canister's alcohol level is always the same - 0.082 percent. The public defender's office noticed that in police logs on the accuracy tests dating back to 2010, the devices were listed as reading the canister gas exactly correct every time. At least some devices' readings should have been incorrect, Adachi and Gascón said.
In January, several attorneys for the public defender's office noticed that in police logs on the accuracy tests dating back to 2010, the devices were listed as reading the canister gas exactly correct every time, Adachi said at a press conference Monday.
In practice, at least some devices' readings should have been incorrect, Adachi and Gascón said. When the readings are wrong, the machines have to be recalibrated so they produce accurate results.
"It would be mathematically impossible for that to occur," Adachi said of the consistently on-the-button accuracy checks. "The results that we have here plainly show that the accuracy testing was not being done."