←back to thread

The Therac-25 Incident (2021)

(thedailywtf.com)
449 points lemper | 3 comments | 27 Aug 25 06:57 UTC | HN request time: 0s | source
Show context
zackmorris ◴[27 Aug 25 16:59 UTC] No.45042158[source]▶
Our power went off a couple off weeks ago due to wind probably knocking a branch into a power line. Now our Frigidaire microwave runs with the door open.

Supposedly there are mechanical switches that prevent that, but evidently "modern" microwaves can control the gun through the logic board.

The engineering failures that led to this, from conceptual to design to internal control, boggle my mind. I'm not even sure where to send a complaint or if it would result in any kind of compensation. Because billion dollar corporations know that they'll never have to face any kind of corporate death penalty because they're protected by limited liability. So we'll just buy another $150 microwave instead.

Are smaller companies better at engineering safety? Evidently not.

replies(2): >>45042535 #>>45044489 #
bongodongobob ◴[27 Aug 25 17:34 UTC] No.45042535[source]▶
I have a microwave from the early 80s. If I stick a pencil in the door latch I can get it to run with the door open as well. It's not the demon core. Just don't stick your head in it when it's running.
replies(1): >>45047993 #
1. simoncion ◴[28 Aug 25 03:19 UTC] No.45047993[source]▶
> If I stick a pencil in the door latch I can get it to run with the door open as well.

"The safety interlocks don't work when the operator intentionally goes out of his way to defeat them." isn't a concern. There's only so much you can do to prevent someone who's dedicated to disabling them.

"The safety interlocks fail dangerous because of an unexpected power cut." is a huge concern. What else did the manufacturer skimp on, or -worse- simply fail to understand was important to do for the safety of the operator of the device?

replies(1): >>45048313 #
2. bongodongobob ◴[28 Aug 25 04:13 UTC] No.45048313[source]▶
It's crazy to me that you are ruling out a power surge frying a board. Same thing could happen to the 80s model as well. You have not root caused it and are making up a failure mode that fits your point. Hell, the hall sensor could be fried and that's pretty damn mechanical. Again, your microwave isn't a demon core. Inverse square law applies. Don't put any limbs inside when it's on, and it really isn't that dangerous, so I'm not shocked they didn't apply aircraft safety design rules.
replies(1): >>45048477 #
3. simoncion ◴[28 Aug 25 04:47 UTC] No.45048477[source]▶
> It's crazy to me that you are ruling out a power surge frying a board. Same thing could happen to the 80s model as well.

I'm taking OP at their word, in part because I don't feel like doing free-of-charge remote diagnostics on a microwave over a communications channel with such an absurdly high RTT. Based on your poor choice of counterexample, I was assuming that you didn't understand the difference between "as safe as we can reasonably make it" design and "it only appears to be safe" design (which is the worst kind).

> Don't put any limbs inside when it's on, and it really isn't that dangerous...

If true, then I'd wonder why they bother with any shielding at all. They'd save a bunch of money by putting a clear glass or plastic plate in front of the door, rather than all that metal, don'tcha think?

> Again, your microwave isn't a demon core.

Lots and lots of dangerous things are less dangerous than plutonium cores that are just raring to fizz a little. That doesn't mean that the safety mechanisms mandated to be incorporated in their design are obviously superfluous.