Hacker News new | ask | show | jobs
by swamp40 4096 days ago
Not sure I'd want that thing sitting on my stove overnight.

If you turn the knobs on my stove too fast, it skips the piezo ignition and just starts spewing gas out.

Wouldn't surprise me if this safety issue has been what has kept other companies out of this space.

But congratulations on their one day goal achievement.

If I may suggest an improvement, how about a manual "off" switch so people can sleep more easily?

And maybe you could add a gas sensor to detect any problems, and try to turn the knob off? And a loud alarm if that failed?

That could turn the product from a liability into an asset.

1 comments

They require you to turn the stove ON first, and then take over from there, so it bypasses the ignition skipping and "auto on" concern.
Yes, I saw that.

But what if that $2 processor makes a poor decision, rotates that knob on accident and kills your family?

I totally understand your concern. That's why we NEVER turn on the stove for you. All modern stoves require you to push the knob in to move it from the off position. The Meld Knob is mechanically incapable of defeating this safety interlock.
Thank you for explaining. Is it possible to misadjust the collar when installing, so as to accidentally defeat the safety interlock?

For instance, if you push in, then tighten down?

I'm an engineer so when people ask if something is possible my honest answer has to be that anything is possible.

Would you allow me to rephrase your question? I'd rather answer the question, "What is the likelihood that the knob can be installed incorrectly or that the mechanical interlock will fail in a way that would allow the knob to unexpectedly turn the burner on when not intended?"

My answer to this question is that it would be EXTREMELY improbable that this would happen. I say this for the following reasons. 1) The part of the knob that allows the user to push the shaft inward to override the safety interlock is a completely separate piece from the part that turns the knob; 2) Both of these pieces are made of smooth, lubricious plastic such that it should be very difficult to install the knob in a way that they would bind; 3) assuming the user did install the knob in a way that made this possible it would be very obvious to the user that things are not right; 4) the knob software is written in a way that the knob will refuse to move from the off position without physical intervention from the user.

What I've described is generally called a defense in depth approach. You put a bunch of safety mechanisms in place such that all of which have to simultaneously for a bad event to occur.

Hopefully that helps.

Thanks, JJ. Good luck, and congratulations on your day 1 funding.