Actions
Forum Thread Details
  • Replies 8 replies
  • Subscribers 591 subscribers
  • Views 318 views
  • Users 0 members are here
Related

Unexpected PID Controller Explanation - In a video about... a moka pot?

stanto

I like coffee.

I find James Hoffmann crosses over into my interests pretty well as a person on youtube, and surprisingly, ends up crossing over into the educational realm, and so, an impromptu introduction about PID controllers.

Worth a watch:

One of the first 'IoT' devices was a controlled coffee pot, watched by a webcam or something. Probably controlled over RS232, one that now monitors the pressure and temperature too? That feels like a level-up.

Kinda want one. It doesn't already exist, right?

Parents

  • Interesting! I'm team "keep the original, real deal", for several practical reasons. In addition, the coffee is already at a high bar with a Bialetti, versus your average coffee shop. But, on the cup-is-half-full side, to me it seems the value of the mokabot is more in allowing people to experiment with example control systems and tweak to their hearts content; It's fun having a little testbed for a real-life scenario when learning; could be very attractive in universities for instance, although they may prefer to just simulate since that's easier and doesn't take up lab time.

    In a similar vein, me and Jan played around with PID practically, not too long ago (there were lots of blogs but I'm mobile, harder to dig them up), we used different devices and masses in our testbeds, in my case the testbed target was an aluminium block of a particular size, and I found PID alone didn't control things well enough; together with ChatGPT a load of modifications were made which I didn't understand, but resulted in very precise temperature, but only for similar ballpark aluminium masses, otherwise things need to be tweaked again.

Reply

  • Interesting! I'm team "keep the original, real deal", for several practical reasons. In addition, the coffee is already at a high bar with a Bialetti, versus your average coffee shop. But, on the cup-is-half-full side, to me it seems the value of the mokabot is more in allowing people to experiment with example control systems and tweak to their hearts content; It's fun having a little testbed for a real-life scenario when learning; could be very attractive in universities for instance, although they may prefer to just simulate since that's easier and doesn't take up lab time.

    In a similar vein, me and Jan played around with PID practically, not too long ago (there were lots of blogs but I'm mobile, harder to dig them up), we used different devices and masses in our testbeds, in my case the testbed target was an aluminium block of a particular size, and I found PID alone didn't control things well enough; together with ChatGPT a load of modifications were made which I didn't understand, but resulted in very precise temperature, but only for similar ballpark aluminium masses, otherwise things need to be tweaked again.

Children
No Data