Custom HVAC Car Project

Discussion in 'General Off-Topic' started by Michaelflat, Jun 2, 2019.

  1. Michaelflat

    Michaelflat
    Expand Collapse

    Joined:
    Jul 10, 2014
    Messages:
    1,543
    So i think i will be driving a Honda Fit/Jazz for a first car, and well it has manual aircon controls.. (all mechanical)..

    I want to use an Arduino (or raspberry Pi, not sure yet, but i'm leaning towards Arduino.. but i have experience with python, i could learn arduino i guess..).. But to control it i will need servos to control the heater flap and etc.. Will this be an absolute nightmare? Or do you think this will be feasile.. (only controlling heater flap nothing else yet.. semi-automatic..).

    I've got the control logic thought out, and will be able to have what i feel is a fully functional system with 2-3 sensors (2 temperature+humidity sensors and 1 photo sensor but i might forgo the photo sensor as it might be a hassle). (for external temperature i might add another sensor, or see if i can get weather data to the arduino!, that'd please the exam board also :p )

    Soo.. would it be a huge hassle or not? Relay for aircon control sounds nice and easy, but servo for heater flap sounds a bit nightmare-ish..
    --- Post updated ---
    *edit*i think i might need to get photosensor and outside air sensors as due to the Arduinos limited ram, you can't parse large amounts of weather data... idk though, i was planning on getting forecast data for each hour for 24hrs.. but i digress it's probably best not to do that approach anyway, best to use sensors!

    3 DHT22s then, and one photo sensor.. The little arduino will be working it's socks off :p

    DHT22 list:
    Intake air
    Supply air
    internal air

    Also think i will do it on Arduino, Pi is not suited for this..

    So yeah, any annoyances, or ideas for a climate control system for car hvac, reply here :)
     
    #1 Michaelflat, Jun 2, 2019
    Last edited: Jun 4, 2019
  2. SixSixSevenSeven

    SixSixSevenSeven
    Expand Collapse

    Joined:
    Sep 13, 2013
    Messages:
    6,960
    Very doable. the Arduino is more than capable for this project. It won't even be working its socks off.

    Only things the pi gets you are python and processing power for the web. Also the pi cannot *reliably* read dht22s due to Linux multitasking interfering with gpio timing. The pi cannot read the photosensor as it has no analogue inputs. The pi cannot reliably operate servos due to Linux multitasking interfering with gpio timing (it'll work but the servo jitters, pi goes to send a 1.5uS pulse but gets interrupted mid way and really sends 1.51, then next time 1.58, 1.52).

    Common workaround, and also a nice but not necessary demo. The pi has SPI, uart(serial) and i2c. An Arduino can operate with all of those. i2c and serial in python on a pi are super easy. An Arduino is really good at controlling servos, has analog inputs, can interface with your temperature sensors. You can have one as a slave from the pi, but in your case, I just don't think you need the pi at all
     
    • Agree Agree x 1
  3. Michaelflat

    Michaelflat
    Expand Collapse

    Joined:
    Jul 10, 2014
    Messages:
    1,543
    yeah arduino is getting more do-able, and yeah it's too much of a faff having a full computer there, like you said the scheduler will mess the gpio up, and also the bootup time of the pi is too high, arduino is more suited for the application..

    Also looking at the requirements for the coursework means i probably won't do it, rather a javascript game with animations and peer-peer multiplayer (cs teacher also recommended that)

    But i will think about doing this project still, might do it anyway just out of fun..

    Thanks for reply :)
     
  1. This site uses cookies to help personalise content, tailor your experience and to keep you logged in if you register.
    By continuing to use this site, you are consenting to our use of cookies.
    Dismiss Notice