Forum Replies Created

Viewing 15 posts - 76 through 90 (of 95 total)
  • Author
    Posts
  • Brett
    Keymaster
    Post count: 101
    in reply to: soues vide how-to #3780

    have you seen this blog post?: http://www.unmaintained.com/index.php/ospid-sous-vide-open-source-high-tech-cooking-on-a-budget/

    it’s not quite step-by-step, but he does give a parts list. I might recommend using this thermisor sensor instead of a thermocouple, but other than that it’s pretty straight forward. and of course, we’ll be happy to answer any questions you have along the way. (and then turn that conversation into a proper how-to)

    Brett
    Keymaster
    Post count: 101
    in reply to: Ramp Soak and stuff #3735

    what sort of accuracy is lost by moving to integer from floating point with the PID library?

    in theory, an int and a float have the same effective resolution (4bytes,) but how you handle the int determines what resolution loss you get if any. I’ve made great strides on the firmware front however, so this conversion won’t be necessary. the code is now smaller, and uses far less ram even after adding the profile code.

    just the front-end support code and some testing, and we’ll be ready for release.

    Brett
    Keymaster
    Post count: 101

    @szeryf, in the firmware you must set usesimulation = true. this will enable the onboard simulation.

    Brett
    Keymaster
    Post count: 101
    in reply to: Ramp Soak and stuff #3733

    How’s things going with the firmware Brett

    we should be fine. the menu code is insanely bloated (classes, unnecessary sacrifices in the name of flexibility, floats.) I’m slimming that down first, in the hopes that that is all that is required. if not, then I’ll need to create an integer version of the pid library. It’s something that I’ve been meaning to do anyway, but it will take some time.

    Brett
    Keymaster
    Post count: 101

    There is an error message to do with rxtx when this happens, will try and capture it…

    Excellent. I have encountered this a couple of times myself but have never been able to reproduce. I’m assuming it’s either a ram issue on the osPID, or just a weakness in the Comm protocol I’ve devised.

    Brett
    Keymaster
    Post count: 101
    in reply to: Ramp Soak and stuff #3731

    2-3 weeks is my current plan, barring any unforeseen day-job craziness.

    I’m going to need to revise my time estimate on this front. turns out the Profile Code pushed the firmware into insufficient-ram territory. luckily the current code is pretty inefficient, so there’s plenty of gains that can be made to make room. the down side is this will mostly come from switching form float to int math, which essentially means a complete firmware re-write. fun

    Brett
    Keymaster
    Post count: 101

    when I try and click to edit a form field, it becomes a drag and droppable item!

    this is actually part of the controlp5 library. I haven’t looked into turning it off. I believe it’s alt+rightclick drag that does it

    Brett
    Keymaster
    Post count: 101

    yes it’s still a little clunky. at the very least a little “refresh ports” button would be nice. we’re working on it.

    Brett
    Keymaster
    Post count: 101

    Using Win 7 executable. Everything works ok, but cannot change any values in text fields.

    interesting. I’m developing on a windows7 box and haven’t seen anything like this. can you try this procedure and tell me if it works?

    1: click on the field (Setpoint, say)
    2: backspace to clear the value
    3: type in the new value and hit Enter
    4: click the corresponding “Send” button to get the values down into the osPID. for the setpoint this would be Send_Dash.

    there may be a big-boy-java front-end in the future. the fact of the matter is that it was far quicker to just port the existing PIDLibrary front-end, than to start over from scratch.

    Brett
    Keymaster
    Post count: 101
    in reply to: Ramp Soak and stuff #3729

    any ideas when the firmware will be out.

    2-3 weeks is my current plan, barring any unforeseen day-job craziness.

    Brett
    Keymaster
    Post count: 101

    I wrapped the thermocouple in a condom

    that is, without a doubt, the funniest thing I’ve heard all week. love the idea though. a balloon would probably work as well. nice thin layer, water pressure presses it up to the element… nice work!

    Brett
    Keymaster
    Post count: 101

    How about tool dip? although that’s pretty close to hot glue.

    for water temperature, insulating the thermocouple a little probably won’t have too much of an effect. you’re introducing a 2-5 second lag in a process that moves on the order of minutes

    Brett
    Keymaster
    Post count: 101
    in reply to: Ramp Soak and stuff #3726

    for example I might like to ramp from 32 degrees C to 47 degrees C in 45 minutes.

    ah ok, you may be able to approximate the behavior you want with the next firmware update. designed mainly for reflowing, you will be able to specify a recipe of SP ramps, steps, and/or waits that you can execute with a single command.

    Brett
    Keymaster
    Post count: 101
    in reply to: Ramp Soak and stuff #3724

    I have tried to run osPID_FrontEnd.pde from Processing but it complains about missing stuff so I will follow that up later

    ah, I forgot to answer that part. make sure you’ve installed the controlP5 library as dictated in the ReadMe.txt that’s included with the code.

    Brett
    Keymaster
    Post count: 101

    (we really need to write a manual for this thing. I’m currently deciding which wiki we should use for that)

    Dashboard > M = Mode. this can either be MAN(ual) or AUTO(matic). sometimes you want to just tell the output to be a certain value. “hold a 30% output” for example. when the pid is running it will just overwrite this value with what it thinks the output should be. switching to manual turns off the PID algorithm, allowing you to do this.

    Config > W = Output Window (in Sec). the pid says 43.5% output, but what does that mean? how does this translate into relay action? you set a time window (5sec by default.) 43.5% then means “the percent of 5 Sec that I will remain on.” the smaller this number the less bumpy the current going to your process will be, but the more cycles you will put on your relay.

    ATune has no sub options. when you click it it initiates the auto-tune and switches to the word “Cancel” which allows you to cancel the auto-tune.

Viewing 15 posts - 76 through 90 (of 95 total)