Testing An Open Source PID Controller With Steak

Sous vide cookers aren’t anything new, but [Phil] wanted to build the first sous vide using the osPID, an open source PID controller just released in the last month.

The build uses the osPID Open Source PID controller we saw last week that comes with inputs for a thermocouple and pair of relays capable of switching a hot plate or immersion heater. The osPID is based on the Arduino and was created by [Brett Beauregard], author of the Arduino PID library.

Getting to the meat and potatoes of the build, [Phil] connected a 300 Watt immersion heater to the osPID and put the heater in a bowl of water. A delicious looking cut of beef tenderloin was shoved into a ziplock bag and suspended in the bowl of warm water for a few hours. With the heater and thermocouple attached to the osPID, the temperature was set at 130° F and the entire device was left alone for a few hours. Looking at [Phil]’s recipe for tenderloin with lemon parsley butter whets our taste buds, so we’ll hope [Phil]’s dinner came out just fine.

20 thoughts on “Testing An Open Source PID Controller With Steak

  1. People think I’m anal enough already about following recipes. You know, measuring *exactly* 2.5 cups of water, etc. I can just see their reaction when I start complaining about “not following the temperature profile close enough.”

    1. Actually sous vide is about simplifying the cooking procedure.

      The idea of the PID controller is to keep the water bath at a very exact temperature and hold it there. No strange ramps up and down or any exact programming. Once the food is at the target temperature for enough time that is needed to be safe, it can stay there for hours (or even days, for some recipes).

      1. They seem good enough to me; different sensors, autotune, C/F, dual alarms, cool/heat etc. I’ve got heaps of them, they work fine.

        The problem is you don’t know exactly what you are getting, I’ve yet to find a cheap one that does bands (heats and/or cools to keep with a temperature range, say 20-30C).

        The only real difference between the cheap units is firmware, and even two units with exactly the same specification code will have different features.

        The cheap units are also relay only (and only 3A at that), but hacking in a better relay or a SSR isn’t hard.

        For a sous vide cooker I’d just cough up the $20 (or less if you look around, eg DX).

        A DIY unit that does rank/soak with comms and a fancy display – now that’s newsworthy (not to detract from the nice job done).

      2. yeah that’s a fair assessment. I guess in your case the extra money is for openness, support, and that front-end. if you don’t need that stuff the osPID isn’t for you.

        it’s conceivable that we can push down the price over time, but we busted our asses to get it down to 85. Standard kit multiplier and a 100-unit run, there’s only so much you can do.

        1. that’s an interesting thought, but I feel that would be like swatting a fly with a buick. also, there’s no way I could get those parts at their cost. add the UI and IO elements and we’re talking quite a bit more than $85.

          perhaps down the road we’ll release a “mega” version or something, but for this iteration the idea was to beat the functionality of a $200 pid controller, which I feel that we have.

          as a side note, the Arduino PID Library has been exhaustively explained and there’s been talk of porting it to java etc. if that happens there’s no reason someone else couldn’t do precisely what you describe.

      3. osPId is a bit of overkill for controlling a cooker, unless as a learning exercise. Most people only need a basic PID, so for under $20 buy makes more sense than build especially if time is short.

        10 years ago basic PID’s way more than $20, so DIY made more sense then. I built a couple.

        Fancy PIDs are still expensive, so I’m still interested.

    1. The MAX6675, although it’s pretty pricey, does get the job done for most temperature PID applications such as for cooking or espresso machine control. When you’re pulsing a heater over a fairly long period of time, you really don’t need to be sampling at frequencies of over 4hz.

    2. The fact he is using a 300W heater in a bowl holding at least 2 liters of water (+ food) means that in 1/4 of a second, the temperature can increase by a maximum of no more than 1 hundredth of a degree C.

      I think the reason you are laughing is because you don’t know what the hell you are talking about.

  2. osPID is ideal if you need ramps.
    Just holding a set temp is an option that can be solved in other ways.
    But as a ramp controller its just as good as an omron E5CK (once they get an input card that can handle other thermocouples than K.)
    Yay osPID !!

  3. What about using a very simple 2 stage controller?

    Simply add a second mains input via a light bulb as a dropper, have your controller switch mains or mains+bulb to the element.

    Plus you have visual indication of the cooking process which is useful.

Leave a Reply

Please be kind and respectful to help make the comments section excellent. (Comment Policy)

This site uses Akismet to reduce spam. Learn how your comment data is processed.