How to best analyze sous vide timing data?

Discussion in 'Physics & Math' started by Jennifer Murphy, Jun 22, 2018.

  1. Jennifer Murphy Registered Senior Member

    Messages:
    182
    My son, who is into cooking, told me about his sous vide device. For those who are like I was and have no idea what sous vide is, It's a method of cooking food under water in a vacuum-sealed bag. Sous vide means "under vacuum" in French.

    It's a big deal because it makes it possible to cook almost anything to an exact temperature and hold it there for fairly long period of time (1-2 hours). If you want a steak medium rare, for example, set the sous vide unit to ~132°. The steak will be medium rare in about 45 minutes and will stay that way for 1-3 hours.

    In any case, I bought one and tried it out. What I immediately noticed is that it heats the water fairly quickly until it gets within a few degrees of the setting and then it goes very slowly. So I ran a few tests. Using the same pot and amount of water, I started the unit at one of several initial temperatures and timed how long it took to get to a target temperature depending on how much over the target temperature I set the unit.

    Here is the data so far:
    Code:
                                 min
    Start  End  Setting  m:ss    less
     80°   100°   100°    8:35   0.00
     80°   100°   105°    4:27   4.13
     80°   100°   110°    4:53   3.70
    
    100°   120°   120°    9:06   0.00
    100°   120°   125°    5:02   4.07
    100°   120°   130°    5:26   3.67
    
    120°   140°   140°   12:02   0.00
    120°   140°   145°    5:26   6.60
    120°   140°   150°    5:56   6.10
    
    140°   160°   160°   10:42   0.00
    140°   160°   165°    6:30   4.20
    140°   160°   170°    6:57   3.75
    
    160°   180°   180°   14:40   0.00
    160°   180°   185°    8:53   5.78
    160°   180°   190°    8:58   5.70
    
    My question is how to best present the data.

    Just by eyeball, it looks like the unit heats at full speed until it gets within 5° or less and then it slows down. I've tested by setting the target temperature at +0°, +5°, and +10°. My next test will be at +3°. If that is similar to +5°, I'll try +2°.
     
  2. Google AdSense Guest Advertisement



    to hide all adverts.
  3. Michael 345 In Aust : found it :) Valued Senior Member

    Messages:
    5,661
    I would pick a mean temperature and go with a graduated colour pie chart

    Please Register or Log in to view the hidden image!

     
  4. Google AdSense Guest Advertisement



    to hide all adverts.
  5. origin In a democracy you deserve the leaders you elect. Valued Senior Member

    Messages:
    10,343
    Umm-um, boiled steak!

    Please Register or Log in to view the hidden image!



    In all seriousness that sounds interesting.
     
  6. Google AdSense Guest Advertisement



    to hide all adverts.
  7. Dr_Toad It's green! Valued Senior Member

    Messages:
    2,161
    I've thought about buying one. With that said, for steaks you cook to the desired internal temp, then sear it for proper color and plate appeal.

    No more overcooked steak!
     
  8. Jennifer Murphy Registered Senior Member

    Messages:
    182

    Please Register or Log in to view the hidden image!

    I'm not sure what you mean by that. I plotted the time it took the unit to heat the water 20° starting at 80°, 100°, 120°, 140°, and 160° (dependent variable) against the number of degrees over the target (ending) temperature I set the unit to. I've attached that graph. It looks like the unit starts slowing the heating process when the water gets within 5° of the target temperature.
     
  9. Jennifer Murphy Registered Senior Member

    Messages:
    182
    The food is not in contact with the water, as my original post made clear.
     
  10. exchemist Valued Senior Member

    Messages:
    7,226
    This seems a very odd way to plot what looks perfectly simple, unless I'm missing something.

    Surely the most obvious and informative way to do this is to plot temp as ordinate and time as abscissa and then see how the gradient changes as the target temperature is approached? You should get a steep rise from left to right and then the gradient will become more shallow as the target temp is approached. You can indicate the target value with a dashed line or something. Or have you some special reason for not doing that?

    Also, the way you label it, the thing is over the target temperature for some reason. Yet your description of how it works is more intuitive, in that it heats rapidly until it gets to a few deg below the target temp and then heats slowly until it gets to the target, presumably to avoid overshooting. Do you mean this? Does it do that, or does it overshoot and then come back?
     
  11. Jennifer Murphy Registered Senior Member

    Messages:
    182
    I am not interested in how long it takes it to get from T1 to T2 in the absolute. I am interested in how long it takes to get from T1 to T2 if I set it to exactly T2 as compared to if I set it to T2+1, T2+2, T2+3, etc. If I am correct that the device turns down the heat as the water gets close to the target temp, then it ought to get there more quickly if I set the unit to +1° than if I set it to +0°, and more quickly still if I set it to +2°. So I ran a test where I started it out at 80° and timed how long it took to increase in 20° increments to 100°, 120°, 140°, 160°, and 180° with the unit set at the target temp. Then I repeated that but setting the unit at the target temp +1°. And so on. I have taken a couple more readings and improved the labels on the graph. See below.

    Please Register or Log in to view the hidden image!


    It seems pretty clear that the unit starts turning down the heat a bit when the water gets to the target temp minus 5° and then it turns it down a lot more when it gets to within 1° of the target.

    This information is useful for a couple of reasons.
    1. Setting the unit to a few degrees over the target temperature will save several minutes on the time it takes to get there.
    2. When the food to be cooked is added, it usually lowers the water temperature, so having it a few degrees warmer will offset some of that. Then the unit can be set to the true target, which it will maintain.
    Cheers
     
  12. CptBork Robbing the Shalebridge Cradle Valued Senior Member

    Messages:
    5,682
    You could try starting with hot water off the get-go and let the sous-vide device heat it up the last little bit, maybe? Which type of unit does your son use, has he got one of those new Joule thingies? I like doing my steaks via the standard reverse sear method, they call it the poor man's sous-vide, comes out amazing.

    Edit: Oh and also, I doubt it makes any noticeable difference if you're in a rush and you toss your food bags into the water before it's been heated by the last few degrees.
     
    Last edited: Jun 24, 2018
    Jennifer Murphy likes this.
  13. Jennifer Murphy Registered Senior Member

    Messages:
    182
    That was one of the purpose of the timings -- to see if it was worth doing that. My conclusion is probably not for lower temps (up to ~140°).

    The Anova.

    I don't eat steaks. So far, I've used it for fish, chicken, and boiled eggs.

    Yes, probably so. On the other hand, my data suggests that setting the device a few degrees high will get it to the target temp more quickly and it will probably be closer to the right temp after the bags have been added, which lower the water temp a bit.

    Here's my final data. It looks like the unit starts turning down the heat when it gets within about 5° of the target temp and turns it way down when it gets within 1°.

    Please Register or Log in to view the hidden image!

     
  14. Jennifer Murphy Registered Senior Member

    Messages:
    182
    PS: I also did two additional timings to see if it makes any difference whether the tub is covered. It doesn't unless the target temp is over 160° or so.

    Here's that data:

    Please Register or Log in to view the hidden image!

     

Share This Page