Sunday, November 27, 2011

Universal Temperature Controller (aka the sous-vide-o-tron Mark I)

So I decided to finally jump onto the sous vide bandwagon.  Admittedly, I'm a bit late, but after seeing various hacks here and there for them, I decided to give it a try and see whats what.  Now... A little background about me and food.  I can't cook.  Well, I can make eggs over easy, sometimes do pasta, and have been known to whip up a random veggie or meat dish out of sheer desperation, but on a whole I am not the most useful person to have in the kitchen (other than maybe as chief dish washer...)  I have also failed more dishes than succeeded, and I have set off the fire alarm at least once in every place I've lived which had a kitchen (this is not an exaggeration...)

So why sous vide?  Well, really the only reason is that its a good excuse to do another electronics project and have it actually do something useful.  It might also improve my cooking, or at least allow me to make things that aren't burnt/tough/dry/etc.  So after digging around the net for various plans and ideas, I settled on the simplest (and cheapest) possible setup for my Mark I build.

Parts List:
PID temperature controller (mine was a Sestos D1S from ebay) $40
PT100 temperature probe (also from ebay) $10
Solid state relay (came with the controller)
110v AC outlet and cover (Home Depot) $1
Old SCSI external hard drive case (had it lying around) free?
Misc wires, connectors, heatshrink etc (mainly Harbor Freight) $10

The Build:
I went into this wanting to build a device that wasn't just for sous vide but rather could be used as a general temperature controller.  This meant it had to be a standalone device and have temperature probe input(s) and AC output.  It had to be decently good looking (a rats nest of wires is not very appealing in the kitchen).  My first idea for an enclosure was a simple plastic box, but unfortunately I seemed to lack such a box at that time.  Plan B was to gut an old PC power supply I had lying around and use that as an enclosure, but as I was searching for a suitably old power supply to destroy I found a external SCSI hard drive enclosure I had stashed away.  I've used these in the past for various projects and they're nice because they normally have 12v and 5v power built in as well as many points to screw stuff into.  This case I had stripped bare who knows how long ago but it still had its power switch and power plug, which was all I really needed.  Success!

Assembly after that point was straight forward.  Crimp terminals onto wires, wire according to drawing on temp controller.  The area in the back of the case where the SCSI connectors originally were turned out to be the perfect size for a standard AC outlet, and the panel in the front was also a good fit for the controller (after some convincing with a rotary tool).

I had to cut up some of the drive mounts a bit as I had mounted the controller to one side of the panel and the screw terminals on the back were blocked.

Finally, I decided that instead of having a temperature probe permanently connected, I would rewire it to a phone jack and make that modular as well.  This will be beneficial later if I ever decide to change to a different thermocouple.

The female end of the phone jack is mounted where the fan used to be (its actually glued to a chunk of the front panel I cut off in order to install the controller.  Reuse is AWESOME.)

The finished product:

I still lack a proper sous vide vessel.  That being said, a crockpot is decent (and cheap) substitute assuming you pay attention to some details.  First, a crockpot was not meant for quick heating or cooling.  This means that in order to not wait hours for your water to heat up you should probably boil some on the stove and add it to the pot, then mix in some cold water to arrive at a temp close to what you need to cook at.

Another issue is that the default PID values probably will not work very well for such a slow heating/cooling vessel.  My first temperature test the water was consistently 4C over and could not figure out why.  Then I realized it had an autoadjustment routine so I poured in some more water and let that run overnight.  After that it has been +/-0.5C.

First cooking subject was an egg at 63c for an hour.  The results were quite surprising.  The yolk had a consistency of custard while the whites were still pretty runny.  This was mainly due to me accidentally cracking the egg when I dropped it into the pot, but nonetheless the product tasted good.

Second cooking subject was chicken at 61c for two hours.  This was the real test.  I purposely made the chicken as simple as possible with just some salt and pepper rub before sticking into the zip-loc bag and vacuum sealing it (zip-loc vacuum seals are a cheap and effective alternative to a full blown vacuum sealer).  Once they cooked for a couple hours I gave them a sear in some olive oil and they were ready to serve.  The results were very successful, to say the least.  The chicken was moist all the way through and tasted like, well, like perfectly cooked chicken.  Nothing fancy, just chicken done right.

Next Steps:
The project is currently in a working state, but I have many improvements I would like to make to it.  The first thing would be a proper heating vessel.  I'm debating between going either with a coffee urn or a cooler and some immersion heaters.  Cost wise the cooler would be slightly more expensive, but it would also be easier to clean and maintain, as well as having a much larger volume in which to cook in.  On the other hand a coffee urn would blend in to a normal kitchen very well and at the end of the day you still have a working water boiler for coffee or tea.
Being that this controller is pretty universal I'm already looking into building a smoker out of a weber I have out back and a hot plate.  I do have an extra k type thermocouple which came with the controller and that would work just fine as the temp probe in a smoker.

I'll keep this blog updated with further cooking results and other random hacks I do.

No comments:

Post a Comment