PID temperature controller, works!
Hello,
Thought I'd post this in case anyone else is interested. I got this handy-dandy "PID" controller, it adjusts electrical output to a target heating (or cooling) device based on realtime feedback from a temperature probe. It took me a day to figure it out and get it set correctly, but it is now calibrated for the heating device I am using. For me, the controller only takes a few minutes to go from a room temp to the set-temp, and is accurate to appx +/- 0.5 degrees of the target temp. Most of the time it is spot-on. It has a nice feature of defining schedules in advance with 6 steps max. For example, if you want something to be held at 50C for 2 hours, then at 60C for 2 hours, you can pre-program it and walk away. Each step has a max of 9999 minutes. It has many ways to configure it, you can either set the P. I. D. values directly, or use a "Auto-Tune" feature (automatic calibration). Ultimately, I had to lower the max electrical output by 80% to my heater, effectively turning my 1000-watt burner into a 200-watt burner, then used the auto-tune, and it calibrated itself perfectly.
The controller was inexpensive, less than $200 USD. My heater is a cheap $15 single-coil burner from the local drugstore.
Here it is, holding a mash steady at 60C. (60.2C at the time I took the photo). The tip of the probe is metal (stainless steel I think..) and is submerged entirely in the mash.
This is a similar one, less features but only $50 that i used to control my homebrew refrigerator - can be used to control a proofing box too...
http://www.amazon.com/Refrigerator-Freezer-Thermostat-Temperature-Controller/dp/B004B4HAPO
Looks handy, but the amazon link says its range is:
RANGE: 30° F to 110° F (+ - 2° accuracy)
So, at least for mash-making purposes, I'm not sure that would work, since you usually want higher temps than 110F. But you're right for basic cooling or proofing the less-expensive Amazon model would work fine.
Thats great! Now, how did the mash experiments turn out? You need a way of measuring the available sugars after holding the mash at different temps. It's called a refractometer, I think, and it measure the amount of sugar in brix.
Syd
Hi Syd,
I have not had time yet to go through all of the mash experiments, it will take some time. I am using my taste buds as my refractometer.
But, since you asked, I tried this yesterday with some white-flour (KA AP). After the initial mash was done, it tasted good, but after the final mash, it was awesome.
Initial mash:
3 hrs @ 63C
1 hr @ 70C
Then took half of that mash, added more flour+water to it, and did a 2nd mash:
3 hrs @ 40C
2 hrs @ 60C
2 hrs @ 70C
---
Anyway, I have ordered a few books on mash-making and brewing. I don't plan to brew but trying to learn more about enzyme action. It is amazing to me what you can do with just flour, water, and temperature. :-)
I am currently in the process of trying out a Rye mash on the following schedule -- a bit simpler.
3 hrs @ 40C
3 hrs @ 60C
3 hrs @ 70C
Then take out half, add more flour+water, and repeat the above.
My goal isn't to create a highly fermentable liquid extract for brewing (containing primarily maltose sugar). My goal is to maximize the level of glucose, which is sweeter than maltose. There is an interesting thing with Maltase enzyme... but it is only active at lower temperatures.
No doubt I am doing things wrong, wrong, wrong .... which is fine with me. :)
In any case, it is fun to try new things.
Very interesting stuff gvz. These are experiments I have wanted to do myself for a long time. Like you, I am not a brewer (although that does appeal to me, too) but I am interested in mashing only insofar as it affects my bread making.
A question: that initial 63C...did you bring that up slowly from room temp or did you scald it with hot water, thereby gelatinising the starch immediately? If you brought it up from room temp you would have probably had some beta-amylase activity, albeit very minimal, but if you scalded it, then you would have de-activated the beta-amylase almost immediately.
However, once you added more flour and water to that first mash and maintained it at 40C, you would very likely had a fair amount of beta-amylase activity. Beta-amylase cannot convert undamaged starch grains. It has to rely on alpha-amylase degrading them first or find mechanically damaged grains or gelatinised starch. You had plenty of gelatinised starch and therefore plenty of food. Even alpha-amylase, as I understand it, doesn't do a very good job of converting undamaged grains to sugar and much prefers damaged grains.
I do know that scalding or mashing too much of the flour is not a good thing for bread making. It denatures the gluten and bread volume suffers. A little, on the other hand, provides food for the yeast and also acts as a natural anti-staling agent. The magic no. it seems is around 5 or 6% of the total flour in the recipe.
You mashed that for a long time: a total of 11 hours. At what point have the amylases done their job? There must be a limited amount of food for conversion into sugars and at some point it will all be exhausted. It would be a waste of time and energy to keep it heated beyond that point. But I guess these questions can only be answered if you have means of measuring available sugars. I am sure this kind of research has been done but it is going to vary from flour to flour. It would be helpful to get this kind of information for the flours you used on a regular basis.
Syd
Hi Syd,
I have been experimenting by starting out at room temps. The mash/PID controller only takes a few mins to bring to a target temp, whether it is 40/60/70C.
Well, my initial white-mash was great. Then I did an initial Rye mash, which was incredible. I could have just eaten it raw for breakfast or something, it was that tasty.
Then I took half out, mixed in more Rye and water, and re-cooked it a second time for another 9 hrs / overnight on a 3/3/3 schedule of 40/60/70C. This morning, bleary-eyed, I woke up and just tasted it without smelling it first. Yuck. Spoiled. It had that leuconostoc smell to it. Reminds me of when I once tried a warm soaker for a long time and grew a bunch of nasties. Ugh.
Well, live and learn. :-)
I'm hoping some of the books I ordered on mash making will give insight on optimal durations.
Let me know if my current understanding (below) is off the mark:
The item from Amazon provides an ON/OFF signal. As such it works with traditional refrigerator compressor motors (and also with other kinds of heating and cooling equipment). It has less finesse, possibly overshooting the desired temperature noticeably the first time, and possibly cycling back and forth from several degrees above the desired temperature to several degrees below it. Virtually all existing oven and refrigerator controls are like this; the relative lack of finesse isn't large enough to cause problems.
A PID (Proportional-Integral-Derivative) controller provides a continuous signal -stronger whenever more is needed and weaker whenever less is needed. It's very smart about its job if properly set, coming to temperature quickly yet not overshooting, and holding the desired temperature very closely. Its continuous signal works with heating elements for things like mashes and proofing chambers, and probably also can be made to work with the solid-state cooling elements used in some wine coolers. A traditional refrigerator would not work with a PID controller; if for example the intention was to at the moment provide 2/3 the cooling, so the motor was sent 80 volts, it would freak out over "another brownout", and if not protected might even burn out completely.
Thanks Chuck. You described it perfectly.
For cooling apps, this unit has a desensitation thing which is intended to prevent too much of a rapid on/off cycle which might damage mechanical compressors. So you can adjust it.
I haven't posted about this before here but I have a PID I have been using for cooking Souis vid. Loosely translated souis vid means "under vacuum" I believe. Basically you first season and seal the product in a plastic bag, place the bag in the water, bring water up to the desired final temperature of the item and hold that temperature until the entire piece is stable. As stated above, the PID controls the temperature to within .5 degrees or closer. For example, I place a steak in a vac-u-sealed bag and hold it at 140F for about 1 hour but longer if necessary due to timing. The meat is now a perfect medium rare edge to edge. Upon removing from the bag I sear the top and bottom on a hot grill just briefly for the expected color and caramelizing. So instead of having the meat over done on the outer 1/3 of the thickness it is perfect and tender through out. This process works on almost every food product but is most beneficial when you want to cook and hold a quantity of meat to a certain point and be able to serve a large crowd quickly. Searing 20 NY Strip steaks for 1 minute each side is easy and if you have a "well done" request it's easy enough to leave one on longer. I do it because the meat is delicious and perfectly cooked. Think duck breast, chicken , all kinds of slow cook meats like ox tails and brisket, pork ribs and shoulder. I recently did a souis vid brisket that took 36 hours. The seasoning was in the bag and after a searing it was unbelievable. Duck breast done souis vid is the best there is. Leave the skin on and after the water oven treatment, sear it for 2-3 minutes in a very hot pan makes it crispy and delicious while the meat is perfectly done and tender all the way through.
It is extremely important to follow food safety requirements and get the temperature up to speed quickly and keep it there. The souis vid process was developed by French chefs and is in wide use by many high end professional chefs all over the world. Contrary to what you might think, the food safety issue of bacteria spoiling the meat isn't a problem. NASA uses the process on the orbiting space station and the military has used it for years, especially the Navy.
I also use my controller unit to proof a sheet of rising dough with a small heating pad in the cool months. I have a plastic sheet pan cover I put over a jelly roll pan, place that on the heat pad and cover it all with a towel to hold the heat in. Works great. My PID was only $50 and there were other parts like a solid state relay switches and such. It was a do it yourself project. There are plenty of books on the subject at Amazon if you are curious.
Eric
gvz, Chuck, Eric
I can get an RKC PID controller relatively inexpensively here in Taiwan. Here are the English specs of a clone made in China. I am completely ignorant when it comes to technical things, so need to ask adviceas I don't want to buy the wrong thing. You guys seem to know a lot more about it than me. Would this model suffice for doing mash experiments such as the ones gvz has outlined above?
Many thanks,
Syd
Hi Syd,
I am no expert of PID controllers either, so I can't really say if the RKC PID is any good or not. Does it need to be assembled?
I was quite lucky with the one I purchased. Meaning... at first, with my setup, it was grossly overshooting the target temp, no matter what I tried. Then I found in the manual, the key that made it work -- lowering the max electrical output. From the doc:
Output power reduction. It is expressed as a percentage value.This function will allow you to control the maximum output
power delivered by the heater. For example, if you set Out=50
and your heater is 1000 watts, the output will use 50% of
the 1000 watts as the full output. It thinks the 1000W heater
as a 500W heater. When the PID algorithm determines 50% output
value, the actual power output will be 250 watts. This
function can be used in two situations.
1) When you have a very powerful heater and using a very
small pot of water to cook at very low temperature, for
example, a 1400 watts heater with a one litter (1 qt) pot
of water at 130 °F. The heater is too powerful for the
small water volume. The moment it is on, it releases too
much energy to cause the temperature to overshoot. Although
it is still possible to stabilize the temperature with
proper PID parameters, it is much easier to control if you
limit the maximum output to 25%. Ideally, an optimized
temperature control system should consume about 25 % of the
heater power at set temperature (steady state), for example,
if you found out that only 50 watts of energy is needed to
maintain the temperature at 60 °C (141°F), ideally you should
use only 200 watts heater for the job. Too much power will
make the system over react too quickly. Too little power
will make the system too slow in response. By using the OUt
function, you can make the 1400 watts heater to act as a 200
watt heater for stable temperature control.
This was the exact scenario I was having -- the single-burner I have is 1000-watts, and it was far too powerful for a relatively small amount of mash in the saucepan. So, I reduced the max output to 20% of max, turning into a 200 watt heater, and that enabled it to work properly. So, something to consider... I am not sure if all PID controllers have that feature or not.
This is the model I purchased, it is from a company in the USA. There was no assembly required. It came with a sensor probe as well, but I purchased a spare sensor just in case...
http://www.auberins.com/index.php?main_page=product_info&cPath=13&products_id=42
Taiwan eh? A long time ago I spent a couple weeks on business in Taipei. Loved it. Great food, great people.
Thanks, gvz. That is very helpful. It seems that power output reduction is a very necessary feature. I can't make out whether this particular make and model has that function or not. The specs are so technical they are confusing me. I will have to do some more research. Just as well I didn't buy on impulse. Thanks for your advice. :)
Syd
Hi Syd,
Another way to control max output might be also done by using the analog dial on most of the single-coil heaters -- "Low / Med / High". For me, I would rather set the heater to "High" (highest) and then use the PID controller to specifically set a max-wattage value, so that the P.I.D. calibration values aren't tied to some lower position the knob was turned at... it could be problematic if the knob were ever changed (say you cooked something else on it), then went back to mash making, but couldn't get the former knob position replicated perfectly... if that makes sense.. ?
I should clarify ... the knob on the heater I have, is a continuous turning control. It has labels for "low / med / high" but you can turn the knob continuously between them, half-way in between Low and Med, etc. Other heaters may have knobs that click into place for 3 specific settings of "low / med / high". If you have the latter, it might be sufficient to lock it at the Low position and use a PID controller that does not have a specific feature for controlling output power. The "Low" might be low enough, depending on how powerful the heater is (1000-watt, 1800-watt, etc..).
That could just be the answer to my problem. On further investigation, it appears as if the PID controller I am considering doesn't have that ability to control power output. Your suggestion makes sense and is certainly worth a try. I think it could work very well with my cooker. I have a really good cooker. It is an 1800 watt, digitally controlled Gourmet Haloplate. I am assuming it is heated by halogen lamps and it can go from a boil to a simmer to a standstill in, literally, seconds. It is as controllable as gas and it has six digital settings. So if I put it on the second setting would that mean it would step the power down by one third (i.e. to 600w). Because of the heating lamps quick response to the controls I think it could be ideal for this setup.
Syd
Hi Syd,
Can your digitally controlled heater turn on / get warm upon powering it on? Or do you have to push a button first after plugging it in before it will start to heat?
Again, I am no expert of PID controllers, but if I am reading this manual correctly, as the PID controller is increasing temp toward a target, it will provide less and less power to the heating device, until it is at equilibrium with the target temp, in which case no power is provided to the heater. This means your heater will be "off". When the temp begins to drop, the PID controller will begin providing power again. But, if your heater needs to have a button pushed to get it going.. it might not work unless you were standing there all the time to push it's "On" button..
From the doc of my PID controller, it says this about "P." value:
Proportional band. It is in 0.1 degree units. This parameter controlthe output of the controller based on the difference between the
measured and set temperature. Larger the P number means the weaker
the action (lower gain). e.g. If P=100, the proportional band is
10 degree (100 x 0.1=10). When the sensor temperature is 10 degrees
below the proportional band (10 degrees below the setting), the
controller will have 100% output. When the temperature is 5 degrees
below the set point, the output is 50%. When the temperature is
equal to the setting, the controller will have 0% output (assuming
integral and derivative functions are turned off). This constant
also affects both integral and derivative action. Smaller P values
will make the both integral and derivative action stronger. Please
note the value of the P is temperature unit sensitive. If an optimized
P value was found when operating the controller in Celsius, it needs
to be multiplied by 1.8 when changed to Fahrenheit.
Also:
The maximum electric current this controller can handle is
15 Ampere. For 120 volt AC in US and Canada, this limits the
heater power to 1800 watts
So, something to consider. I think these PID controllers might work best with "analog" / mechanical style heaters..
Then again, I could be totally wrong... hehe... :)
On my mechanical/analog heater, there is a red LED light on the front of it that lights up when it is heating. When I am using the PID controller, the light flickers. It's never solidly lit for an extended period of time, just kind of flickers every so often.
A PID controller controls the heat output (power) of the burner by turning it on an off according to the feedback input it receives from a temperature sensor (thermocouple). The burner sees it as an on/off switch--one that cycles on and off very quickly.
A 1200 watt burner will put out 1200 w-hr of power if left on for one hour. If hooked up to a PID that decides it only needs a 1/4 of that to maintain a temperature, the PID will cycle the burner on and off so that the burner is only on for about 15 mins over an hour period, e.g. 1 out of every 4 seconds or whatever as determined by the configuration settings.
The output voltage of the Auber temperature controller, which has a PID controller at its heart, is constant. The current is also constant--determined by the wattage and setting of the burner.
Course, it does say in parenthesis... ("assuming integral and derivative functions are turned off") when referring to 0% power output at equilibrium.
I did not have to play too much with the integral/derivative settings -- this unit has an "Auto Calibrate" feature, and it figured out the P. I. D. values on its own... (once I lowered max output). The integral and derivative functions do not appear to be turned off for me. So, I am not sure if it is truly cutting power output to zero when the temp has achieved its set temp, or if the temp has wandered slightly above the set-temp and it is cooling down. I suppose if I had a voltage monitor or something I could try to determine that. All I know is, the LED power light on the heater isn't lit most of the time during PID operation, as I mentioned it just flickers on/off every few seconds.
If the LED power light is flickering on and off, that would indicate that the power was, indeed, being cut off entirely. That isn't going to be good for my digitally controlled heater. I ran a quick test on it and if you pull the plug out of the wall socket while it is on, you have to press the power button and reset the temperature setting again after you re-insert the plug. Well I guess I could buy a relatively inexpensive analogue coil heater and that should solve the problem. It would be a pity, though as my haloplate heats up and cools down much more quickly than a coil heater does. You would think that it would make sense for the power to be controlled by a rheostat whereby the resistance is increased and the power output decreased. It can't be all that good for the heating unit to be constantly turning on and off like that. Perhaps having that kind of function would increase the cost of the unit. I have no idea, I am guessing wildly here. Thanks for pointing that out. Clearly, I need to do more research.
Syd
Hi Syd,
Come to think of it, I do have a "wand" type of voltage monitor, I forgot I had... it doesn't report a number, just beeps at you if it detects electricity nearby. (It's for a basic checking of electrical sockets in your house, or the circuit breaker switchbox, etc)... But I may try it the next time I mash, just for curiosity sake.
Turning on/off a digitally controlled heater may very well be hard on it over time. Maybe also for the cheaper analog coil ones. Fortunately, the one I am using cost $15 at the corner drugstore.. so far it seems OK. If its lifetime gets cut in half, oh well. :)
I'm also not an expert at all with these devices. It sounds like the unit gvz has is a smarter unit than my assembly. Where his actually raises and lowers the base voltage and then controls the power cycle off and on, my controller turns the power off and on via a solid state relay. The unit would handle low current devices but on advice of others who suggested using a relay I chose to let the unit switch the relay off and on as the PID sensed under or over temperature. I couldn't comment on the one you are talking about. My assembly is a total home built, right down to the project box and wiring. I've always been handy with electronics and have the experience and shop tools to do these projects. If you are inclined to try a DIY home project, search for "DIY souis vid machine". If I were still working, I'de buy the unit gvz is showing. Much less trouble.
I'm also just learning to brew my own home brew. The mash sounds interesting.
Eric
Thanks for that advice, Eric. Yes, it seems the unit gvz has bought is worth the extra money. Being able to control the power output will help the controller maintain the temp in a very narrow range without overshooting too much in either direction. I don't want to spend too much money on one but I don't want to buy one that I can't use, either.
Best,
Syd
In my application, I am heating water with bags of food (usually meat). I use a tank heater to warm the water that has the capacity to boil 5 gallons or at least get it quite hot. It's the kind of thing farmers will put in a stock trough in the winter to keep it from freezing. Anyway, for my use it is important to provide circulation of the water so I avoid hot spots and cold spots. All of the water must be the same temperature. It doesn't take much of a pump, I use an aquarium type submersible pump that holds up well as long as I don't try to cook at over 180F. That melts the plastic and ruins the pump.
So as for cooking mash for brewing or baking I think you have to consider stirring or somehow blending the hot bottoms up to the mix on the top. Maybe you will have to be near and stir it every 30 minutes.
Eric
Important considerations Eric. I am guessing a good quality pot with excellent insulation will come in handy here. Perhaps a rice cooker or an analogue slow cooker would be even better than a heater and a pot. Slow cookers are generally made from very thick ceramic and have good insulating properties and don't create hot spots. A lid and stir every now and then should even the temperature out as well. Mashing in a water bath might be the ideal, though. You could first create a roux of water and flour, then seal it in a plastic bag and immerse it in the water bath just like you do in the sous vide cooking. I would worry about heat and plastic together, though and the plastic imparting some kind of smell to the roux.
Syd
Syd,
I have used a slow cooker, rice cooker, stock pot and insulated coolers of varying sizes. The largest and most effective was a large 12 gallon fish cooler. It worked well keeping the water stable while cooking 3 full briskets.
Eric