I have a DIY cooking project with which I'd appreciate some help.
I'm trying to hold several gallons of water at 130F with minimal variation. A digital thermostat regulates the temperature (to within .3C), an aquarium air pump provides circulation, and a small immersion heater (Norpro 559) heats the water. The latter draws 300W on a standard 110VAC line.
So far so good.
The problem is that even though the heater powers on for less than a minute (out of even ten), when it turns off, the residual heat overshoots the mark, causing twice the range in temperature (.6C) necessary with the thermostat.
So I'd like to prevent the heater from getting so hot, reducing the wattage by approximately half.
I'm pretty sure that wiring in a resistor into the hot lead to the element would do it, but I don't know the specifics. Resistors are available in a huge array of OHM values and watts and what is utterly bewildering to someone like me is probably mindlessly simple to many of the experts who frequent this group.
Where is the sense point for the digital controller? Perhaps moving it to somewhere closer to the header might reduce the overshoot. AIUI, some home thermostats have a small local heat source (e.g., a hot resistor) near to their temperature sensor so that the sensor gets hot faster than the room air, so it turns off the heat just a little sooner and lets the residual heat "coast" to the final room temperature value.
Ultimately, you'll probably need a more sophisticated control setup, with, say, a triac controlling the heater so that you're not running a bang-bang control (full on - full off) but one that can taper off as it gets closer to the set-point, or turn on just a little if just a little heat is needed.
Zeroth question - other than you fussing about it, is this really all that big of a deal? Few cooking or culturing processes care all that much about a lousy degree, F or C.
Get another heater just like the first one. Run the two in series. That's exactly the size and power handling you'll need for a resistor, and it's already packaged/built. Each heater will only deliver 75 watts at half voltage, but the two together will deliver 150. If you find that
150 is still too much (and it probably is if you're at 10% duty cycle with 300W), move one heater into a separate tank of water. Or just get a smaller heater to begin with - seems like 50W would be about right (you have 10% - one minute out of 10 - with 300W - so 30W would do, and 50 W gives you a bit of reserve for variations, along with a heater that will be on more than half the time, rather than 10% of the time.)
If you really think you need super precision, a microcontroller and PWM control are the way to go. But it could be that you really don't need so much precision. I don't know what you're trying to culture at that temperature, but things like yogurt cultures really couldn't care less about a few degrees variation (and happily work down at 105-110F.)
Well, this is why PID controllers exist. The thermostat sounds like a simple heat-proportional "P" feedback loop closed on the temperature, and you have a need for the differential signal to be added in (the "I" is for integral, probably not relevant to this problem).
Well, a series resistor wouldn't necessarily dissipate as much as the heater, and you can get (or build) series resistors at just about any dissipation value you want. A series resistor that's designed to dump it's heat into some medium that you want to warm up is called a "heating element".
Using a diode to reduce the power in half would be a Bad Idea. It rectifies the AC on the power line, and forces DC current to flow through the transformer that's supplying the house. That can make the transformer core saturate, which makes the power company cranky.
If, for some reason, the OP can't take the suggestion to run two heaters in series, to get a smaller heater, or to just call the temperature overshoot acceptable, then driving the heater with a lamp dimmer may work out well, by allowing him to turn down the heat delivery at will.
--
Tim Wescott
Wescott Design Services
http://www.wescottdesign.com
Do you need to implement control loops in software?
"Applied Control Theory for Embedded Systems" was written for you.
See details at http://www.wescottdesign.com/actfes/actfes.html
You move the sensor closer to the heater elements or get a digital heat controller that has a PID controller in it.
With a PID ("Proportion-Integral-Derivative"), you can use the "D" parameter to slow down the heating cycle when it starts to approach the SP (Set point).
Dropping the heating wattage may not be such a good idea, it's better to have some reserve.
Of course, if you were using a RTD or simple PTC/NPC, one could whip up a basic voltage comparatar with some Lead (Derivative) in it to throadle it back when it gets close to the set point.
Making the physical bulk of the heater as small as possible, for a given power , will reduce overshoot. A long,thin, well insulated heater wire looses the remaining heat very quickly, and causes less overshoot. Increased turbulence will distribute heat faster to all parts, and will let your sensing circuit react faster. Last, by using a dimmer to reduce power in reverse proportion to the present temperature error, a balance may be achieved, and the power never goes completely off, eliminating overshoot.
Not sure I like the idea of pumping 72* F air bubbles through a system I'm trying to keep at constant 130*F. There is a better way to mix your water. Mikek
You can do this a few ways--suspend the heater inside the tank to maximize convection, use a mechanical agitator, or stuff the tank fairly tight with bronze wool.
Cheers
Phil Hobbs
--
Dr Philip C D Hobbs
Principal
ElectroOptical Innovations
55 Orchard Rd
Briarcliff Manor NY 10510
845-480-2058
email: hobbs (atsign) electrooptical (period) net
http://electrooptical.net
I don't care to have any drug research done on my computer. What's he cooking Rich? Mikek (-: PS. Guess I could look up acetone and alcohol boiling points. Hmm.... Acetone 133*F Ok, what dissolves in acetone?
Jesse wrote in news:Xns9EDE83B3A34Bjesseatnospamdotcom@127.0.0.1:
Many thanks to all who troubled to share their thoughts. I thought I would be lucky to get even a single reply, but in only a few days there have been fifteen from almost as many individuals.
Rather than reply to each individual message, I'll try to clarify matters with a single response. I hope everyone who replied will see this contribution to the thread.
First, to respond to Ecnerwal
And Rich Grise
Over the past several years, within the general realm of "molecular gastronomy" exists that of "sous vide" (French for "under pressure") cookery.
Misleading nomenclature to the contrary, what this is really all about is instead of cooking, for example, a beef roast at
350 degrees and removing it when it reaches the desired internal temperature, it's wrapped and placed in a water bath of the desired temperature (130 for example) to begin with. It's then cooked not just for hours, but sometimes for days.
Okay, what's the point?
There are several advantages (google to acquaint yourself with them) but among the most important is that, as in pit BBQ, this "low and slow" cooking has a tenderizing effect, offering the chance to turn the lead of cheap chuck into the gold of rib eye and having it STILL come out medium rare.
Being a somewhat new approach to cooking, however, it simply isn't known what effect temperature fluctuation has on the process. The general consensus is that less is better. I knew when I bought the thermostat that it only allowed a .3C variance and was content with that.
Many have used a PID controller, as whit3rd noted, but I thought that even though it allowed much greater temperature stability, its complexity was beyond my capacity to properly configure.
When I discovered that my STC-1000 thermostat exhibited a small but significant overshoot, I thought it was probably acceptable but I wondered if there was an easy way to reduce it.
One obvious approach was to keep the heating element from getting so hot. I now realize that a simple resistor simply won't work, it would have to be massive to reduce the current by the required amount.
I'd like to use a less massive heating element, as Sjouke Burry pointed out, but finding something ready made that can be immersed in water, isn't so easy. A home aquarium heater can be hacked to remove its thermostat, but this isn't a project I'm comfortable doing.
Rich Grise saw a good solution
It looks like $5 spent at Home Depot should do the trick.
Mikek had another insight...
And John Fields...
To mix the water.
Originally, I tried placing the stockpot on a hot plate (connected to the thermostat) without the bubbler, hoping that natural convection would do the trick, but the overshoot was pretty bad. Adding the bubbler helped, but using the immersion heater and bubbler was the best solution so far.
Cooling the water by bubbling room temperature air through it, I knew, wasn't such a good idea, but the only economical alternative I could think of was something like a submersible tabletop fountain pump, like a Sunterra. The question was whether it would tolerate such hot water. I emailed the company and inquired, but received no response.
Rather than delay any longer and worried about mixing electrical current and water, I thought the air pump, despite its disadvantages, was probably the better option.
Perhaps some who have followed this thread may have a thought on this topic. Would one of these tiny submersible pumps function at 130F temperature? For very long? Would it simply fail and need to be replaced or result in some disaster?
Anyway, for less than $50 - compared to $1400 for a Fisher laboratory circulator or even $400 for a retail Sous Vide Supreme - I'm cooking sous vide.
And very happily too. Like pit BBQ, of which I am also a fan, my few attempts so far have revealed that it can create genuine culinary magic.
Thanks again to all for so generously sharing their knowledge.
Very interesting Jesse, Do you worry about some kind of bacteria growing at 130F? And how long do you have to let the meat cook? I was the lunch cook at a fairly nice restuarant for about a year. The first thing I did when I came in the morning (~5:30 AM) was to start the prime rib for the evening. Started at ~220F for a few hours and then dialed down to
170F IIRC.
I think you are obsessing excessively about a few degrees, of temperature fluctuation.
George Herold wrote in news: snipped-for-privacy@a26g2000vbo.googlegr oups.com:
Thanks for the reply.
Bacterial growth is of great concern, but has been studied extensively - google Doug Baldwin, for example, for more info. From all available info, both theoretical and practical, the
130 mark is accepted as safe.
As for the concern over temperature fluctuation, you may well be right. I wouldn't call my concern "obsessing" and I'm content with what I already have, but if was easy to keep the range a little smaller, and it seems that it is with a dimmer, I'm willing to polish the setup to make it a little better.
Lastly, please realize that heat transfer in liquids is MUCH greater than that in air, some 23 times as much, as I understand.
Putting your hand in a 350 oven for ten seconds is no big deal. Doing so in the same temperature oil is rather different.
ElectronDepot website is not affiliated with any of the manufacturers or service providers discussed here.
All logos and trade names are the property of their respective owners.