Fermentor temperature controller

[RadicalEd]

Although I've yet to brew a batch, my engineering instincts have kicked in (and my sub-zero computer cooling experience ) and I have plans in my head for a pretty solid temperature stabilizer designed to keep a fermenter (be it conical or carboy) at a very specific temperature. The interesting thing is that I think I could design it to both heat and cool, so that the owner could simply set it, turn it on, and forget about it.

Do you guys think that there may be a niche market for this, given that I think I could put them out for well under $200?

Being an inexperienced brewer, though, I lack once crucial piece of info: how much heat does fermenting beer create per gallon? With that I can zero in much closer on production costs.

Thanks in advance!

 

[kvh]

Here's a bit about it:

https://www.homebrewtalk.com/showthread.php?t=31506&highlight="heat+fermenting"

I actually think I've heard people talk about as much as 6 degrees hotter. This is probably over a 5-gal batch, generally.

kvh

 

[RadicalEd]

Thanks for the link, but sadly they only talk temperature, not heat . They are related to each other, but just knowing one does not give you the other. I.e a well insulated heat source providing 10 watts of heat will likely be much warmer than a 100 watt heat source with a huge heatsink attached.

To properly design this system, it must be capable of removing more watts of heat (or BTU/hr if you're more comfortable with that) than the yeast is pumping out. If you can establish that, it becomes much easier to maintain any given target temp.

Keep 'em coming!

 

[Thalon]

I doubt cheapskates like myself would go for it, as there are much cheaper altermatives already out there. Here's plans for a simple and cheap (~$50) fermentation chamber that is thermostat controlled and uses gallon jugs of ice to keep it colder than the ambient temp:

http://home.elp.rr.com/brewbeer/chiller/chiller.PDF

Someone else took the idea and modified it to use your house's air conditioning vent as the chill source, thus driving cost down even more:

https://www.homebrewtalk.com/showthread.php?t=32651

As long as the ambient temp is higher than your target brewing temp, either of these should work fine.

 

[FlyingHorse]

Do you guys think that there may be a niche market for this, given that I think I could put them out for well under $200?

I'd be interested to see how you're going to do that, when the cheapest available mass-produced corny-sized refrigerators (ie the Sanyo 4912) go for $160-180 on sale. And they only cool.

There'd probably be a pretty small niche...I think most HBers don't need to manage the temp in both directions (everyone's looking for cooling; warming generally isn't a problem. Maybe there are a few hardcore folks out there who would like the capability, but that's a small slice of the folks who are members here, which is a smaller slice of the homebrew universe in general. Worse yet, the fanatics who want that much control tend to be DIYers who can build (or have built) their own solution.

Not trying to p*ss all over your idea, but you asked

 

[RadicalEd]

Thanks for the additional input. One of my goals here is to stop myself from getting too far involved if it really isn't a viable option. If nothing else it may just turn into a personal experiment when I have more equipment, or a class project (electrical engineering), if I can pull that off .

I'm in the process right now of putting together "best case" and "worst case" scenarios for pricing, as well as designing a basic control circuit.

EDIT:

Ok after a little bit of research and SWAGing, I've come up with some preliminary figures:

For best case scenario, wherein I'm able to get parts at optimal prices and a smaller power supply will do, parts could be as little as about $70.

For a worst case scenario, wherein I'm not getting optimal prices and need more power supply oomph, I'm looking at $135 for parts.

I can take a few "shortcuts" from here, ie using thermistors over thermocouples or just cooling instead of heating too, which could knock off $2-7 and ~$12 respectively.

But still, one of the largest issues confronting me is how much heat the yeasties make; if anyone can get me that info I could go ahead and design a unit with available part specs.

Thanks again to everyone responding!

 

[malkore]

its probably something the upper echelon of home brewers would love.

the middle tier usually has a freezer with a temp controller on it for kegging that they could use for lagering/temp control.

the lower tier isn't ready to invest another $300

I think its a cool idea, and if you make one for yourself, share it with us (after you file a patent

 

[RadicalEd]

That's kinda what I'm thinking now. Thanks for the support, Malkore!

 

[WOP31]

One problem with trying ot design it around how much heat a certian strain produces is that they are all different. You would need that info for each yeast strain. plus each brand of yeast. Your best bet is to contact wyeast and white labs and nottingham, and so forth.
For example I made an IPA and a scotch ale and had them both in the same closet and the IPA kept itself around 68-70deg, whereas the Scotch ale, due to the yeast being more aggresive and active, had to have a towel draped over it in a bucket of water, just to maintain the mid 70's on it, otherwise it would have been in the 80's.
Sounds like a good idea, and if you could keep the price down I am sure that people would buy it. Keep us updated i know that everyone around here loves new equipment.

Cheers

 

[the_bird]

$200 would be a real sticking point for me; the $80 digital Johnson Controller will heat or cool, and if were to use the $24 Thermowell, I'd be able to dial in that temp based on the wort temp, not the ambient (anyone use the Thermowell, BTW?). How would your system be an improvement over this?

 

[olllllo]

$200 would be a real sticking point for me; the $80 digital Johnson Controller will heat or cool, and if were to use the $24 Thermowell, I'd be able to dial in that temp based on the wort temp, not the ambient (anyone use the Thermowell, BTW?). How would your system be an improvement over this?

Plus you can use it on anything from a dorm fridge to JohnBeer's walk in (window AC).

 

[the_bird]

I wasn't even thinking that this would replace having to acquire a fridge - but fridges can be had so cheaply that their acquisition is really a non-issue. Dorm fridges are what, $50 on Craigslist? I got my lagering fridge for free, just bought a $50 analog temp controller, that's working great.

 

[RadicalEd]

Wop: I don't really need to know the heat of each individual strain; as long as I know the maximum heat output of any strain, and can remove more than that amount, then I can use the magic of cycling to maintain a target temp.

Bird: The particular advantages of my unit, as I see it at least, are that you can individually control the fermentation temps of each of your fermentors, and that you don't need a large and bulky fridge or chillcloset in order to accomodate those massive conicals (which are likely to be the primary market, simply due to their size and the reletive cost of this unit to their own). If you can fit one conical in a dorm fridge, that's $50 (fridge) and $50 (controller) for a total of $100 for the average joe, so if I can make this for under that, then I think it would actually be the economical option for some. Admittedly, though, I don't think that this unit would be able to bring and keep a fermentor down to serving temp (ie 40*f); it just won't have the power.

Ok, for now let's assume that this unit will only be cooling, with heating left as a future option to pursue. With that, part costs are looking to be around $60 in an ideal situation.

But before throwing around too many figures, I'll contact a yeast maker (brilliant idea, BTW) and see if they can give me any figures on heat output. If it turns out to not be much, then the optimal price is quite within reach; but I think that even if it isn't under $100 is still quite doable.

But please keep the dicsussion coming! Any and all constuctive criticism will either help me get out now or establish a stronger product .

 

[the_bird]

I strongly suspect that the "how much heat does fermenting beer create" is going to vary quite a bit. One would certainly expect a very aggressive fermentation to produce more than one that's slower and more subdued. In fact, I would almost expect that the ambient temp - which impacts how aggressive the fermentation occurs - is going to also impact the amount of additional heat that the fermentation process adds.

FWIW, I've heard a 7° figure in the context of fermenting a big Belgian, I suspect the answer is usually going to be around 5° - 7°.

So what's the "magic" behind your system? If it's peltier coolers (which I know next to nothing about), lots have tried that with little success; might be worth a search.

 

[RadicalEd]

Again, as long as I know the maximum heat being produced, it is a simple matter to desing the system to hold a higher load, and then cycle the system to reach your target temps.

Also again, any temperatures by themselves mean very little; unless you know the thermal conductivity of the fermentation vessel (not a fun figure to find, mind you) temperatures by themselves mean next to nothing. What I need is the rate of production of heat, ie watts, BTU/hr, calories/hr, etc. Put a 10 watt light bulb in a styrofoam container and I can almost guarantee you that it'll be warmer than a 100 watt bulb in the air. A 7 degree delta in a metal fermentor is a lot different than a 7 degree delta in a glass fermentor; metal conducts much more heat.

Well, I admit peltiers play a big part of my design (not much magic to 'em, though ), but where I hope to innovate is how you attach it to the fermentor. The big problem here is contact; you need a very good contact between the two surfaces in order for the heat transfer to occur efficiently. With different sized and shaped fermentors, that is almost an impossible task (needing custom design and lapping), but I might have hit upon a way to create a good contact with most any fermentor. I'm partially bouying my hopes on the fact that a few commerical producers are using peltier devices, or TEC's, such as Morebeer.

An interesting side note about TEC's is that if you reverse the current through them, they switch the direction that they are moving heat; ie they switch from cooling to heating.

But yes, I'll take a look at some of the TEC exploits around here. Thanks for that.

EDIT: ok I've taken some time to look through other pelt threads, but in each one there is at least one simply explained cause of failure, ie PSU not big enough, bad surface contact, not enough surface area, etc. Yuri was on the right track with his, but he hasn't been working on it recently. I'm hoping that my experience cooling computers (I'm actually building a unit to run my computer at -30*C via chilled liquid) will allow me to avoid many of the same pitfalls.

 

[MadWeezel]

Heat of fermenation:

http://oz.craftbrewer.org/Digest/messages/2004/January/msg00311.shtml

Does this mean i get a free unit for finding the answer?

 

[RadicalEd]

Not a free unit, but I'll give you a HB (or five) once I start and you're in the area .

So let's say that you start with a nice heavy wort at 1.1 OG, which is fully fermentable and will fully attenuate. So:

5 gallon batch * 0.10 g/ml (glucose contribution) * 3785 mLs/gallon * 1/180 g/mol = 10.5 mol glucose

Given that 67KJ are released in this process/mol, that's 10.5*67 = 703.5 KJ released.

Spread that heat over the course of 7 days (please correct me if that's a very incorrect assumption to make) of fermenting time, which is 86,400 seconds/day* 7 days = 604880 seconds, so rate of heat production is: 703500 J/604880 seconds=1.16 watts.

Now let's throw in a fudge factor, to say that most of the fermentation happens in 3 days: that would be 1.16*7/3=2.71 watts.

Let's further extend that to a 20 gallon batch, so 2.71*20/5=10.8 watts.

So giving some very serious leeway in terms of fermentable sugars and fermentation times, the most heat a 20 gallon batch can be expected to produce per second is 10.8 watts.

What tells me is that most TEC attempts to this day have been flawed in such a way that they are not efficiently removing the heat from the fermenter, as the TEC's themselves are fully capable of removing that heat. That gives me hope that a TEC solution is a viable one.

 

[kvh]

...

Let's further extend that to a 20 gallon batch, so 2.71*20/5=10.8 watts.

So giving some very serious leeway in terms of fermentable sugars and fermentation times, the most heat a 20 gallon batch can be expected to produce per second is 10.8 watts.

What tells me is that most TEC attempts to this day have been flawed in such a way that they are not efficiently removing the heat from the fermenter, as the TEC's themselves are fully capable of removing that heat. That gives me hope that a TEC solution is a viable one.

This is why you went to school for electrical engineering, and I went to art school.

 

[Lil' Sparky]

RadicalEd, I would say that very rarely does an ale take 7 days to vigorously ferment. The main part of the fermentation usually takes place between 2-4 days, sometimes even 1 day, and that's when most of the heat is generated. After that, fermentation is still taking place, but it's much slower and I doubt much heat at all is being produced.

 

[Thalon]

But he's right in that you have to design it for the biggest load it will experience. They have to design air conditioning systems for stadiums based on it being completely full of people, even though most of the time it's empty and during the few hours a game is on it might be half full. You don't want to under-design the fermenter to fall behind in cooling power during the vigorous initial fermentation just because its sweet spot is for the rest of the 3 weeks in the fermenter.

 

[Lil' Sparky]

Right. I was trying to say he was probably underestimating the biggest load, since he was overestimating the time of the vigorous fermentation.

 

[enfox]

I have an iceprobe (fish tank cooler) installed on the side of a corney and I can get temperatures down to 60 degrees with not much trouble. If you can make a unit that is as clean as that and does not require drilling the fermenter and cheaper I think you may have something.

 

[RadicalEd]

Well, that 20 gallon batch at 10.8 watts was adjusted to brew vigorously for 3 days. If we shorten that to 1 day, we're looking at an absolutely blazingly fast fermentation for such a high starting gravity (but it is better to over estimate than under) and we get 32.4 watts.

Compared to 10.8 watts, that is a sizable amount, and would require a decent sized TEC to get that cooled, at least a 40-50 watt model. I'm currently looking for one so that I can "play" with it. TEC's are funny things, and different models react differently to different voltages. Oddly enough, it is more efficient to buy low wattage units with max voltages near what you can provide (ie 12V) than to get a high wattage unit made to run at 24 volts. I think that that may have lead to some of the difficulties that many have been experiencing; they bought big and powerful TEC's, but were unable to feed those hungry beasts enough volts.

 

[Thalon]

Right. I was trying to say he was probably underestimating the biggest load, since he was overestimating the time of the vigorous fermentation.

Ahhh okay gotcha. Maybe it's a good thing I didn't go into electrical engineering too.

 

[WortMonger]

This is a interesting thread. I am still waiting on Yuri's little project with anticipation, lol. I personally think (not having any experience in the matter) properly matched TEC's hooked to a matched probe-type heat exchanger only placed going down through the lid would be great. I mean the heat rises anyway, right? I just figured a big conical could be wrapped with insulation and made to look professional of course. I like that everything producing/removing would be on top and easily installed through a flat faced lid/ or dome and would be easily insulated around and made to look nice too. I am toying with the idea because of wanted to ferment large batches at different temps and being able to use my chest freezer for dedicated lagering. This of course is when I get my conical. Like I said though, I have no knowledge and will definitely keep up with the intel on these threads.

 

[RadicalEd]

You know what, wortmonger, you just may be right. The key to this system may just indeed be a top-down unit; there's a whole heckuva lot less to worry about than a side-unit. My only problem with a top-down is that is will make cutting up your top completely necessary; for some people that may not be a problem (ie they are making the top and thus one more cut isn't a big deal) , but for others (including cornie fermenters, ie) chopping up their tops isn't an option.

I'm on the verge of buying a lot of 10 TEC's to play around with, and I'll start with the side-mount. But I'll give the top mount a try if I can't get good results with the side mount. I'll have to improvise a test though, as I don't happen to have a conical lying around . I'll probably slap it on a gallon jug and calculate the effective heat removing capabilites, ie how long to reach a certain temperature.

 

[WortMonger]

I think drilling a larger diameter hole in a corny wouldn't be too out of reach for the average person. Someone always knows someone with a drill press. I wondered, for a more advanced thought, if a hydrogen filled aluminum cylinder hooked to the cold side would be even better heat transfer. I mean if you were machining your own cold probe anyways whats a little threading work and adding a valve to add your gas to the sealed cylinder? A little ambitious, but I am just thinking that every little bit of efficiency helps and machining projects are fun.

 

[jdoiv]

This is an interesting discussion. Question for you though. Is your idea to have a probe go into the wort to pull the heat out or just using a way to make a better connection with the fermenter? If it is using a probe, then going through the top would be a good way of doing it as long as the probe reached to the bottom of the fermenter. If it is by contact on the side of the fermenter, then putting it at the top would not be efficient has you have a lot of air and headspace (I'm guessing the probe is what you are thinking of).

As a side note, when I managed kitchens in a former life, we would put our soups and sauces into an ice bath to chill them down. To speed the cooling and lessen the chance of spoilage, we would leave a large metal whisk or spoon in the center of the soup container. This acted as a heat exchanger to pull the heat from the soup and speed up the cool down process.

A device that could fit through the opening of a carboy and cool it to a specific temp and allow for an airlock, would be a nifty little invention.

 

[WortMonger]

Yep, in my thinking I would want it to nearly touch the bottom of the conical part or bottom of the fermentor. Being straight vertical it wouldn't collect any sediment and should act as a heat elevator lifting it out the top of the vessel to the moon, lol. You could even make it kind of star shaped lengthwise for more surface area on the probe (like fins on a amp only going around). I also think the weight of the TEC and lid would keep a nice seal if the manufacture was that simple (I will fit-out for pressure if I build though). Cleanup would probably be a soft washcloth if I had the probe polished. I found that very good information about polishing the surface of where the probe meets the thermal lube and then the cold side of the peltier junction. I just figure it will look better if the beer side of the probe was nice and shiny for looks and ease of cleaning and inspecting. I think it would be cool if someone could make a much smaller diameter one for carboys, maybe one with blowoff escape and temperature probe built in. All you'd need is a couple of blankets wrapped around that bad boy and viola!

 

[RadicalEd]

The biggest problem surrounding a rod-type design is the length and material of the rod. The best option for conducting heat would be copper, but copper has exploded in price over the last couple years. Seeing as how it's also used for drinking water, I doubt that it would add any off tastes, but galvanic corrosion in a steel conical is a very real threat.

Aluminum would be the next best from a thermal standpoint, and is much cheaper, but the galvanic corrosion is even worse.

If you start going much below Alu in thermal conductivity, I'd be worried about being able to transfer enough heat along that rod up to the peltier device, especially in a regular carboy. I'll keep pondering, but I can't really think of any "perfect" solution.

 

[Sir Humpsalot]

Well, since you're running a power source... couldn't you measure the typical voltage through the wort as a result of the dissimilar metals... and then induce an opposite charge of equal magnitude to balance it out?

Oh wait.. I'm complicating things. I'm drunk and don't know much about science. I know they do some parking garage structures that way though.....

 

[Biermann]

Hmm. . . .RadicalEd-- have you looked at the heated/Cooled conicals utilizing pelts from B3? They use a Ranco controller, and a pelt with a huge heat sink lapped onto the side for temp control.

Might be an idea to improve upon.

 

[RadicalEd]

You know, I did see them when I was looking at the morebeer catalog (they gave me confidence that TEC's would work), but I found a few things dissatisfactory with it. Namely, it cost a whole bunch of money for the upgrade ($545 for just cooling, including insulation, on the 7.1 gal), and the conical needs to be on a lapping machine for a very long period of time (ie, only their conicals at time of purchase).

Using a Ranco controller would certainly make my life a whole bunch easier, though. I can make an analog one for a ton less, but the digital capabilities would be nice. It's just a matter of whether one can afford that controller.

Damn Squirrels, that is an interesting idea, and I had considered it, but it would mean current running through the wort, and I frankly am not sure if that would have the potential to harm it. Might be an interesting experiment if someone wants to make a FrankenBeer.

Right now I'm thinking that any rod used would have to be plated with nickel, gold, or silver. That's about the only way I can see to halt corrosion while still being able to use a metal that conducts heat well.

 

[WortMonger]

I read yesterday that copper has twice the thermal conductivity of aluminum. Wow, I am just getting into metal casting more and more and copper is charcoal melt-able. I know it is expensive but properly shaped wouldn't be a lot of material and would make it to the bottom of the fermenter no problem, even a 20+ galloner. I mean you would have to scale everything up/down of course. I also read yesterday that copper actually removes hydrogen sulfite from the beer. I mean breweries still use copper cool-ships as well as primary fermenters so a copper heat-sink rod should have no ill effects. As for galvanic corrosion because of dissimilar metals, it shouldn't be too hard to separate with silicone/etc gasket the two metals. I mean in my mind the thing is on the lid and everything kinda just all fits going through the hole you make. So, it should be easy to separate and seal your fermenter from outside air. I intend a design that uses every bit of the cool side peltier's surface area and the great thermal conductivity of copper. I know this is out of reach for some, but hey if you could mass produce it wouldn't be. I, of course, base this on not knowing how much copper is truly needed. I may be way off and need a big thick rod of copper or something.

How much mass is needed on the copper probe I propose RadicalEd? I am thinking maybe a 1/2" solid copper bar almost the entire depth of the fermenter, surrounded by 2-3" long heat sinks coming out of the bar running the whole length up to the copper peltier connection plate. Then the grease, then the peltier (or stacked pelts), etc. etc.... Investment molds could be made and easily repeatable precision parts could be made as soon as enough scrap copper was rounded up. Fun, Fun.... With smooth parts you wont have to worry about sediment sticking to your sinks and they should clean and polish easy too. Wow, if the science part comes true with you guys information and projects this might be my big melt project. Keep this thread active please, I am getting so many ideas from your info.:rockin:

 

[RadicalEd]

Well, I'm afraid that preventing galvanic corrosion is not quite so easy as putting a gasket it; as long as the two metals are in contact with each other via an ionic solution (and beer is acidic=ionic) then corrosion both can, and will occur. I come from a computer cooling background, and mixing metals without the proper preventative steps is a big no-no. The easiest solution is to add antifreeze, but something tells me we don't want that .

But for the moment let's assume that we can avoid corrosion. I don't think that a metal bar is really the way to go on this one. Your best bet is likely to use concentric layers of copper tubing with a notch cut across them, like this:

viewing it from the side. This should offer significantly more surface area (the really important factor) than a copper rod, and thus eliminate the need for heatsinks. www.mcmastercarr.com is a pretty goo place to get metals, and I hear www.onlinemetals.com is pretty good, too.

This assembly would then need to be brazed (high-temperature special soldered) to a "cold plate," a plate of metal that rests against the cold side of the pelt. I'd then recommend using Arctic Silver Thermal Epoxy to glue the pelt to the cold plate, since non-epoxy solutions will require a hold-down mechanism. Throw some epoxy and a heat sink down on the hot side of the pelt, and you're pretty much set.

And I'm glad I can help .

 

[WortMonger]

What about silicone coating the copper or would that not be smooth enough to let yeast slide?

Or, is it really going to corrode in 7 day ferments? I mean you will be removing the un-siliconed copper and cleaning it as well as cleaning the stainless steel and reclassifying it before the next use anyways? I just wonder. I need to do more research on how much copper is too much copper. I think though if short term exposure like I said would be ok. It is the stainless we are worried about right? Questions questions.

 

[RadicalEd]

I'd avoid coating the copper with anything other than a metal plating; just about anything else is going to really hurt the heat transfer characteristics, risk making the beer undrinkable (I wouldn't get silicone anywhere near my beer), and if not perfectly well covered, risk corrosion regardless.

Galvanic corrosion is always a risk. Even if it's just for 7 days at a time, you still could encounter corrosion. Just as important as destroying your equipment, it could easily destroy your beer, too. Remember how it works; the ions in the liquid pick up metal atoms and ferry them around the liquid. So it introduces unwanted metals into your beer, too.

I think most people would worry more about the stainless as it is the much more expensive component, but corrosion would affect both equally.

 

[jdoiv]

Would you have galvanic corrosion if you were using one of the plastic conicals? Only one metal would be in the wort. This could be a good solution for the plastic conicals.

 

[RadicalEd]

Aye, there can only be galvanic corrosion if there's two metals in ionic contact. So the boys with plastic can freely use copper .

 

[WortMonger]

Gotcha on the corrosion, man that is good to know. Electroplating the copper would be the way I went then. With what I don't know, but plating is deffinitley the way I would pursue based on what you have said now. So, what do you recommend for a stainless conical? Chrome plated? Nickel? I still think copper for the heat transfering would be best only now I won't worry about corrosion.