Marty Tippin (martyt@pobox.com)
Original: Feb 16, 1996
Last Update: August 22, 1998
August 21 1998: Since my
interests have shifted away from homebrewing, I've sold all my
equipment and am essentially out of the hobby for now - but these
pages will remain online for the foreseeable future. And I'll
still be around to answer questions, so don't hesitate to e-mail
me!
-Marty
This document is intended for those who are thinking about (or are in the process of) building a 3-vessel, converted 15.5 gallon "Sanke"-type keg brewing system, similar to the types sold commercially by Brewer's Warehouse, Pico, Sabco and others.
I've recently completed building my system and wanted to get what I did down on paper (or, in this case, on silicon) in hopes that it would make the process a little easier for others. (Of course, I also want to show off what I've made - I'm pretty proud of it…;-)
The first question you have to ask yourself before starting a project like this is, Why?? For me, the answer still hasn't become crystal clear - I guess it all started on an impulse when I had the opportunity to acquire some used kegs in good shape. After that, I was able to fabricate all kinds of reasons why I needed a converted keg system - some of the reasons were real (I was tired of doing my mashing and sparging in the kitchen and having to lug everything down to the garage to do the boiling and chilling. My spousal unit was also tired of the way I always managed to leave the kitchen in less-than-pristine condition after brewing a batch…) and some of the reasons were mostly imagined (I would make such fantastically better beer with the new system that I'd be sought out by all the microbreweries in town to be the brewmaster…)
Seriously, though, I was looking for a more convenient, compact and self-contained brewing setup that wouldn't require me to lift, carry, stoop or otherwise exert myself (can you say, "couch potato?") My usual routine of heating the mash on the stove, putting it in the oven for a while, heating it some more, waiting some more, heating sparge water in every empty container in the house, ladling the sparge water over the grainbed for an hour, hauling everything downstairs, boiling for 90 minutes, cooling with an immersion chiller and pouring off to the fermenter usually took about 7 hours and left me so tired afterwards that I could scarcely lift a glass of homebrew to my trembling lips. While I anticipate that the new system will make things easier, I don't expect that it'll shave off too much time - hopefully, though, I'll still be able to see straight when I'm done with a batch. And I won't have messed up the kitchen, thus staying in the good graces of my wife…
I've also been considering moving from 5 gallon to 10 gallon or larger batches - I drink a fair amount of brew and give even more away to friends, relatives and anybody else who'll take it. A 5 gallon batch really doesn't last that long. With the converted keg system, I can do a maximum batch of about 13 gallons or I can continue to do 5 gallon batches without changing anything other than the volume of water and amount of grain.
Another question you should ask yourself is whether or not your bank account is up for this exercise. I initially estimated that the whole system would cost around $400 to build. That was a double-edged sword: On the good side, it allowed me to talk my loving wife into letting me get started (if I had told her it was going to cost as much as it did, she never would have given the green light.) On the bad side, I've spent somewhere between $750 and $800 at completion, nearly twice what I estimated. But that's still a lot less than you could buy a comparable ready-made system for.
So now that you've made up your mind to go for it, let's get started!
I'd like to thank Kirk Fleming for his article, "A Simple Recirculating Wort Processor." That article was the single best reference I had while designing this project and I stole -er- used many of his ideas. Teddy Winstead's "Half-Barrel Keg Conversion FAQ" was also an invaluable source of information. I don't intend to rehash all the information contained in those two documents; instead, I suggest you read them first to familiarize yourself with the many of the more important issues involved in this task.
I'd like to thank Harlan Bauer, Jeffrey Green and everyone else on the rec.crafts.brewing newsgroup that were so helpful while I was solidifying my design. And lastly, an extra special thanks to my brewing partner, Robert Swenson, for his advice and assistance during the construction of this monster.
Most of the information I used in assembling this system was gleaned from The Brewery, which is to my knowledge the single largest repository of homebrew-related information on the 'net. If you've never been to The Brewery, check it out as there's tons of great information there.
I'll assume you've already got three 15.5 gallon "Sanke" type kegs - as I said above, I happened on to mine mostly through chance, and I leave obtaining them as an exercise for the reader. Refer to either of the two articles above for more pointers.
The keg conversion is based mostly on Kirk Fleming's design - rather than having a nipple (a piece of pipe with male threads) welded into the side of the keg, Kirk recommends using a coupling (a fitting with female pipe threads on both ends) instead. This affords a great deal of flexibility that a nipple never would - if you decide you don't like the design, take out whatever you've threaded into the coupling and replace it with something else.
I found a local welder who specialized in "sanitary" stainless steel welding and had him cut a 12" hole in the tops of the kegs and weld a total of 5, 1/2" NPT SS couplings: two each on the mash/lauter tun and the hot liquor tank (one for the dip tube assembly and one for the bimetal thermometer), and a single one on the boiling kettle (for the dip tube assembly). The couplings were obtained at a local company that specializes in stainless for about $20. The welding cost just under $120.
The couplings for the dip tubes were welded about 6 inches from the bottom of the keg. For the thermometers, the couplings were welded about 4 inches from the bottom of the keg and about 6 inches to the right of the other coupling. I had the couplings welded flush to the inside of the keg, which turned out to be a minor mistake: The welder splashed a little slag into the threads of the coupling and I spent a couple of hours with a carbide bit on my Dremel tool cleaning them up. I'd recommend that you have the couplings inset about 1/4" inside the keg to prevent this problem.
The kegs were still in pretty rough shape when I got them back from the welder, so I spent a few hours with a grinder cleaning up the edges where the holes were cut, using a wire brush (use a stainless steel one!) to shine up the welds and a Brillo pad, mixed with liberal amounts of Comet cleanser and elbow grease, to scrub the inside and outside.
One other addition I made to the kegs was to add a sightglass assembly so I could tell what volume of liquid was in the kegs. The assembly is made from a 1/8" NPT to 3/8" compression brass elbow, a length of 3/8" OD polyethylene tubing (the milky colored stuff used to run water to an icemaker), some 1/2" OD aluminum tubing (which happens to be 3/8" ID) and a couple of braces designed for 1/2" tubing. First thing was to drill and tap a 1/8" NPT hole in the side of the keg, near the bottom, and install the elbow - use care when tapping the hole as it's easy to strip it out and not get the threads properly seated - I practiced first on the cutouts that I had saved from the welding. Next, I attached the polyethylene tubing to the compression fitting and measured a length of aluminum tubing to slip over it. I clamped the tubing in a vise and cut away about half of it longitudinally - the leftover aluminum tubing slides over the polyethylene tubing to give it support. The clamp then goes over the tubing and a small bolt is used to attach the clamp to the upper skirt of the keg. The setup works like a charm - no leaks and I can easily read the exact volume of liquid. Total cost for all three kegs: about $15.
The next major obstacle was to figure out how I wanted to mount the kegs. This decision had to be made at the same time I decided whether I would use gravity only, or one or more pumps for moving the liquids. I could have done a 3-tier, gravity fed system, or a single-tier, multiple-pump system, or any of a hundred other designs. So I started researching and trying to figure out the pros and cons of each system I ran across.
It quickly became clear that I didn't want a 3-tier gravity-fed system: in order to position the boiling kettle so that you could use gravity to drain the wort into the fermenter, the other two kettles had to be so high that it would be difficult to deal with the mashing and sparging. Likewise, I didn't think I could afford the two magnetic drive pumps that would be required in a single-level system (one for pumping sparge water and another for pumping sweet wort into the boiler). But I could swing getting a single mag drive pump for recirculating the sweet wort during mashing, pumping the sweet wort to the boiling kettle during the sparge, and for pumping the bitter wort out of the boiling kettle after boiling. I saw no need to have a pump for the sparge tank - gravity could do it's thing there with no problem. So I settled on a two-tier system with the boiling kettle and mash/lauter tun on one level, and the hot liquor tank on the upper level.
Having leaped that hurdle, I was faced with what material to build the rack out of: Wood was not an option since I planned to use 3 separate propane burners for heat. Angle iron was a possibility since it would be relatively easy to drill holes and bolt the thing together with little effort. But I wanted more of a challenge. So I decided to use 16 gage, 1 inch square steel tubing. My brewing partner, Robert, claimed that he knew how to weld ("I took a class back in high school…") so we could rent a welder and put the whole thing together. After calling around I found some tubing for $.30 per foot and a 110V AC wire welder that I could rent for about $40 per day.
A short note on welding: As it turned out, neither Robert nor myself knew much about using a wire-feed welder, but we were brave (or maybe stupid) enough to give it a try anyway. We practiced for a while on some scrap metal and figured out the basics of using the welder. By the time we finished, we had actually laid down a few nice beads - but the majority were quite ugly even though they were functional. If you know someone who can really weld, see if you can rope them into helping you. If not, don't be afraid to try it yourself (assuming you've at least seen someone else weld before) - just be careful since you're dealing with a lot of electricity and a lot of heat... This is another of those areas where you should recognize when you've gotten in over your head and it's time to get someone else to do it.
After doing some sketching on paper, I cranked up a shareware CAD package and whipped up the following drawings of my system, which I later used during fabrication.
The first is a front view and gives a good idea of the overall construction and dimensions of the system:
Next is a top view showing the support structure for the kegs:
Note here that the right 1/3 of the drawing is actually the upper level - the second 16 inch bar from the right is duplicated on the lower and upper levels. If I were going to change anything here, I would make the center opening for each section about 2" larger all the way around (to 8" by 8") to keep the burners from heating the tubing.
The side view is rather boring and is omitted here - the overall depth of the rack (as can be seen on the top view) is 18". The bottom of the rack consists of 4, 16" pieces of tubing running from the front to the back. And on the lower level below the hot liquor tank, I welded in some pieces of tubing to allow me to lay stuff (like my counterflow wort chiller) on it.
The braces for the burners are made from strap iron, 1/8" thick by 1-1/2" wide. I took a piece 18" long and bent it 6 inches from one end, then welded the 6 inch section to the rear leg of the rack, angled at 45 degrees toward the center of the rack. More on the burners in a little while.
After all the welding was done, I cleaned it up with a 4" angle grinder (rented for $10), a wire brush and some sand paper. I shot the whole thing with a coat of heat-resistant paint to keep it from rusting - the heat resistant paint was a nice color and wouldn't burn off when heated, so I decided to use it all over.
For a finishing touch, I added a set of 4 chair glides (those little adjustable feet on the bottoms of most older furniture). The glides were metal and slick on the bottom so I could slide the rack around, plus they had a 1/4" stud sticking out of them for adjustment - I leveled the rack (we didn't weld it quite flat…) and locked them in place with a couple of 1/4" nuts. It looked pretty sharp when I was done!
The whole system is quite stable and would take a considerable effort to tip over - I was concerned about having the hot liquor tank so high, but even full of water it shows no signs of instability. The welding was less than perfect and the kettles are a little wobbly on the rack, but it's nothing to be concerned about. Otherwise, though, I expect this rack to last at least as long as I'm interested in homebrewing…
When we last saw our kegs, they had a 1/2" NPT coupling welded into them. The next logistical nightmare was figuring out how to plumb the whole thing. Stainless steel would have been the preferred option for someone with unlimited financial resources. If I ever meet that person, I'll let you know. I decided that copper, brass and bronze would be adequate for my needs (and a helluva lot cheaper).
For the insides of the kettles, Kirk Fleming recommends using a 1/2" NPT brass nipple screwed into the coupling, with a piece of 1/2" rigid copper tubing soldered inside the nipple (the inside diameter of the nipple is the same as the outside diameter of the copper tubing (5/8") - I still don't know why they call either of them 1/2"…) The nipple is threaded into the coupling from inside the kettle (to keep gunk out of the threads of the coupling), and one half of a 1/2" copper to copper union is soldered to the copper tubing about 3" inside the kettle. The other half of the union is soldered to a short piece of copper, an elbow and still another piece of copper to complete the dip tube assembly. The dip tubes extend to about 1/4" from the bottom of the keg, exactly in the center and allow you get all but about 1 cup of liquid out when draining the kettle.
The copper tubing soldered into the brass nipple sticks through the SS coupling to about 1" outside the kettle. Onto this, I soldered a 1/2" bronze ball valve (about $5 each). Half of another copper to copper union completed the plumbing for the kegs. Each one consists of this same configuration (except the hot liquor tank, which has a compression fitting instead of a union on the outside).
I had decided to use a single magnetic drive pump, mounted below and between the boiling kettle and mash/lauter tun. After looking around at several different pumps (Grainger, Cole-Parmer and others sell them), I wound up getting one manufactured by March Pump Co., the MDX-MT3 model. It's rated for 200F and pumps about 8 gallons per minute (gpm). While not technically listed as "food grade", it is advertised as being suitable for use with weak acid solutions such as fruit juice. I also talked to some of the engineers at March and they told me the components of the pump are impervious to nearly every chemical there is and I shouldn't have any problems using it with hot wort. The temperature rating is adequate, since that's the maximum continuous duty temperature the pump is rated for - I'll only be making short excursions above 200F and don't expect to have any problems. The pump was about $130 at a local March distributor.
A short note on pumps: A magnetic drive pump utilizes a magnet (duh…) to couple the pump mechanism and the electric motor - that way, the entire pump is separated from the drive motor and there's no chance of lubricants, etc. from the motor leeching into the liquid being pumped. The pumps are of the centrifugal flow variety, which means there's a gadget inside that spins and throws liquid entering at the center of the pump toward an outlet at the edge of the pump. What I didn't understand - and hopefully am making clear here - is that the pumps won't build up excessive pressure if you restrict the outflow, so even though a pump is rated for 8 gpm, you can safely restrict it to just a trickle. The thing you DON'T want to do is to restrict the intake side of the pump - apparently this makes the pump work way too hard and will eventually break it. Be sure to read the manual that comes with your pump and follow the directions and you'll be fine.
For the intake side of the pump, I could have used braided vinyl tubing, but was worried that it would collapse under suction and would generally be a bad solution. I also didn't want to have to mess with hooking up and unhooking the tubing from one kettle to the other. So I decided to use 1/2" rigid copper tubing between the boiling kettle, mash/lauter tun and the intake side of the pump - the ball valves on the kettles would be used to control which kettle was the source of input to the pump and would be either open or closed (they're not used to control flow). Plumbing this part of the system was basically an exercise in cutting copper pieces until they fit properly and soldering them all into place.
On the output side of the pump, I placed another ball valve (this one would be used for flow control) an a short piece of copper tubing to attach a length of 5/8" ID braided vinyl hose (around $1.25/ft.) which is then used to direct flow to whatever kettle I'm aiming for. The braided vinyl isn't a great solution (it's not rated for liquids over 140F, smells funny after you use it, and I've heard it tends to delaminate over time) but will have to do until I save up the money to buy some silicon or teflon tubing.
The pump has an open air-cooled electric motor, and is mounted to the front center leg of the rack. I used an old 3 lb. coffee can to make a cover for it so as to avoid splashing any more water on the motor than was absolutely necessary. To control the pump motor, I wired up a waterproof switchbox that has a lever on the outside which flips the switch of a normal household on/off switch mounted inside the box. The motor is wired to the switch and I added another 10 ft. of heavy duty extension cord and a waterproof plug on the other end. I can now safely turn the pump on and off with no worry of electrocution. Wiring cost: about $20.
For the hot liquor tank, I attached a 5/8" compression to 3/8" compression reducer, to which the sparge water sprinkler is attached. The sparge water sprinkler consists of a piece of 3/8" soft copper tubing that extends horizontally from the compression fitting to the far side of the opening in the top of the mash tun, then makes a 90 degree bend down into the mash tun, followed by another 90 degree bend back toward the hot liquor tank, and then into a large horizontal spiral that makes about 2 complete revolutions. I drilled a series of 1/16" holes in the spiral, about 1.5" apart, and crimped the end of the spiral with some pliers. The sparge water runs down the tubing and sprinkles nicely and evenly over the grainbed. My first prototype for this gadget had too many holes in it (they were spaced about .5" apart) and at the flow rate I wanted to use, all the water ran out of the spiral in the first 6 inches or so. The second prototype worked perfectly. If you build one of these, play with the hole spacing to get the distribution you like - it's better to start with the holes too far apart than too close together.
I decided to purchase a ready-made, folding stainless steel perforated false bottom from Sabco (about $60) for use in the mash tun - I couldn't find perforated SS locally at all, and the price for a square piece that I had to cut myself wasn't that much less than I could buy a finished one. The Sabco model folds in half for easy insertion and removal, and fits the inside of the keg perfectly. Another option would have been to construct some sort of EasyMasher-like apparatus, but I was concerned about grain scorching on the bottom of the kettle as well as the ability to recirculate the mash - the false bottom solves both of those problems by keeping the grain away from the bottom of the kettle and making a nice filter for recirculation. It works like a charm.
For the boiling kettle, a SS false bottom to filter the whole leaf hops out would have been overkill, but I did need something that would keep the pump from plugging up. I wound up using a variation of another of Kirk Fleming's ideas - you can buy an aluminum (Yes, aluminum. No, I don't expect to get Alzheimer's disease - I'll be more likely to die from cirrhosis of the liver from all this homebrew…) pizza pan from most discount stores (around $12) that's about 12" in diameter and is perforated with several hundred 3/32" holes. I bought one of these and bored a hole in the center of it for the dip tube, then placed it inverted on the bottom of the kettle - the curvature of the pan keeps it about 1 inch above the bottom of the kettle and makes a nice area of filtered wort for the dip tube to suck up. I added the dip tube and a SS hose clamp to keep the "mash strainer" stuck on the bottom of the kettle so it wouldn't float up during the boil. After the boil, I let the wort settle for 10 minutes or so, recirculate for a couple more minutes and then start pumping through the wort chiller. It works like a charm!
Speaking of the wort chiller, I made a counterflow chiller using an old 5/8" garden hose the dogs had chewed up, 15 ft. of 3/8" soft copper and a couple of 1/2" tees, caps and short pieces of 1/2" rigid copper. I based my design on some counterflow wort chiller plans by Drew Lynch but decided that 50 ft. of tubing was overkill. It turns out that 15 ft. is plenty - I can chill wort from boiling to around 55F with no problems using our winter-time tap water. Sanitization is accomplished by recirculating the boiling wort through the pump and counterflow chiller for about 10 minutes before starting to chill.
I purchased two Ashcroft 3" dial bimetal thermometers with a 3" probe from Sabco ($33 each) for use in the hot liquor tank and the mash/lauter tun. You'll recall that I had an extra 1/2" NPT coupling welded into those tanks, so adding the thermometers only required that I screw them into the couplings. For the mash/lauter tun, I stole another of Kirk's ideas and rigged it up so I could place the thermometer inline with the plumbing coming out of the mash/lauter tun and monitor the temperature of the recirculating wort. This was quite simple, and only required that I replace an elbow on the mash tun side of the plumbing with a tee, and add a 1/2" copper to 1/2" NPT (F) coupling (refer to the schematic below.) I purchased a 1/2" NPT brass plug to place in either the coupling on the mash tun or the copper coupling, whichever one wasn't being used.
The following schematics should give you a good idea how everything fits together. First is the plumbing for the boiling kettle, mash/lauter tun and pump:
Next is the hot liquor tank - this is pretty basic.
Finally, a side view of a typical kettle showing the false bottom and dip tube assembly:
Photo of Plumbing for Boiling Kettle / Mash Tun / Pump
Photo of Sight Glass Assembly and Thermometer Mount
I wanted each keg on my two-tier system to have its own burner and a way of controlling all of them separately. I've been using a King Kooker with the 170K BTU ring-type burner for the last year and was quite satisfied with its performance, so I called Metal Fusion (makers of the King Kooker, (504) 736-0201) to see if they'd sell just the burners. Much to my surprise, they do sell the burner assemblies, for only $15.50 each plus shipping. So I ordered two of them to add to my existing one and I was all set.
Since I already had the one burner, I also had a high-pressure LP regulator and hose, which has a 3/8" flare (F) fitting on the end. I didn't want to buy two more tanks and two more regulators, but was worried about whether I could run all three burners at the same time from a single LP tank and still have adequate gas flow. Having no way to verify whether I could or not, I decided to assume that it would work and went out to buy gas fittings. I had no idea what I really needed, and spent almost two hours at the hardware store laying fittings out on the floor and trying to figure out how it'd all go together (the salesman who kept trying to help me finally decided I was nuts and left me alone…) - I decided that it would be easier to use 1/4" soft copper tubing and the appropriate reducers to get from 3/8" to 1/4" from the LP tank to the gas line and then from the lines back to the burners. The one tool I had to buy for this part of the construction was a flaring tool which puts the appropriate 45 degree flare on the copper tubing to mate up with the flare fittings.
The following schematic shows how everything went together:
When it was all said and done, I had spent about $50 for the parts to plumb the gas lines - it probably could have been done cheaper if I had stayed with 3/8" copper all the way through, as the reducers and unions I needed to get back and forth between 1/4" and 3/8" were around $4 each.
The completed system works like a charm - I can crank all three burners to full blast with no gas shortage problems, and the needle valves allow me to control the flame down to just a trickle for temperature maintenance during mashing.
A word of caution is in order here: BE SURE to use some soapy water and check, double check and triple check every fitting on the gas line system for leaks BEFORE you light anything - some of it took quite a bit of extra tightening of some fittings to get them leak-free since I did a less-than-professional job of using the flaring tool. I'll leave the preaching about how bad it is to store LP tanks in the house to someone else, but suffice it to say it's not a good idea. And if you really don't know what you're doing, either find someone who does or at least have the work checked over by someone who knows about gas - it'd really suck to blow your house up because you didn't know what you were doing.
Photo of Burner Support and Plumbing
So now that it's done, how well does it work? I'll be updating this section of the document from time to time as I use the system, make modifications and try new things. Check back here every few weeks to see what's changed.
In the interest of keeping this document to a reasonable
length, I've moved this section to it's
own page.
Here you'll find links to other web pages describing brewing systems similar to mine and/or based on my design. If you've got a web page you'd like me to link from here, send me some e-mail!
Jeff Kane - As far as I know, this is the first system to be built based on my design. Jeff made some changes to the keg support structure in the frame which may be worth considering before you launch into building your system.
This document is copyright © 1996 by Marty Tippin (martyt@pobox.com). However, you may freely copy and distribute it for any non-commercial purpose, provided this copyright notice remains intact. I didn't pay anything for the information I used to design my system and you shouldn't have to either - if this document is helpful to you, then feel free to use or modify any ideas contained in it! Anyone wishing to publish this information or otherwise use it commercially (whether in part or in whole) should contact me for permission first.
By way of disclaimer, I am not responsible for anything boneheaded you do that gets you hurt, arrested, expelled or even deported. (thanks, Beavis & Butthead…) Use some common sense and know when you're getting in over your head and you should be fine. Please be extra careful when working with the gas as it's clearly the most dangerous part of this whole thing.
And if there's anything blatantly wrong or otherwise in need of correction in this document, or if you've got suggestions, questions or just need to unload on someone, please let me know - there's a link to my e-mail address at the top of this document. And if you decide to build a system based on this design, I'd appreciate if you'd send me some pictures and let me know how it worked and what you did differently!