Eco Sustainable Village

Note: Eco Sustainable village Foundation is studying the viability of such units to be used as a cottage industry in Eco Villages.

160-liter per day batch processor

Chuck Ranum runs Triple-R Recycling in North Dakota, and makes bio diesel from waste cooking oil for himself and for the members of a small local cooperative. They use it in their vehicles, for home-heating, for farm-tractors, and they can't get enough. Chuck will produce about 2,000 gallons this year, and is planning to increase production by four times, to 160 gallons a week.

Chuck Ranum:

My 1983 Chevrolet van has a 6.2 liter V-8 with 117,000 miles on it. I burn 100% bio diesel during the warm months of the year, and a mix of bio diesel and petro-diesel in the winter.

I've heated my home in North Dakota for over a year now on 100% bio diesel, and supplement with electricity. As North Dakota winters can get into -30 to -40 below zero F, we use 600+ gallons every winter. It's just a standard forced-air furnace, the ordinary 20-year-old pressure pump injection, spark ignition type that has been around forever. Homebrew bio diesel needs to be washed, or the excess methanol will eat the seals in furnace pumps in short order.

My main bio diesel feedstock is WVO. I collect it from one hospital, one bakery, three convenience stores, and five restaurants.

I picked up almost a dozen open-top 55-gallon barrels from an engine re-builder. He had chemical shipped to him in them, and they were headed to the junkyard. He gave them to me, and all of them came with rubber-edge lids. They don't need cutting and the top edge is rolled over for safety. One became my reactor tank, one became my preheat tank, all but one of the rest were fitted with hinges and latches left at different WVO producers.

Everyone was perfectly happy to get them, as grease pickup was costing them $40 a time. I haul the grease for free, and my barrels take up much less space than the old dumpsters.

One of the barrels is bolted onto my trailer. My '78 Jeep has an on-board air compressor. A standard York reciprocating A/C compressor can be used as both an on-board air and suction pump. They will generate over 120 psi, at a fast enough flow rate to run most air tools (engine at idle to 1200 rpm max.) and enough suction to flatten a 55-gallon steel drum like an empty beer can faster than you would believe!

I have one of the barrel lids fitted with a 1-1/2 inch ball valve, PVC dip tube, and 10-foot heavy-duty hose. I clamp the lid onto the WVO collection barrel, put the end of the hose into the barrel on the trailer, and pressurize the collection barrel with my compressor. The grease is forced up the tube, through the hose into the trailer barrel with no bucket-spilling or back straining at all. I can pump 'gloppy' grease, about the consistency of mayonaise, or even thicker. I take 50 gallons out in less than 5 minutes.

When I get home, I put the same lid on my trailer barrel and push the grease into my preheat barrel, through a mesh screen that takes out the french fries and broken spatulas.

In the winter I cheat, and drop off empty barrels. I haul the frozen-solid ones home and put them in my heated garage until they will pump.

I'm going to get a 250 gallon trailer soon, so I can start collecting from restaurants in the nearby cities as well.

The processing setup

My processor has a capacity of 42 gallons a day (160 liters). I will do 30 to 50 42-gallon batches this year. I run a titration and a test batch before every single one. We use a 1 hour mix time and 20 hour settling time.

The WVO comes from the trailer outside via the large grey 1-1/2 inch grease hose which leads into a bucket hanging over the preheat tank. This is the screen prefilter, made from a closed-top 5 gallon pail with a cut-out bottom replaced by steel mesh to get the big chunks. The lid is cut part-way, to allow scraping of the screen with a long wooden spatula while it is doing its job. The grease hose is inserted through the round opening, which holds it rather nicely. I can filter 50 to 100 gallons before cleaning, which is no more than tapping on it upside down and shaking the chunks out through the larger opening in the top.

The 55-gallon preheat tank (middle) has the lid propped open here. Both the preheat tank and the reaction tank (left) are insulated with Reflectex brand insulation (basically bubble-wrap sandwiched in between aluminum foil). Both tanks have a 1500-watt heating element installed on the back side, wired to an ordinary water heater thermostat attached in the middle of the barrel. The thermostats are simply held in place by a few wraps of black electrical tape that goes all the way around the barrel. It might not sound secure, but it works very well.

Four inches up from the bottom on both tanks I cut a hole and welded a short piece of pipe fitting, sized to fit the electric water heater elements. This required enlargement with a pipe-tap to allow the element to seat completely, preventing leaks. Correct wiring diagrams come with new thermostats.

I have to drain and scrape the bottom of the preheat tank every 400 gallons or so to keep the buildup from reaching the heating element. The element itself has a round piece of screen surrounding it to limit food particle buildup.

The oil leaves the preheat tank via a valve 4 inches above the bottom to miss the settled chunks and water, and to a 40 micron filter assembly and down to a pump that sends it to the reaction tank through a large black hose. This is just a cheap clear water pump out of a catalog, and seem to work just fine in this application.

The reaction tank is also a standard 55-gallon open-top drum. I stood the barrel upright on soft ground, put a six-foot long 1"-diameter pipe in the middle of the bottom, and put the full weight of a tractor loader bucket on the top end of the pipe. This bowed the center of the bottom about 1". I drilled a small pilot hole dead center and used a step drill to enlarge the hole to 1 inch. A standard 1-inch tank fitting is welded to the bottom. A short nipple threads into this. A ball valve shutoff is connected directly to a dual valve. One hose is the glycerine drain, the other goes to the pump, which sends the bio diesel through a 10 micron diesel fuel filter, and on to the wash tank. This is also a cheap clear water pump.

I use a chemical resistant hose between the reaction tank and the pump, and between the pump and the wash tank. The methanol content in the fuel destroys ordinary Goodyear fuel line in a few months. I use ordinary fuel hose for the glycerine drain, it gets gummy pretty fast, but holds up well enough for that purpose.

The reaction tank mixing motor was salvaged from a hospital boiler room renovation. It is a sealed-type unit, safe for mixing volatile chemicals. It's 1/3 hp, mounted on the wall above the reaction tank, with a commercial stainless steel shaft and 3-blade 4-inch diameter mixing propeller, 1/3 up from the bottom, off to one side 3 inches, and set at about a 10 degree angle. It was all surplus, so I got it for free.

The prop spins at a high enough speed to "howl", even 28 inches below the surface in 42 gallons of oil. The sodium methoxide is injected slowly over a 10-minute period directly into this vortex. This seems to be a violent enough mixing action, as I've had almost no failures in over 2000 gallons of product so far. I use a factory-made slip-on plastic lid, and cover the slot for the shaft with a plastic bag. Very little vapor escapes.

I buy seven 55-gallon drums of methanol at a time for $100 per drum, plus a $20 deposit per drum. I store them outside city limits.

I made a dip tube/valve assembly that fits one bung on the methanol barrel, and put a tire valve in the other bung for an air-hose. Methanol is pumped from the methanol barrel to the sodium methoxide mixing tank.

The 10-gallon sodium methoxide mixing tank was a stainless steel cannister from an old coke machine, also salvaged. The mix motor is also salvaged. The air pressure supply tube is coiled hospital oxygen tubing, salvaged. It allows me to move the methoxide tank around without having any hoses on the ground to get stepped on.

The tank comes with a stainless steel dip tube that reaches almost the very bottom. This is hooked to a teflon and stainless steel 1/4" line that feeds into a stainless steel fitting at the top edge of the reaction tank, right next to the 1-inch hot-grease inlet. I tapped the dip-tube fitting to accept the threaded end of the feed line. The other fitting was a seal on the gas inlet (these vessels operate with CO2 pressure). I modified it to accept the stainless steel propeller shaft.

I pump 8.4 gallons of methanol into the mixing tank and then pour the pre-measured NaOH into the methanol while it is agitating, over about 5 minutes. This is the only time the possibility of contact with sodium methoxide exists, and I wear safety equipment. I seal the vessel with the original cap and continue to mix for 5 minutes. Then I pressurize the tank with a tiny air pump (7 psi). This forces the sodium methoxide up the dip tube, through the teflon/stainless line and injects it into the stirring hot WVO in the reaction tank without my being exposed to it.

The plastic glycerine jug is a 5-gallon jug that the restaurants get their grease in. I have one of them save them for me. They are nice and clean, and the glycerine drains into them and sets up nicely. Later, they can be cut open very easily, to allow the solid glycerine block to be dropped into a melting tank to be fed to our new, preheater equipped, used oil furnace to heat our building next winter.

The wash tank is made from an 80-gallon water-heater. I cut the bottom off, turned it upside down and welded legs on. It had fittings in all the right places, and has a beautiful bowl bottom. It would, in fact, make a great biodiesel reaction tank! The steel is far thicker than a barrel, and there are fittings for heating elements already in, as well as a fitting on the bottom (was the top) that valves could screw right into. I use a large aeration stone I bought from an aquaculture supply house (attached to the yellow hose). I intend to get rid of the hose and attach the stone through a fitting on the bottom of the tank. In order to do this, you must use a check valve, or water and fuel will backfeed to your compressor as soon as you shut it off!

During the summer the washed biodiesel goes into black barrels in the sun to get rid of the last of the water.

I just made a deal to put 5 Lennox solar water heater panels on the roof (surplus again, circa 1978). Using a small circulating pump and a coil of tubing in our WVO preheater, we should eliminate about 50% of our "wasted" electricity use on the preheat side. If this is sucessful, we may do the same for our reactor vessel. This would almost bring us to the point that we could supply our remaining electricity needs with a few solar panels.

How to make a cone-bottomed processor

Maybe the best thing (the only good thing?) about fossil-fuel petroleum is that it comes in 55-gallon oil drums, which duly become empty, and magically transform themselves into the worldwide mainstay of appropriate technology and do-it-yourself tinkering. And of backyard biodiesel-making -- the perfect mixing vessel.

Well, almost perfect. They need a bottom drain, and for perfection the drain should be at the end of a cone, replacing the flat bottom of the drum.

Here's how to make a cone bottom for your 55-gal oil drum biodiesel mixer.

You need cutting gear to cut out the drum bottom and cut the cone to shape from a piece of flat steel sheeting, and welding equipment to join up the cone and weld it to the bottom edge of the drum.

The problems tend to arise when drawing out the shape of the cone-to-be on flat steel. Trial-and-error doesn't work very well, usually leading to a lot of annoying grinding to make it fit, or a botched job, wasted steel and wasted time and effort. This is how to get it right.

You'll also need a calculator and a big compass.

If you don't have a compass big enough to draw the circle you'll need, you can improvise one from a steel nail, a felt-tip marker pen and some cord that won't stretch. Tie the cord tightly to the nail at one end and the pen at the other, close to the point in both cases, with the length of cord between them equal to the radius of the circle you're going to draw. Take some trouble trying to get it precisely the right length. Make a little dent in the centre of the steel sheeting for the point of the nail, then carefully draw your circle, holding both pen and nail firmly vertical.

Use 16- or 18-gauge flat steel for the cone. A 3/4", 7/8" or 1" valve will be fine.


(Drawing not to scale.)

You can adapt this method to any size of drum just by changing the measurements and calculating accordingly.

The outer-edge diameter of a standard 55-gallon drum is 22-3/4 inches -- check it, this is the critical measurement.

How deep do you want your cone? The deeper the better, because the deeper it is the steeper will be the sides, and the better it will drain. If it's 12" deep, the sides of the cone will have a slope of just over 45 degrees, not very steep. At 15" deep the slope is about 52 degrees, steeper. But deeper and steeper also means the whole contraption will be higher. The drum is nearly 3ft high, plus 15" for the cone, plus another couple of inches for the valve, a couple more for a hose connection, and you need enough space to put a bucket underneath -- another 12" at least... that's 65" already, up to your chin if you're a six-footer.

So we've taken a 12"-deep cone as an example. That's the second critical measurement.

In the diagram, "r" is the radius of the oil drum bottom, measured to the outer edge -- that's half the diameter, 22.75 ÷ 2 = 11.375".

h = 12" -- "h" is the height (depth) of the cone.

sh = 16.535" -- "sh" is the "slant height" of the cone, the length of the sloping side.

You calculate this by Pythagoras's theorem, which states that the square on the hypotenuse of a right-angled triangle is equal to the sum of the squares on the other two sides.

If you're among the math-challenged, relax, it's easy -- your calculator will do the work for you.

The red triangle in the diagram is a right-angled triangle -- the angle between the two sides "r" and "h" is a right-angle, 90 degrees. The side opposite the right angle is "sh", the slant height of the cone, and that's the hypotenuse. To calculate its length, square the other two sides. That means multiply 12 x 12 = 144; multiply 11.375 x 11.375 = 129.390625. Add the two answers together: 144 + 129.390625 = 273.390625. That's the "sum of the squares on the other two sides", which equals the square on the hypotenuse. So the square root -- on the calculator -- of 273.390625 is the length of the hypotenuse: 273.390625 = 16.53452827. Shorten it to three decimal places (add one if the fourth figure is 5 or more) = 16.535". That's the slant height of your 12" deep cone -- the third critical measurement.

To make a cone all you have to do is draw a circle on something flat, cut a pie-slice out of it and join up the two straight edges. But if you want the bottom of the cone to fit something specific, and for it to be a particular height, you have to know how big to draw the circle, and how big to make the pie-slice.

The radius of the cone circle you have to draw is the same as the slant height of the cone -- sh, 16.535".

So now you know the size of the piece of flat steel sheeting you'll need: 16.535 x 2 = 33.07" square. So make it 3ft square.

Take a long rule, or anything with a straight edge that's long enough, and draw a line across the plate from one corner to the opposite corner, and a second line joining the other two corners. Where they cross is the centre of the plate. Draw your circle from that point, radius 16.535". Okay, so you can't measure .535" on your ruler. It comes to 17/32", but 9/16" will do; 5/8" will also do, or make it 16-11/16", which will give you a small margin of error.

Now you need to know how big to make the pie-slice cut-out.

The edge of the cone must fit the drum. So the cone circle minus the pie-slice must match the circumference of the outer edge of the bottom of the drum. Circumference = diameter x pie -- that's on your calculator, or just multiply by 22 and divide the answer by 7.

The diameter of the drum is 22.75", circumference is 22.75 x = 71.5".

The diameter of the cone circle is 16.535 x 2 = 33.07, circumference is 33.07 x = 103.934".

The arc (a section of the circumference) of the cut-out pie-slice sector is 103.934 - 71.5 = 32.434".

Now you need to know the angle of the cut-out sector so you can measure it off with a protractor. Divide the arc by the circumference and multiply by 360: 32.434 ÷ 103.934 x 360 = 112.34 degrees. Make it 112-1/3 degrees.

Now you can draw in your pie-slice that you're going to cut out. Here's a protractor you can print out if you don't have one.

Useful to know: the volume of a cone is 1/3r²h. For a 12" cone on a 55-gal drum, that's 7 gallons (US).

Two more things to consider: first, you might want to leave a tab for some overlap when welding the two edges of the cone together, as marked by the dotted line on one edge of the pie-slice in the diagram.

Second, you might want to pre-cut a hole for the valve, as in the diagram. Or just saw off the point of the cone at the right height after it's assembled. Depending on what kind of valve you get, you'll either weld the valve on direct, or it will screw onto a short length of steel pipe, and you'll weld the steel pipe on. Whichever, stand the valve or pipe on top of the cone, straight up, and draw a line under it around the top of the cone where you'll make the cut. Make sure the hole is narrower than the outside diameter of the pipe/valve so you'll have something to weld it onto.

If you'd rather pre-cut the hole, measure the ID (inside diameter) of the pipe or valve, add half the thickness of the pipe or valve wall. Say the answer is 7/8". All the angles are the same as for calculating the cone, so you can do it proportionately: 7/8" is the diameter of the hole, divide by 2 for the radius: 7/16 = 0.4375. It's the slant height you want: 16.535 x 0.4375 ÷ 11.375 = 0.636", which is slightly more than 5/8", so make it 5/8". Draw the circle for the pre-cut hole for the valve with a radius of 5/8".

Before you start cutting and welding, double-check all your calculations and measurements. When you're satisfied, make your cuts, then weld it all up and you're done. You'll find it a lot easier to bend the cone evenly to shape if you use a roller of some kind. This is actually a good reason for pre-cutting the valve hole because the roller can fit through the hole while you're rolling it. Try a piece of 7/8" steel waterpipe about two feet long or more.

You'll also need a stand for it. Steel piping or angle iron will do, four legs, firmly welded to the sides of the oil drum and joined by cross-struts just above the level of the valve. Make it strong -- a full processor will weigh more than 400 lbs.

A useful refinement is a sight-tube, leading from near the bottom of the cone vertically up the side to near the top of the drum. Use translucent PEX tubing (crosslinked polyethylene) or high-density polyethylene (HDPE), thin-walled for better visibility, both of which are heat- and chemical-resistant; 1/2" or more ID, preferably more. Get right-angled plumbing fittings, drill holes for them and weld them in. Use strong stainless-steel gem clips to connect the hose to the fittings, a good idea to use two clips at each end, with heat- and chemical-proof silicon sealer as a gasket compound.

A final refinement -- make it last. Mild steel drums, cones and fittings rust, especially in contact with sodium methoxide, and sulphuric acid, if you use the Foolproof acid-base bio diesel method. Depending how much you use your processor, it might not last much more than a year unless you protect it.

We recommend POR-15. We've used it and it works exactly as claimed -- no more corrosion. It's not very cheap, but cheaper than replacing your processor, cheaper than stainless steel too. You should only need a pint or two.

[ Home ]

Send mail to ghazi@wavepowerplant.com with questions or comments about this web site.
Copyright © 2006 Wave Power Plant Inc.
Last modified: 04/08/06