My experiments on burning neat vegetable oil in this machine were not a complete success. I'll be honest, I had hoped to fit this burner to our boiler but despite all the hard work and careful design, other problems appeared.
Ignition could be adjusted to be 100% reliable from about 150ºC electrical preheat upwards. The flame is instantly stable with a clean, frighteningly sharp cutoff with the Protec valve fitted, but the pump pressure and air setting have to be increased substantially to get a good clean flame, which of course increases the oil consumption, flame length and power.
At 10 bar [150psi] , occasional Roman Candles, [ larger, slower-burning droplets emerging from the flame tip, too big to burn in time inside the flame] occurred. These are furnace varnishers! Increases in oil pressure solved that problem, only to create another. A very fine haze of unburned microscopic 'motes' of oil appeared, [clearly visible in the darker 20 bar flame pic, where I deliberately increased the pressure to get a good pic ] , as a 'halo' surrounding the flame.
Theoretically, the oil droplets need to be between 20 and 40 microns in diameter. The very nature of the vegetable oil appears to be the problem, for even when heated [to 150ºC or over] to obtain a viscosity similar to diesel oil, the long-chain molecular nature of the fuel seems to produce droplets in a wider range of sizes than the ideal and they take longer to burn. Above 40 microns, tendency for Roman Candles. Below 20 microns and the micro droplets are travelling so fast they can go right through the flame without burning at all!
The killer problem, however, is the necessary burner idle times, when vegetable oil can set like varnish on or in the nozzle itself. I lost two nozzles within days, clogged with resin overnight. Even though I get my nozzles from the States legally tax and duty free via the internet at US$4 each, [ US $8 for a Protec insert ] such losses are unacceptable. I took to cleaning out the Vegetable Oil from the burner to conserve nozzles by running a neat diesel wash through the pump at the end of each experiment, which turned out to be a EUREKA MOMENT!
Accidentally burning a small can of wash-oil, [after my carefully annotated oil-soaked tape labels all fell off, the story of my life!], I found that vegetable and diesel mixtures can be burned without preheating at all. Trials, using cleaned dog-meat cans as fuel tanks, with the pump flexibles dunked in them, established: 10% VO in diesel burns superbly; no changes in pressure or air required. 20%; a slightly longer flame, air up a tad. 30% required further slight increase in air. 40% produced rare Roman Candles, but increased pump and air pressure solved that, even longer flame.
50% and 60% VO required slight [80ºC] preheating or an initial 15 second neat diesel burn to ignite, produced either a smattering of Candles and/or a slight oil haze. At 70% VO, 100ºC preheating was essential for assured ignition, both candles and/or haze a lot worse. Within wide adjustments, up to a 40% VO cut can be burned consistently, with no ejection of unburned oil.
Back at the Drawing Board, a cunning plan began to evolve. I now intend to use a regular, unmodified pressure jet burner, with a regular off-the-shelf, low power [150W] Danfoss FBHB5 inline 70ºC oil preheater lifted from a doner burner [for consistent fuel supply parameters] and to burn a mix of VO and #2 diesel at about a 33%-40% VO cut. Fitting the Danfoss LE pump and Protec anti-drip valve will aid cleaner burns. I reckon I'll need to increase pump pressure slightly to about 11-12 bar, but have ordered some 0.4GPH x 80º hollow-cone nozzles from my supplier in the States to cut back the power slightly.
This is because a VO/#2 flame is longer and there is a limit to the size of flame that can be accomodated in a given boiler firebox. The visible flame should not impinge on boiler tubes or furnace plates, btw, as this causes sooting. To mix the oils, I am building an insulated, 150W heated [@ 50ºC] 30-liter 'day-tank' in stainless steel for the VO, [to prevent any fat granules in the used oil clogging filters], supplying a small variable-output washing-machine grade peristaltic pump. This will accurately dose about 10-15 ml of VO per minute, teeing into the Burner's oil pump diesel inlet line, via a small additional 200 mesh filter = a 33-40% mix at c. 20kW power.
The Danfoss pump will be reconfigured to run single-pipe supply, to avoid pumping VO back to the diesel tank from the LE pump's relief valve. The day-tank will need refilling about every 7 days or so. A simple delay circuit at burn-end will allow the burner to run on for about 60 seconds on diesel only, thus flushing the system at each burn, allowing an assured diesel-only start and preventing nozzle varnishing. Plus, 100% #2 oil can be burned if required at the press of a switch.
Here's the schematic:
Hopefully, the new burner will be fitted and running our heating this winter. More pics and results anon....