Being a Wizard, once I heard rumors of the holy grail, I had to begin the search. Don't be surprised if I add a solar panel to the rig too
A clear water pump would likely work well to provide both the required 90psi, and some heating. Unfortunately it might also emulsify any water.
My local farm supply has adjustable spray-rig pressure relief valves that can easily regulate line pressure to 90psi, recirculating the excess oil back to source. They only cost $12 each.
I may order a Dieselcraft centrifuge to play with. You're a wizard, indeed.
One trouble with many water pumps is the temperature rating, which depends on what kind you are talking about. I like a gear or vane pump the best since most can take up to 210 F, which is good if you want to run multiple passes. And good for the holy grail discussed above.
Gear and vane pumps don't mind the extra viscosity of cold WVO, while many types of water pumps don't like it at all.
Using as a FE -- Steam tables indicate that at 90 pounds pressure water has to be hotter than 331.1 deg f to boil. Your flow rate of .9 G/M will take a lot of heat to get the oil to that temp prior to the centrifuge (roughly 11,000 watts, takes 50 amps at 220 volts to get this much heat) but if the unit can operate at that heat (probably can) I would think it would work the same as any FE setup. The water should burst into steam as it exits the units 2 output nozzles.
Clear water pump pressure -- This is a standard centrifigul pump design, even with there output completely blocked they will likely make no more than 40 pounds pressure, usually not more than about 30 pounds pressure.
The power steering pump Idea is probably the least expensive approach, I didn't expect they would output this high of a flowrate at pressure, will check these out myself. Most of the power steering pumps on older american vehicles are actually vane style pumps rather than gear pumps but they may work also, will have to test this.
Don't forget that you are not dealing with pure water. Unless you have washed the wvo before this step it will be a cocktail of water,sugars, salts, acids, etc.
Thanks for the good engineering data. That makes it seem like the holy grail is within reach. Maybe on a vehicle, you could do a FPHE preheat before the centrifuge/flash evaporator, and use much less than 11KW. What starting temp were you using for that figure? How many KW would it be if the starting temp was 160 F for example?
Another option is you could do multiple passes instead of the 11KW, which of course gets you superfine filtration to boot. As long as its not causing an emulsion that the flash evap can't handle. Have you tried running an emulsion through your flash evap?
And of course the fun mist wash before this which Dana wants to be sure we dont forget
You made me realize this is probably a vane pump too, I never took it apart. So I dug up my old mitsubishi shop manual, and yes it is a vane pump. And it can do 1000 psi with blocked flow, thats what they spec for testing it. And they say don't completely block the flow for more than 10 seconds (they don't say what happens then, probably pump damage.)
Today I ran this rig for 45 minutes with the same oil (= 3 passes) and got a small amount more black goo out of it. Now its really clean.
This heated it up to 120 F for mistwashing. Washed for several hours using about 3 gallons of water, the water came out totally clear and neutral PH. Complete seperation between oil and water, no emulsion. I should stick with this restaurant.
Tommorrows test will be to run this oil again now that it has a small amount of water added and see what happens. I will leave some of the water in the barrel and be sure to stir and suck it into the centrifuge.
I did some calcs and starting at 160 F, it would take 7.78KW to get the oil to 230 F which at my altitude ought to be enough to make it flash evap (boil = 200 F.) This is not counting that the pump and centrifuge are adding a little heat to the oil also, since they both heat up about 10 F higher than the oil. I will test this once I get the seal on the second water heater element working.
If Dana was right and water through this rig makes an emulsion, this will see if a flash evap will remove that emulsified water too.
11KW -- This was a realy quick estimate based on my FE. I am applying 1125 watts of heat (220 volt 4500 watt heater running on 110 volts) at about 90% duty cycle (thermostat only shuts off for short time, then back on again). the flowrate in my unit is 5G/H, your flowrate is about 50G/H, should need 10 times as much heat.
My concern with applying that much heat is that you will likely polymerize some oil on the surface of the heating elements, this plasticised oil will eventually flake off and can clog an arifice. I see this problem even with only 1100 watts but the plastisized flakes only cause a problem when heating up the cold heater when starting a dewatering run (does it every time though), this is the reason I now have a high-temp filter just before the orifice, much less burnt fingers from unscrewing the clogged hot orifice.
In my unit the oil is raised from ambient temp up to approx 300 deg f in a single pass. Measuring the temp of the oil as it exits the flash orifice will indicate if you have enough heat energy stored in the oil for complete dewatering of that pass (flashing water to steam takes its energy from the heat in the oil, this cools the oil). If the exiting oil is above 212 deg then all the water should have flashed off and left extra heat in the oil. On a few occasions I have had to put 325 deg in the oil to insure hotter than 212 deg f oil out, this was with oil that looked clear (no visable water). I would think an obviously visable emulsion will contain a LOT of water, way more than I have ever tried to dry, This will take a LOT of excess heat or several passes to flash off I would think.
It will get interesting with that much water, I had only a couple drops of free water get into the heater pipe once, made so much steam that I could not see the flash evap unit, scared me enough I ran..
The amount of excess heat above boiling is based on the amount of water you need to flash off, more water needs more heat to keep the exiting oil above 212 deg to insure all the water is turned to steam in a single pass, multi-passes should allow less heat needed for each pass, just never thought about it before as I was shooting for dry oil on the first pass.
All of my oil is water washed and settled warm for at least 7 days (14 if doing a baking soda water wash), usually way more than that, then recirculated warm continuously through several filters for at least another 72 or more hours before it goes through the FE, there is absolutely NO free water visable in it, This oil almost always produces wisps of visable steam with an occasional momentary burst of steam, even on the rare occasion when no steam is actually visable I have held a cool mirror to the steam outlet, it foggs instantly and tiny water droplets form in only a second or two.
It is odd but the water is not evenly distributed in the oil, watching the impact point of the exiting hot oil stream against the bash plate I see the impact area constantly changing size as more/less water is flashed off, you can even hear the change in the audable "hiss" from the steam release.
Power steering vane pump limitations -- These must be able to make high pressure, the older units on my vehicles have a pressure bypass, don't know what it is set for but if you turn the wheels all the way to one side plus a little you can hear the pump change tone as the bypass opens.
I suspect the main limitation with these is overheating. Popular Mechanics magazine once had a set of plans available that used these pumps to power a hydraulic cylinder for a hydraulic lifting/dump unit for a pickup or trailer bed. It was slow due to the limited volume of oil the pump would move but it worked. Feedback from folks building the setup indicated that pump life was short due to overheating when using a power steering pump.
Overheating shouldn't be a problem at 90 psi and 1gph. They operate at much higher pressure in a vehicle. I've used PS pumps before for pumping and never saw any issues. Using it as a hydraulic pump would put a much heavier load, equivalent to cranking the wheel back and forth fast while not moving. 90psi is more along the lines of the load it sees on the highway.
No problem with overheating the pump, like JOAT says this is an easy load for it.
No problem with clogging, since any chunks from the heater or anyplace else get stuck in the centrifuge right before the orifices. I proved that since I found lots of shiny metal particles in the rotor goo on my first run, which I suspect were the (probably zinc) coating on the water heater elements flaking off in the first 30 minutes. Now they are a dull grey instead of the shiny look they had when new.
I still don't want to add heat so fast that it polymerizes the oil since who knows what other problems that adds. But multiple passes is probably how I will operate this, so I think I can get the heat up there with 2- 4500w elements.
Catching particals before the orifice -- So obvious I overlooked it.
90 pounds pressure -- Probably right, to low a pressure to cause much of a strain, good, this gives us another source of cheap pumps.
Polimerization -- I suspect the only problem might be in direct contact with the metal of the heating element, it is the hottest thing as it is slowly (relatively) transfering heat to the oil, as long as the centrifuge catches any chunks I guess you can apply as much heat as is nescicary. Might have to pull the heating elements every so often and clean/scrape them off to insure the best heat transfer ?
EUREKA! (is that what a wizard is supposed to say when an experiment succeeds?) I ran the same 15 gallons of oil which now after mistwashing had lots of water, visible water, left in there on purpose to see whether it can remove it. Mist washed in the same barrel the centrifuge runs in.
I hooked up a 2x6 concentrating parabolic solar panel to add the equivalent heat of my 2nd 1125w heater element which I don't have installed until the seal is fixed. Plus I just like solar and had it laying around.
Right away it formed a terrible emulsion, with 1" of beer foam looking head, and I thought I just ruined the cleanest WVO I had seen (clean since it had previously been through the centrifuge for about 2 hours = 6 passes.) The milky emulsion cleared up in about 15 minutes of running. It started puffing out many clouds of steam really quick at about 120 F, much more than the first time I ran the oil. Note -it had not puffed any steam after 15 minutes into the original run a few days ago, which is another thing that made me suspect this was steam and not atomized oil. It also fits Tim's description of how his flash evap. steam releases, you can hear little random puffs just before each steam cloud is visible.
So I left it run for about 2 hours, it slowly got up to about 200 F, and I waited until it was not puffing anymore, then did the pan test. No bubbles at all, EUREKA! no water. This thing just might be the holy grail! More testing to come...
Leaves the question why would it dewater so well at a low temp?
I haven't seen it up close, but assume the jets that drive the centrifuge are right by where the oil exits. Normally they would be submerged in engine oil being piped back to the engine. However, in this app, since they eject more or less directly to the atmosphere, perhaps there is an extreme low pressure point occuring as they exit the fast spinning nozzle, allowing the water to boil off at a low temp as though in a partial vacuum?
Awesome Holy grail!!! This is coool. Thanks for sharing.
low temp steam -- Looking at the steam table, water will boil at about 120 deg f at around 26.5 inches of mercury (12.77 pounds of vac) this is a very hard vacuum to reach but don't know, spinning nozzles with high-speed flow may produce it ? At 200 deg f it only takes 6 inches of mercury (2.94 pounds vac) for water to boil, this is a pretty light amount of vacuum so it would be much easier to produce by the ventury action of the high-speed jets.
I always see a bit of low temp steam when I first fire up my FE unit and have wondered about it. The oil is not up to temp yet when first firing it up but there is always a bit of visable steam being released. I intend to test this more, made a few posts wondering if some of the energy to produce the steam might be coming from the pressure drop across the orifice, no one commented on this at the time and I don't know the physics that might be going on here, seems reasonable that any type of energy, either heat or pressure drop, could be used for steam production (both are stored energy) but I don't have a clue how to equate a pressure drop into BTU's or some other energy equivelent to calculate its use in flashing water into steam.
Once I get time I intend to do more testing by dropping my temp down in increments until I don't get dry oil, then start increasing the pressure to see if the added pressure will boil off more water. Was thinking that it might be possible to use pressure-only on ambient temp oil. For testing at more than about 200 pounds pressure I will need a sturdier unit than my present schedule 40 pipe unit. My thinking is to use a true hydraulic pump (power steering pump probably, driven by a salvaged 90 volt DC veriable speed treadmill motor) and normal high pressure hydraulic hose, simply put an orifice at the end of the hose and see what happens.
It sure would be nice if total dewatering can be done (on a visable heavy emulsion yet) with only 90 pounds pressure and a bit of heat..
The vacuum theory sounds reasonable, the high speed jets of oil might be working sort of like a ventury (jet) pump, small high speed jet of oil pulling the air away from around the nozzle faster than it could rush in from the outside, high speed spin may be helping with this by creating an outer boundry layer of air ? The max vacuum possible will only be about a 15 pound pressure reduction so the boiling point should be the same as if you increased the pump pressure by 15 pounds on a static orifice.
High heat is a power hog to produce and is the hardest thing to supply a lot of for a mobile unit, pressure is fairly easy to make using a lot less power (but this "less power to make" probably also equates to "less power to flash off steam", power-is-power), would be nice if dewatering can be done using radiator water temps at some higher pressure, polishing (fine filtering) using the centrifuge in the same setup would be a plus, especially as it also eliminates orifice plugging. Probably still will need some sort of course filtering before the oil reaches this stage but it sure sounds promising.
Quick ? Will this method remove sugars and acids or just water? Seems washing removes the acids and sugar. Any help appreciated.
What did this experiment use as a source of the wet oil - mist washed oil!
I don't know the physics either, but was guessing similar to what you say. Another form of energy is that the nozzles are spinning at 6000rpm, so its different physics to your setup. It definately is steam, it even sounds like ssssss when it releases.
And it seems to produce steam just as easily at 120 F so could be a huge energy saver over getting it to 300 F.
You mistwash before this. My rig dewaters and filters. (edit- see page 12 it may be possible to remove sugars/acids without a mistwash if you capture the output at pump shutoff)This message has been edited. Last edited by: SunWizard,