I've been wanting a centrifuge, but they're too pricey for me. I considered making one from scratch and had mocked up the plans, but hadn't gotten around to making it happen. I picked up a large used medical centrifuge thinking that'd start me on the path to modifying it for WMO processing. My biggest reason for it is to remove water, removing some solids is a bonus. I haven't yet built the continuous flow rotor for it, for now I'm using the stock baskets and running tests to see what it'll do. It's currently making about 2500 G force and is pretty impressive, but I was hoping it would separate emulsified or suspended water. (I'm not sure what the proper term is for oil with water fully mixed in that makes it cloudy and doesn't seem to want to separate.) I didn't think it would, but the sellers of centrifuges say their's will and they don't make more G's, most are less than mine. I've played with a few samples and it's not looking good. Once it's modified I expect somewhere near 4000 G. That would speed up the process, but won't change physics. Does anyone know if it's possible to remove this water by gravity? I assume my samples of 'emulsified' oil have something mixed in that's allowing the water to stay in suspension with the oil, alcohol perhaps.(?) Water that's just kinda mixed in does separate really fast. I don't even know for sure that this suspended water would be an issue. As far as combustion goes I'm sure it'd burn, just unsure of it's effects on the pump & injectors. As a note; I've run some pretty heavy gear oil mixes with no preheating at temps as low as 55F. and, to my surprise, it has no noticeable effect on performance.

One other thing I've figured out is that centrifuging oil is NOT the last step. Everyone focuses on the centrifuge removing all particles by gravity, what about contaminants that are lighter than the oil?!? I plan to make my 1 micron filter the last step anyway, but added that for the folks who may not think about it. The centrifuge sellers don't seem to address that at all and I've found lots of stuff floating on my samples. In a continuous flow type fuge this stuff would likely end up mixed into the "clean" oil undetected and pumped into the tank.

 

Need Don of one of the other Guru's to chime in on his one. I do think that having a final filter is a very good idea. I run mine through sock filters down to 1 micron, then use an oil type filter (10 micron, need to get a 2 micron) to clean it up.

 

I bought a used centrifuge and worked with it quite a bit, then sold it. Not really what I was looking for. It won't remove "suspended" water at all, just "free" water. I can do that with settling. I've been working in my lab with ultrasonics lately and have had some success with cleaning up "cloudy" bio mixes. I haven't tried it with a normal batch, just lab sized batches. My plain is to work with it some more this winter, but it looks promising....

 

Fabman, you're likely more educated than I on this stuff, is there something that could be added to make suspended water fall out of oil? I have some different oils, mostly gear oils, that have suspended water. I had the thought that possibly adding lots of water might make it happen, no idea, but with little to loose I'll probably try a batch tomorrow and fuge it to see what happens.

 

Why do you say it won't remove suspended water? I have been reading over at the biodiesel site and they seem to think they get "dry" oil using a centrifuge. Also watched some over at youtube.

I haven't paid any attention to veg oil developments for about five years but I was thinking the centrifuge looked like enough fun to get me back into it. I still have a Mercedes.

 

The only way to tell if the suspended water has been removed is to give it the "pan" test. EVERY batch that I ran through my centrifuge failed the test. As a final test before I sold the centrifuge I ran the oil through 4 times in a row, and it still failed. Not worth my time or trouble.

 

Is this a gravity-fed CF - similar to Simple CF or WVO designs?

What is the temperature of the oil you are processing?

Yes, some emulsions are tough to break. I dunno much about WMO (except that it is not a fuel for MY truck ), but hydrogenated and/or high-fat content VO is MUCH more difficult to dewater than non-PHO oil. Some of the other 'nasties' in your WMO may be making a tougher emulsion to break. In a nutshell, all CF is doing is expediting 'settling' process. There are some terms I can't remember describing the effectiveness of CF'ing a liquid. It is a factor of 'contact time' in the CF and includes G's and flow rates...

As for CF effectiveness, I finally bought one (WVO designs) after seeing several in operation as I've traveled around the country. Most of the people I know are running 3400rpm rotors and putting between 2gals/hr and 10gals/hr thru them. A final filter is simply just a good idea, but one guy has 1000's of gals thru a 1mic bag filter (in a housing).

IMHO, the pressure-type CF's like Diesel Craft are not nearly as good for VO. The biggest problem is they have a tendency to increase the rate of polymerization.

 

I haven't pre-heated any samples because I assume that would only speed up the process, not an issue to me for now. I've done temps from 55-90 or so, but with all different oils so I don't know how much effect temperature has had. It will be a continuous flow gravity fed system once I've modified it. For now it just has 4 baskets that I'm playing with so I fill them, run it, empty them. I'm guessing they hold about 3/4 quart each. After more tests it does seem to be separating the water. The problem is that as it slows back down the fluid in each basket spins and somewhat remixes the water, but there are obvious striations in a circular pattern, so it is separating at least some water. Once I fab the right rotor for it it should make much higher G force than the units I've seen for sale for this purpose. I'm still trying to figure out if emulsified water will even do anything to the injection system. It seems the molecules should be so small it shouldn't have friction issues, but the pressures involved may create cavitation or some other thing that I don't know about.

 

You will need to heat the oil to efficiently break the emulsion. The hotter it gets, the easier it is to break. For VO, if you have water suspended in solid Crisco, its going nowhere fast. But if ya heat it up to a liquid, the water easily falls out. There are more physics involved in an oil/water emulsion, but the idea is the same. The warmer the oil is, the less time+gravity it takes to separate (I still can't remember what thats called...)

As for water in the fuel - baaaaaad things happen when those water molecules rapidly expand into steam in confined spaces, like the injector tips!!! Then there is the EROSION that occurs from the lack of lubrication where the fuel is compressed to over 20,000psi inside the injector and if you were to survive all of that, the corrosion of carbon steel parts in the injectors as well as the fuel rails, etc could eventually lead to problems. I always say 'Choose wisely or PLEASE post pics of the carnage!!'

I'm very interested in your CF set-up! Will you post some pics???

 

I was a machinist mate on real steamers in the Navy for a long time. What I learned about running our centrifuges, old delavalle purifiers, was that they don't work worth a squat on cold oil. OIl samples always came back unsat. So we rarely did that. In our lower levels where we had all that stuff we would usually run the purifier on the main sump directly for a minimum of 12 hours a day. That's usually oil running around 160-170 degrees. It worked great there. Then we'd clean the stacked discs at the 12 hour mark. It's amazing the crap that thing would get on hot oil. Also about every 15 -20 minutes we'd have to crack open the water drain on it on hot oil and gallons would run out. Gallons of water. On a cold storage tank, nothing.

Usually about once a month or so we'd pull all the oil from the main engine sump, about 3300 gallons on most of them, and put it in the settling tank and refill it from a clean storage tank. We never got rid of oil on navy ships. It was all recycled by us and reused.
When in the settling tank we'd heat it between 180 and 220 and run the purifier continuously till the samples came back good. And our oil labs onboard were picky. Especially on the carriers and battleship I was on. They also ran samples of our JP5 which was used by the boilers and the aircraft. In the settling tank at that temp with the purifier going we would have to clean it about every 2 hours. That's 3 times a watch and it got to be a pain but it worked great.
Just food for thought on making one get water out. They do work but the temp makes a big difference.

Oh and we never ran filters. We had a big double strainer on on side of the lube oil cooler but its holes were about 1/8 inch each. It was just big strainer baskets. If we had the main engine open or had made repairs after a wiped bearing then we would run muslin bags in the strainer baskets untill we were sure nothing was left in the oil but that is the only time we had filters. A properly working centrifuge is good enough. It made 3 of my ships run past the 50 year mark with millions and millions of miles on them. They would still be running now if they weren't museums.

 

RRranch, i think these centrifuges are just the spinning drums of the old sharples/delaval fuel/lube oil purifiers that we used. i remember that we had to run fuel or oil at minimum temp to control specific gravity for the dam ring that it exited thru, the heat helped remove water and contaminants from product. they would remove amazing amount of junk from "clean" fuel and oil. the dam ring was orifice disc that was on top of drum and if you had time to play you could change out to get varying viscocity of product. never had problem with fuel using them.

 

Dieselcraft centrifuge works great -My filter and dewater rig - Topic Powered by Social Strata

This is a link to the centrifuge thread at the biodiesel site. I started reading there and followed other links and ran searches. I would like to read your take on this if you are interested.

I was looking at a PS pump on a Mitsubishi the other day and thinking I could have this up and going in no time.

The last time I looked at veg oil my thought was to buy raw virgin oil from the processors for about a $1/gal. My guess is it would still need to be processed before it was reliable fuel.

 

Read this first: https://www.ford-trucks.com/forums/1...ml#post9566168

Post #7 "IMHO, the pressure-type CF's like Diesel Craft are not nearly as good for VO. The biggest problem is they have a tendency to increase the rate of polymerization."

 

Thanks for all the input guys, and I really liked RRanch's input as that is some dewatering experience. I'm currently replacing the water pump on my beloved oll burner. Hopefully while I'm working on it tomorrow I can put a few heated samples in the 'fuge to see what happens. I've yet to get or build a rotor for it, too many other things going on as usual. I'll try & shoot a photo too even tho it's not 'production ready'.