quote:

Originally posted by JOAT:
Just to throw in my 2 cents. With the exeption of my last rather nasty batch, I have not seen any free water in the centrifuge. All other runs showed no water in the rotor with non-mistwashed oil. So I don't think you can count on consistent centrifugal removal with all oil batches. Perhaps if the water is unusually dirty it aids the removal? However this same batch of mine also strongly resisisted evaporative removal.

Here is the main flaw in that, you have not been capturing the output at shutdown. Water will quickly run out the jets in the bottom at shutdown. Also to see it you would need to shutdown often, probably after each pass. If you run many passes it will get evaporatively removed and you won't see it in the output. And you would need to have a large enough amount to actually see, unless you run pan test just on the output. And you have been running horribly gunked up batches where the gunk could be filling the space where water could otherwise collect.

This leads to a new way to operate that I will be testing: Run at lower temps around 120F and collect the shut down output frequently, and pan test it compared to pan test of the before shut down output. This will probably remove water faster than the evaporative technique and if it works will mean we can eliminate the mistwash. We could use the puffing of steam which occurs for me at 120F as simply an indicator of whether it still has suspended water.

And I will wait before sending these samples for testing, since if this proves to work I want the #2 centrifuge only sample to be from the new "capture" method.

I am sort of kicking myself for not testing this sooner, I thought of this and even mentioned it on page 1! Too many things to test, not enough time, not enough bad VO.

I like that water in oil sensor...easy-1, is it? Maybe that will conclusively detect dissolved water in VO, as it says it does. That would be more accurate than a pan test, I think, unless the other contaminants in WVO mess up the indicator...

quote:

I don't see that a discussion of suspended water is anything but very central to this thread.

I agree...but discussing it separately and linking to the separate discussion makes the info much easier for others to find than if it is buried in a 12+ page discussion titled

quote:

Dieselcraft centrifuge works great -My filter and dewater rig

Organizing discussions so those that look for info later can easily find it is the reason we have most of the features included in discussion forums like this. Using them wisely is a small kindness we can easily provide with very little effort.

quote:

Originally posted by B.K. Hosken:
I like that water in oil sensor...easy-1, is it? Maybe that will conclusively detect dissolved water in VO, as it says it does. That would be more accurate than a pan test, I think, unless the other contaminants in WVO mess up the indicator...

Devices such as this ARE very accurrate and I suspect can measure suspended water even if it is highly contaminated with acids/sugars/salts...etc. Unfortunately they are also too expensive for the average individual to afford.

quote:

Originally posted by danalinscott:

quote:

I don't see that a discussion of suspended water is anything but very central to this thread.

I agree...but discussing it separately and linking to the separate discussion makes the info much easier for others to find than if it is buried in a 12+ page discussion titled

quote:

Dieselcraft centrifuge works great -My filter and dewater rig

Organizing discussions so those that look for info later can easily find it is the reason we have most of the features included in discussion forums like this. Using them wisely is a small kindness we can easily provide with very little effort.

quote:

Originally posted by B.K. Hosken:
I like that water in oil sensor...easy-1, is it? Maybe that will conclusively detect dissolved water in VO, as it says it does. That would be more accurate than a pan test, I think, unless the other contaminants in WVO mess up the indicator...

Devices such as this ARE very accurrate and I suspect can measure suspended water even if it is highly contaminated with acids/sugars/salts...etc. Unfortunately they are also too expensive for the average individual to afford.

I am going to look into the price of it again. I could use one of these for the biz anyway.

Guess what? The US distributor is 2 miles from my work place...Cool! List is 2.5K

quote:

Originally posted by SunWizard:

quote:

Originally posted by JOAT:
Just to throw in my 2 cents. With the exeption of my last rather nasty batch, I have not seen any free water in the centrifuge. All other runs showed no water in the rotor with non-mistwashed oil. So I don't think you can count on consistent centrifugal removal with all oil batches. Perhaps if the water is unusually dirty it aids the removal? However this same batch of mine also strongly resisisted evaporative removal.

Here is the main flaw in that, you have not been capturing the output at shutdown. Water will quickly run out the jets in the bottom at shutdown. Also to see it you would need to shutdown often, probably after each pass. If you run many passes it will get evaporatively removed and you won't see it in the output. And you would need to have a large enough amount to actually see, unless you run pan test just on the output. And you have been running horribly gunked up batches where the gunk could be filling the space where water could otherwise collect.

Actually, I pull the CF as soon as it stops spinning and break it open in a separate pan, still draining.

Any water in the upper chamber above the holes in the separator plate would not drain out till you open it up.

I usually pull the CF down every half hour or so till solids slow down, 1/8 to 1/4 inch sediment in most cases.

Not all my batches have been horribly gunky, I just post those pics for dramatic effect

EDIT: even tho I saw no free water previously, there was steam coming out of the hot sediment, which indicates some trapped moisture at least....

JOAT, you need to capture from the moment the pump stops. Because in my tests the water is thin enough to start draining right away, and so does some particles of the black gunk in the rotor.

I just ran the first test of my new capture method, on my bad #1 oil from above that I had saved, which had lots of suspended water according to a pan test. And the filtered VO passed the pan test with no bubbles after 1 pass at only 120F temps. And the stuff I captured from the rotor is horrible on the pan test, very crackly and dangerous, and I see some small globules of free water in the bottom even.

Its looking very promising, I suggest you test this same way on a large batch, I don't have any large batches around right now. I went and ran everything I (and my restaurant sources) have through the rig already since it was so fun

quote:

Originally posted by SunWizard:
JOAT, you need to capture from the moment the pump stops. Because in my tests the water is thin enough to start draining right away, and so does some particles of the black gunk in the rotor.

Only from the lower chamber tho. The bottom half of the upper chamber is prevented from draining.

I'll have another batch running mid week...

quote:

Originally posted by JOAT:

quote:

Originally posted by SunWizard:
JOAT, you need to capture from the moment the pump stops. Because in my tests the water is thin enough to start draining right away, and so does some particles of the black gunk in the rotor.

Only from the lower chamber tho. The bottom half of the upper chamber is prevented from draining.

I'll have another batch running mid week...

Good, what will that batch be?

From a visual estimate of how much I get when I capture, I think its more like 1/4 is prevented from draining, so its quite a bit that drains. Remember you got to subtract what is below the seperator, it drains. And of what is above the separator, its about 1" up to the bottom of the holes. Try full capture you will see. And when it stops, the water and sludge inside all the top 3/4 recombines sloppily and violently since on mine it hits resonance at certain speeds as it slows, so lots can go out the jets.

I hope this isn't too horribly far off topic, but since Mr. Dieselcraft is checking up on this thread, and others here have ties the business and know how this thing is constructed, would it be possible to make one that is more of a "true" centrifuge design? By this I mean still using the jets to spin the rotor (no seals, simple, inexpensive, etc etc), but maybe having it spin the heavy junk (water, goo, probably some oil as well) out a different hole off the side of the unit... much like the industrial centrifugal seperators? Wouldn't this be achieved by modifying the housing? Just thought I'd throw that out there because it was mentioned a short time ago that there was an initiative to do some R&D for an SVO-specific unit.

quote:

Originally posted by SunWizard:
snip...
And when it stops, the water and sludge inside all the top 3/4 recombines sloppily and violently since on mine it hits resonace at certain speeds as it slows, so lots can go out the jets.

Hmmmmm, that has me thinking. One other change just prior to noticing water in the housing was I tweeked the rubber mounts till the unit had minimal vibration. I wonder if vibration is a factor in preventing water separation. Will have to keep testing of course, just seems like the list of variables keeps growing

Speaking of more tests, talked to Mr Dieselcraft todday and sounds like over a dozen WVO'ers will be getting their units next week. Be interesting to see the results other users get.

Also neglected to mention, usually with such a cruddy unwashed batch my layered paper towel suction filter would have to be changed every 5-10 gallons, even with 3 stages of settling. Made it thru the whole 45 gallons it reaches with no restriction. My house water filter (5-15 micron spun) and Fleetgaurd FS1000 on the pressure side of the filter pump are still showing no restriction after about 150 gallons. Usually there is some pressure increase by now. Certainly a good indiator the Dieselcraft is pulling a lot more of the 5-20 micron stuff out than 3 stages of 1 day heated settling was getting.

Hey SunWizard and Joat or anyone else who has a Dieselcraft. Can you descrobe the path of the WVO through the unit based on the Dieselcraft diagram:

http://www.dieselcraft.com/productinfo_OC_1.html

Describe where it enters, the path it takes from there and how it sets what part(s) spinning, where the solid collects, where the oil exits etc.

Any idea what is the dwell time of the oil in the centrifuge before it exits? I.e. approx. how many revolutions does the oil stay in there before it exits?

Could you also spin a batch of your conventionally filtered oil that's ready to burn and see how much (or how little) crap is in it? It would be an interesting test.

The oil enters through the shaft in the center, comes out of 2 holes in it where it lubes the bearing and enters the rotor through 2 holes about 3/4 the way up. The rotor fills up and spins 6000rpm, and most of the solids collect on the walls of the rotor cover. It then goes through the 4 holes on the seperator plate into the lower chamber, where some more seperates on the outer edges of that chamber before it exits the jets at the bottom.

The dwell time at 1gpm is about 6 seconds = 600 revs. if you just divide the capacity by flowrate, but the heavier goo and water can dwell there for hours. So its hard to tell if dwell time means much.

where did you guys get your pumps for your centrifuges? the guys through the company aren't that helpful, they just tell me to look on the website and there is nothing on there about where to get one (i get the point that I have to get one producing 90 psi).

We are useing a Power stearing pump belt driven from a old motor we had laying around.

quote:

Originally posted by liveoily:
where did you guys get your pumps for your centrifuges? the guys through the company aren't that helpful, they just tell me to look on the website and there is nothing on there about where to get one (i get the point that I have to get one producing 90 psi).

Please read my initial post on page 1 it shows the pump and motor and explains.

We are woriking ferverishly on getting pump and motor data, part numbers and an order system in place. By Friday we will post all the needed data so you can get the needed pump with one phone call or click of the mouse. :-)

I unfortunately haven't been able to read through all the pages of this thread, but it seems to me there's a bit of a catch-22 here:

1) If there's any water present in the oil, an emulsion would be formed (we've had that happen with solid/liquid centrifuges, and it was a BIG problem)

2) Water removal is easiest once the oil has been adequately filtered (and you can simply keep it heated and the water will fall out).

It seems to me that water-free oil with a lot of solids in it is a pretty rare event. Maybe I'm wrong. Perhaps WVO-ers can cherry-pick off the top in a way my company can't.

Anyway, my 2 cents.

My own suggestion to the firm making this product is to consider a method of separating the "drive power" from the "filtered oil". The reason I mention this, is that it has been noted that one pass (or even 2) was not enough to remove all of the "centrifuge extractable" particles. This is "likely" an indication that the residence time of the oil in the unit is too short.

There is also the minor point that running the oil through the pump might have the unwanted effect of mixing up the particles, water, and oil really well.

quote:

Originally posted by HarryN:
My own suggestion to the firm making this product is to consider a method of separating the "drive power" from the "filtered oil". The reason I mention this, is that it has been noted that one pass (or even 2) was not enough to remove all of the "centrifuge extractable" particles. This is "likely" an indication that the residence time of the oil in the unit is too short.

There is also the minor point that running the oil through the pump might have the unwanted effect of mixing up the particles, water, and oil really well.

The number of passes would depend on what micron level you want to get to. They indicate it will remove down to 0.1 micron, and I haven't seen anyone filtering that small before this unit. So getting every "centrifuge extractable" particle may not be what most want. Doesn't hurt to do it if its easy, thats why I have been testing to the limits of many (4) passes. Future tests may give us an idea of the micron level removed with 1 pass, 2 pass, etc.

Just running through any pump mixes all that up really well, so you can't avoid that even if you seperate the drive power. Then does this rig remove it all really well is what we are testing.

quote:

Originally posted by kumar:
I unfortunately haven't been able to read through all the pages of this thread, but it seems to me there's a bit of a catch-22 here:

1) If there's any water present in the oil, an emulsion would be formed (we've had that happen with solid/liquid centrifuges, and it was a BIG problem)

2) Water removal is easiest once the oil has been adequately filtered (and you can simply keep it heated and the water will fall out).

It seems to me that water-free oil with a lot of solids in it is a pretty rare event. Maybe I'm wrong. Perhaps WVO-ers can cherry-pick off the top in a way my company can't.

Anyway, my 2 cents.

This is removing bad emulsions very quickly in my tests. Maybe there are some that it won't but I haven't found one yet. No cherry picking needed, only a 100 micron screen to prevent plugging the jets.