quote:

If anyone has a lab or even some of the testing equipment, that would be the best! Also ideas of how to structure such lab testing would be useful since thats the stage we are at now.

So to continue the testing with lab tests, here is a possible classification of VO types. It would be interesting to see how much of each class of VO people are running into.

Type 1:
No suspended water in a pan test, no mistwash needed, cleans up great with this rig. If you can find this kind of oil, you've got it easy. My main source is this kind of oil. If you can find type 1 oil, using a mobile or any version of this rig is easy.

LAB:
Test starting oil, then test what micron particulate removal we get at pass#1, #2, etc. (edit- and the same oil through a 10 micron clear goldenrod with new filter to compare)

Type 2:
Has suspended water in a pan test.

LAB:
Test starting oil, then a sample each after mistwash only, mistwash and centrifuge, centrifuge only. Test for suspended water, PH, and particulate removal. This would also help answer the still raging debate, does a mistwash work?

These tests then could lead to 3 more types which would require more testing to determine, and I question how the line could be drawn (so safest would always be mistwash) yet here they are:
Type 3: small amounts show in a pan test, requires no mistwash and we get full removal?

Type 4: larger amounts show in a pan test, or a PH test says warning: acid, yet with a mistwash do we get full removal?

Type 5: nasty VO, pan test is bad, PH is bad, can't clean up even with a mistwash. Learn how to avoid this VO.

Possibly the line could be drawn between types 3-5 by doing a PH test on a small sample first. Since if there are acids to start, its not likely to work without a mistwash, but with no acids it may be possible to remove fully after no mistwash.

Being a skeptic, I think the line between types 3 and 4 may be so hard to determine that its best to mistwash whenever you fail an initial pan test. The line between 4 and 5 would be a good one to define if possible, if there is a type 5 even. (edit- and I probably will not have the funding or desire for testing or defining between types 3-5 since its such a grey area, too many possible VO sources to try to make any claims anyway. So the expected result will be the same recommendation said here so much already, if you fail an initial pan test, do a mistwash, or settle some more, otherwise you are in great shape.)

Anyone have ideas of tests to possibly define those lines?

Some good news to report: Dieselcraft is going to develop a complete rig like this for WVO. We created enough interest with this post and the sales it generated to get this manufacturer who seems very capable in engineering to develop for WVO. No details on timing or price, but it probably will be something 110volt, with heater, gear pump, and this centrifuge. They had already done some R&D on this type of rig for motor oil and the interest we created has made them sure to consider WVO in these future products, and speed up the development.

And they will fund some initial lab testing on this which we are working on. And there are other offers of support for lab testing too.

quote:

Originally posted by SunWizard:
Some good news to report: Dieselcraft is going to develop a complete rig like this for WVO....probably will be something 110volt...

Please encourage them to make it NOT a full component, pump and all thing. ONE, lets be real, WVO people have a side to them that likes cost effective ESP. if that means they can adapt their own pump they already own etc. Having to buy something that includes parts you already have will reduce their sales. TWO, having to buy something then undo it (replace pump w/12v one)for mobile use is not that good either.

Also have you passed onto them that one of the aspects of why this particular unit has more value to us than some other types is their little jet's that spin the rotor. For us these jets are (possibly) also acting like flash evaporators when used with heated oil. This is a KEY feature to us unless they provide proof that it spins fast enough to remove all water by centrifical force alone. Granted it is not yet proven it removes all water with the Jet/flash evaporator built into it.

Well they read this thread sometimes, so you probably just encouraged them They will keep offering these centrifuges seperately so we who like to home build our mad scientist creations out of parts in the garage can still do that. The new rig would use the same centrifuge so the jets and rotor are unchanged.

Long term if the WVO market looks good, I think they are responsive and may be interested in learning modifications that would possibly remove water directly by a port on the centrifuge, etc. They always ask me questions about how can they make it better. And they have the manufacturing connections worldwide to do it. We might want to get Bud hooked up with them.

Till resturants start selling their waste oil to veg. fuel corporations, the holy grail will remain onboard filtering. If they want to experement with that I'll offer myself up as a guiny pig since my rig is being built to be a cross country traveler. Some type of On board filtering for this will be REQUIRED. Besides a Unimog going down the road (especially mine) will definately catch peoples attention thus be excellent PR

quote:

Originally posted by jeepin, moggin Jessup (coachgeo):
Till resturants start selling their waste oil to veg. fuel corporations

Yikes I hope not, don't give them any ideas

quote:

Besides a Unimog going down the road (especially mine) will definately catch peoples attention thus be excellent PR

Agreed. I would like to see it, have you got pics of it up here someplace?

quote:

Originally posted by SunWizard:

quote:

Originally posted by jeepin, moggin Jessup (coachgeo):
Till resturants start selling their waste oil to veg. fuel corporations

Yikes I hope not, don't give them any ideas

Well that's a good thing bad thing. If this is happening that means Veg. oil fuel is hitting maintstream which means less dependency on forign oil.[/quote]

quote:

Besides a Unimog going down the road (especially mine) will definately catch peoples attention thus be excellent PR

quote:

Agreed. I would like to see it, have you got pics of it up here someplace?

hmmm. I think it is on this board. Posted before attachments function was disabled. Other places it is posted you have to be a member to view it? I fought against it (Im the moderator for a Unimog forum) but the powers to be wanted only members to be able to view pics.

I contacted Dieselcraft and they agreed to funding 3 samples to be tested for 2 things "Automatic particle count" which gives the ranges of micron sizes of the particles, and "water by karl fischer" which gives ppm water. So I was thinking 1. Original VO (some with suspended acid), 2. Centrifuge only 2 passes, 3. Mistwash & centrifuge 4 passes.

More tests beyond this would be useful but it probably makes sense to see the results of these first, to decide what else we want to test.

I plan to save about a gallon of the original VO incase we need more tests on it after this. (for example using my old rig of heat,settle 24 hrs, filter through 10 micron goldenrod.) Also save some just after mistwash and do a litmus PH test on the VO at each stage and the wash water.

Anyone have other ideas of how to design to get the most out of these first tests?

I have a design suggestion that may appeal, particularly for this planning to build a mobile processing rig :

Have you considered enclosing the tank and return outlet in a simple pressure vessel (a recycled LPG tank for example) and putting the system under vacuum to assist in evaporation?

Steam tables ( http://www.engineeringtoolbox.com/saturated-steam-properties-d_457.html ) give a boiling point of 81 deg C (178F) at 0.5 bar (half atmospheric) and 60 deg C (140F) at 0.2 bar.

Since coolant temps are around this range, a nice flat plate heat exchanger would do the trick to get the temp, and insulating the tank
would help maintain it.

Belt driven automotive vacuum pumps can pull vacuum down to 0.2 bar ( http://www.wabco.info/intl/en/inform.cms.php?keywords=820+001+305+3 ) and could be run from the same drive as the pump.

You'd need to plumb an air circulation system in there to take the steam to an condensation chamber and return "dry" air to the tank.

Anyone else have thoughts on this?

cheers all from downunder,

Richard

Yes I have thought about that, and it probably could be made to work. It would add a good deal of complexity to the setup. I was originally thinking I might need to vacuum dewater before running the centrifuge. Then when I tested and saw how good it was dewatering at much lower temps than I expected, I think the extra complexity would not be worth it. Maybe in a high humidity climate it would be more helpful, since I have very low humidity here which I am sure helps the evaporator work.

Yep, a little more complex, but pretty useful for a mobile rig that has access to coolant heat, but not the electrical power needed to run big wattage elements for extended periods.

I'm still getting my vehicle's WVO system going (IP issues) but since it's a truck/mobile home, I've been putting a lot of thought into how best to process on the road.

I'm keen to experiment with vacuum dewatering on the move once I have the system up, so I'll let you know how I go!

Thanks to everyone for all your hard work so far - hopefully I can contribute my experiences too...

cheers,

Rich

After remounting the unit with 2 piece bushings I have reduced sound and vibration to the point you can barely hear it in the next room.

Ran a pretty wierd batch today and got some interesting results. First, the batch started as a dirty thick emulsion. Turns out the bar had a new cook, used about as much water as oil cleaning the fryer...

Anyway did a 160*F heat and overnight settle, then ran it thru the centrifuge after draining off the water.

The oil still had a light emulsion, kind of cloudy, and a lot of suspended yack. First run about half filled the rotor, but I was interested to find free water caught in there too, about 2 tablespoons. There didn't seem to be free water in the lines or drum bottom, so appears it separated it from the oil. You can see some of the droplets in the pics. Saw this at 60 psi and at 80. Didn't think till too late to put a jar under the outlet and see if I could catch any exiting water. Curious if any makes it through as drops or as emulsion.




Also added an oil diverter mod to send the incoming oil up instead of directly outwards. The idea being to perhaps reduce turbulence and the amount of solids getting thru. Not enough runs yet to tell conclusively but it did seem to layer in the bowl smoother with less in the jet chamber. Basically I just put a couple o-rings on the shaft, then a plastic plug with a hole over it. The oil now moves up inside the plug to the top before entering the main chamber.

Very interesting results. That is some nasty oil! I am surprised about the water you found, did you take the rotor off quickly after it quit spinning? I have noticed it does take a few minutes for the oil to drain slowly out the jets, but it seems if there was water it would come out quickly after it stops spinning. Which is why I have captured the output as soon as I turn it off to see if I catch any water, but have not found any. You had much nastier oil so maybe thats how you caught some, and I bet you would have caught more if you captured everything right after you shut off the pump.

Thats an interesting and simple mod, too bad we can't see whats going on inside the rotor. I wonder if the ribs on the top of the rotor are making turbulence? Or maybe they help get the newly entering oil up to 6000rpm speed quicker. Because at 1gpm the oil doesn't stay in the rotor long.

Here is a mod that I think would be fun to test, change the jets to a smaller size, lowering the flowrate and increasing the time in the centrifuge. Which would also increase the temps which should help the FE action. And it would be easy since they are standard brass jets. Have you measured the existing jets holes yet?

quote:

Originally posted by SunWizard:
Very interesting results. That is some nasty oil! I am surprised about the water you found, did you take the rotor off quickly after it quit spinning? I have noticed it does take a few minutes for the oil to drain slowly out the jets, but it seems if there was water it would come out quickly after it stops spinning. Which is why I have captured the output as soon as I turn it off to see if I catch any water, but have not found any. You had much nastier oil so maybe thats how you caught some, and I bet you would have caught more if you captured everything right after you shut off the pump.

Thats an interesting and simple mod, too bad we can't see whats going on inside the rotor. I wonder if the ribs on the top of the rotor are making turbulence? Or maybe they help get the newly entering oil up to 6000rpm speed quicker. Because at 1gpm the oil doesn't stay in the rotor long.

Here is a mod that I think would be fun to test, change the jets to a smaller size, lowering the flowrate and increasing the time in the centrifuge. Which would also increase the temps which should help the FE action. And it would be easy since they are standard brass jets. Have you measured the existing jets holes yet?

Jets are approx .040"
I'm planning to try smaller ones at some point
If memory serves they are M8, 1.0 pitch

Forgot to mention earlier that it took almost 5 hours to clear the oil with this batch. Even at 6.5 hours it still was marginal per the pan test. Seems to indicate the difficulties a portable setup would see with poor oil, as I did not mist wash this one. Also had baking soda in it, which seems to neutralize some acids but delays dewatering considerably.

I didn't try to catch draining water post spin down, as it is heavier so stays in the rotor below the separator plate holes.

quote:

Originally posted by JOAT:
...Also added an oil diverter mod to send the incoming oil...

Could you add to this post a "before mod" shot and provide a little more explanation. Not understanding your thoughts

quote:

Originally posted by JOAT:
Forgot to mention earlier that it took almost 5 hours to clear the oil with this batch. Even at 6.5 hours it still was marginal per the pan test. Seems to indicate the difficulties a portable setup would see with poor oil, as I did not mist wash this one. Also had baking soda in it, which seems to neutralize some acids but delays dewatering considerably.

You didn't say how big of a batch did you run? A portable setup probably wouldn't be doing baking soda, especially if a mistwash dewaters much quicker, so its hard to compare that batch.

George, you can see some before mod shots on page 1 and 6. There are holes in the center post behind the red part he added, where the oil enters the rotor, and JOAT is directing that flow upwards.

can you guys explain your setup a little more please? Is a pump necessary, what about elevtaing a drum and letting gravity feed into the centrifuge?

Where did you get your pumps? The maker says it requires a pump generating optimally 90 psi, what did you use and where did you get it?

Can I see more pics too please?

quote:

Originally posted by SunWizard:
You didn't say how big of a batch did you run? A portable setup probably wouldn't be doing baking soda, especially if a mistwash dewaters much quicker, so its hard to compare that batch.

My bad. 55 gallon batch, at 170-200 degrees. Mistwashing being a bit inconvenient for portable use, I've been trying different methods. Previously gave up on baking soda, but tried it again with the CF to see if any benefit.

quote:

George, you can see some before mod shots on page 1 and 6. There are holes in the center post behind the red part he added, where the oil enters the rotor, and JOAT is directing that flow upwards.

Hey Coach, what he said. To elaborate, instead of the oil going straight out of the center shaft, it now is directed upwards by the red plug in the pic. Thus it may reduce turbulence, as well as moves it's entry farther from the exit holes visible in the pic below the red plug.

quote:

Originally posted by liveoily:
can you guys explain your setup a little more please? Is a pump necessary, what about elevtaing a drum and letting gravity feed into the centrifuge?

Can I see more pics too please?

Lots of pics on page 1 and 6

Page one has a lot of details. Gravity won't work as you need 35-90 PSI, with 60-90 working best in my tests

quote:

Originally posted by liveoily:
can you guys explain your setup a little more please? Is a pump necessary, what about elevtaing a drum and letting gravity feed into the centrifuge?

Where did you get your pumps? The maker says it requires a pump generating optimally 90 psi, what did you use and where did you get it?

Can I see more pics too please?

Go back to page 1, there are lots of pics there and also pages 6 and 7. Full explanation on page 1. Yes you need a pump that can do 0.9gpm at 60-90 psi. And you need to heat to 120-200 F.