Hi plugly

Many thanks for giving it a try and posting.

Can you say if there was more deposit around one terminal or the other?

I've increased the brightness of your pic, because my screen is a little dim and I couldn't see the soap clearly - here's the tweaked pic...



Could you try again, with only one central electrode and the other side connected to the tank.

Try + on middle electrode, - on container and vice versa, to see the effect.

Have you tried it with glycerol-laden bio yet? Make sure your probes don't go down so far as to end up in the settled layer.

electrostatic air filters should have a nice high voltage DC supply. I'n not sure about the air ioniozers but they also might.

-dkenny

quote:

My test setup consists of a glass jar, into which I place 2 parallel metal plates, each 100mm x 30mm, spaced 30mm apart.

Very cool work, GL! Might I suggest trying this with the plates oriented horizontally, positive on bottom and negative on top? This should speed up separation as well as produce a much sharper gradient.

Also, another ideal source of HV dc power supplies is copy machines. These are connected to the corona wire and are typically rated anywhere from 3-10kV but at very low current, a good thing here. They are easy to identify even for the non-expert in electronics - just look for the thing inside of a copier that is potted in epoxy (usually red-orange).

Hi tesseract

You're right - horizontal plates work very well, the bottom one can sit right on the bottom, the top on on the liquid surface.

I just tried it in the glass jar and it separates nicely.

May not be so effective in a steel drum, I think you'd need non conductive sides.

I'm trying the process now with virgin sunflower oil as the source, all my tests so far have been with WVO.

Thanks!

Hi Graham. Nice experiment. Have you considered ultrasonic seperation? Solid state piezo-electric transducers may prove equally effective, with less current draw and a higher margin of safety. Might be worth a shot. See www.sonosep.com.

Hi welder,

Thanks for the link. I'd not heard of ultrasonic separation before - looks interesting.

quote:

Originally posted by GrahamLaming:
...
You're right - horizontal plates work very well, the bottom one can sit right on the bottom, the top on on the liquid surface.
...
May not be so effective in a steel drum, I think you'd need non conductive sides.
...

Excellent! Every so often I get lucky with a suggestion...

I wouldn't expect this to work too well in a metal drum, but since no heat (or not much) is required, there's no reason to not use a plastic drum. Metal foil can be "contact cemented" on to the outside, but because the current required is so low, I would think that metal dispersion paint (eg: cold galvanizing, copper flake, nickel flake, etc.) could be used as the bottom conductor. To make a connection to the coat, embed a bare #18 copper wire around the circumference of the drum/cone bottom tank in a bit of epoxy.

Current flow should be extremely low - my seat of the pants estimate says that if more than 10mA flows with both electrodes submerged then there are ionic contaminants in the biodiesel (NaOH or KOH; fatty acids;???).

Definitely an excellent "thinking outside the box" sort of experiment!

So, why not make the tank itself to be one electrode and the other can drop in from the top. Add 30KV and clean up the oil.

quote:

Originally posted by Tony from West Oz (The Wizard of Oz):
So, why not make the tank itself to be one electrode and the other can drop in from the top. Add 30KV and clean up the oil.

You could do this, but it probably won't work nearly as well. Grtavity assists in the separation when the charge is applied to the top and bottom of a plastic drum; if the whole drum is one electrode then gravity is working tangential to the electrostatic field.

30kV will definitely give you a jolt, no matter how limited the current is. Static electricity will give you a good zapping at 10kV, and while it isn't lethal, it certainly isn't pleasant. Another thing to watch out for is the whole setup acting as a capacitor - the biodiesel and/or the tank will act as the dielectric and store a charge.

GL - as a fellow electronics guy, try testing your setup with an LCR bridge to see what the capacitance between the two electrodes is. If it's more than a few tens of pF it could give you a real "surprise" after power has been removed...

quote:

Originally posted by tesseract:
GL - as a fellow electronics guy, try testing your setup with an LCR bridge to see what the capacitance between the two electrodes is. If it's more than a few tens of pF it could give you a real "surprise" after power has been removed...

Oh man oh man... am I the only one here who now has images of a biodiesel hybrid that uses the fuel tank as a capacitor for acceleration and regen?

i had a chat to a mate that does chemical mixing for the elector plating industry told him about the tests i`d been doing he said the soap was hitting the + copper side and the hydroxide was becoming copper-hydroxide he said most metals could be used for the probes and copper was proberly the best and the more power used would only effect the time it takes to clear but there would be a cut off point where it would all balance out.
this does seem to match what ive seen even with low voltage the bio clears in about 3hrs and the copper + probe is clean like new piping.

plugly

60vdc, even with the metal plates spaced 2 mm apart, showed no electrostatic filtering effect.

However, the de-meth'd & de-watered (from a 10% water pre-wash) had very little soap in it anyway.

I am using mild steel plates. Graham - have you used steel, or only copper?

Hi Dan

My 1st test was with copper.

Then I tried it in a steel bean can with the can as one electrode, a carbon rod as the other.

I also tried aluminium foil. They all seem OK.

However, I'd expect steel to be a good choice.

I have had some effect below 100v , but not much. It is certainly painfully slow.

For glycerol separation, I found 300V with 30mm spacing worked well.

For soap separation, I found 1000V plus is effective. You can see the soap moving in the bio as you apply the field.

You don't want your plates too close together, because you don't want glycerol or soap to bridge the gap. Their conductivity is high, comp[ared to the biodiesel, and they will tend to either heat up and scorch if you have a fair amount of current available, or cause a large voltage drop, if your voltage source is current-limited, as it should be, to around 1mA.


Hi plugly,

I'm not too sure about the electrolytic action you mention, when applying the field to de-meth'd biodiesel. If water has also been driven off, you should see no electrolysis, simply a bulk particle charge transfer, through the insulating medium of the biodiesel.

Now, you may well see normal conduction in the raw biodiesel when doing the experiment to liberate glycerol, bacause there is water, NaOH etc. present during that time.

So, it may be good to consider which materials would be preferable for the electrodes, as we don't want to introduce any hard-to-remove metal salts.

I'd think steel would be fine, as would carbon.
Any thoughts?

How will you remove the soap once it has accumulated to one area? Sorry if this is a stupid question.

I am envisioning a plate that has a surface which can collect soap particles. You could then disconnect the voltage and pull this electrode out - carrying the bulk of the soaps with you.

here's a link to the drawing:
http://www.biodieselpictures.com/viewtopic.php?p=242#242

Has anyone tried useing this method to dewater WVO ? The oil industry uses it to filter out water and salt so I was thinking maybe we could adapt it to WVO.

Hi CaptnX

Thanks for the suggestion.

I'll give that a try in the morning - I have some wet oil needing de-watering, I'll see how it perfoms.

quote:

Originally posted by trik396:
How will you remove the soap once it has accumulated to one area? Sorry if this is a stupid question.

This technique may be well suited for a countinuous processing rig. If the oil is pumped slowly through a long horizontal pipe with the plates position so that the soap and glycerol falls to the bottom you may be able to achieve seperation by the end of the pipe. The soap/glycerol can be continuously siphoned off and the oil can proceed to the next step.

I'm just spit-balling here but the rapidity of seperation makes me think it is possible. A mechanical engineer can help out with the design. And any petroleum engineers would be invaluable since they have probably already dealt with these problems while we reinvent the wheel.

Just a thought,
Terrapin

I was thinking more of a plate inside a tube. The top of the tube is above the barrel and has a spout leading to the soap bucket/barrel. On either side of the plate is an insulating rubber squeege. Every X amount of time you lift the squeege, the soap is forced above the liquid level and out into the soap bucket.

-Jim

Is it safe to say that pure biodiesel is non-conductive?