Hi George,

I don't think many people have tried this yet - it's outside the usual realm of plumbing and mixing, so I'm not surprised folks are reluctant to respond if they haven't first hand accounts to write about.

I'd be happy to make a few safe, high voltage generators for folks to share among themselves to try this out on small, safe sample batches.

As to KOH, I'm one of those with no experience - I honestly don't know how this works with KOH.
And I don't have the drive to try KOH, NaOH just works so well for me.

Whether KOH or NaOH, you won't need to remove the methanol for the glycerol separation.

But, you do need to remove the methanol for soap separation, else it stays in solution in the methanol.

From what I hear, you don't want methanol in your fuel, I don't really know why, it's just
what I hear.

Maybe because it inhibits soap separation, or presence of methanol in biodiesel is coincident with presence of soap, maybe because methanol and aluminium are co-reactive, if you have an aluminium head. Maybe someone with quality factual knowledge can explain this one to us.

As to the benefits, I'm blown away by the speed of glycerol settling with the HV. It is just so slick and so complete. I'm willing to bet I can get complete glycerol dropout in an 80 litre batch within 10 minutes, before the year is out. My power supply needs matching to my reactor. As it all happens below the liquid line, I'm not overwhelmed by the challenge of making it safe - it is 'do-able' if you keep all spark opportunities out of the flammable vapour zone.

So, you react, throw the switch, have a nice cup of tea (it's an English thing), and then go straight to distillation. Two hours later you have fuel fit for the tank. No water needed, 5% prewash obsolete already with this method. That pleases me, because it is one less step, easier to distil more valuable methanol out. Less methanol in the environment.

The glycerol separation is easy, fast and effective. Proven, done, I'm happy with that.

Need more folks to join in so we can get the water and soap separation side developed more.

I'm on a limited time budget here. I have a day job, 2 young kids, a 100+ mile a week cycling addiction and houshold chores to attend to, plus the biodiesel addiction. 24 hours is too short for a day!

But here's today's recommendation:- If you want to get creative - get a bicycle - cycling just frees the mind - it all happens for me when I'm peddaling like a maniac, doesn't matter, forest or highway, sunny, rainy, hot, cold - all good - endorphins or somesuch. Good stuff - homebrew class A drug which is 100% legal!

Each day is too short, but plenty good.

Come on folks, join in the fun!

Graham,

Thanks for the post. I'm anxiously awaiting my DC power supply so I can start experiments. Your results are very exciting!

I didn't think this worked for water separation? Have you seen positive results for water separation? Have you tried it for settling WVO? It might be similar to the glycerin in that it electrically saturates the sediment causing it to settle out quickly. Ahhhh...I need that power supply!!!

This could easily become the standard for producing Biodiesel!

Kind Regards,
Martin

This is great work once again Graham --

The following link has cheap parts and explanations of how to assemble a high voltage, current limited device. Has a picture and where to get parts from.

7.5 KV current limited Graham, what is your opinion of this device. Hard to go wrong for $10.95 as long as this is safe.

Chris

Hi

There has been success in the oil industry with electrostatic separation of water from oil.

Take a look at http://www.ecs.soton.ac.uk/research/projects/CEC/

http://www.akerkvaerner.com/NR/rdonlyres/62F79C75-EC69-...water%20separator%22

http://www.natcogroup.com/Content.asp?t=ProductPage&ProductID=7

I haven't had much success with voltages up to 2kV.

But glycerol separation is wild!

I haven't tried more than 2kV, so looking forward to your results.

I'd suggest a spacing between electrodes of 30mm per kV as a start for glycerol.

You tell us what you find with 7.5kV and water laden WVO.

I can't comment on how it would work, as there is little tech spec on the web page, and it's above any voltage I've used in these experiments.

Cheers!

Graham,

Great articles! I'm sure there are some tweaks, but it appears the oil industry is doing a very similar thing...or vice versa.

The last article indicates a voltage range of 14,000 to 22,000 volts with AC current. I assume this is why you are building a high voltage power supply? I'll be curious if we can get water to coalesce with lower voltages or if that's the threshold for water.

Also, any idea why they would use AC vs DC? I would think a constant voltage would be more effective than alternating.

Am I missing something here? It seems like high voltage current is too good to be true…rapid water and glycerin settling and soap removal at the molecular level to produce high quality bioD faster and cheaper.

Kind Regards,
Martin

quote:

Originally posted by GrahamLaming:
(SNIP)
So, you react, throw the switch, have a nice cup of tea (it's an English thing), and then go straight to distillation. Two hours later you have fuel fit for the tank. No water needed, 5% prewash obsolete already with this method. That pleases me, because it is one less step, easier to distil more valuable methanol out. Less methanol in the environment.(SNIP)

Graham, thank you for getting back to me on this. I am still puzzled by one part of your process as outlined above: you say that you use the HV to settle out the glycerine, then "straight to distillation." What I'm not understanding is this: are you then distilling out methanol at that latter stage? What about the soap? Are you using your waterless spray technique to remove soap, or does it somehow come out with the glycerine when you apply the HV?

Again, many thanks and apologies in advance for my thick-headedness.

Hi George,

After draining the glycerol, I re-heat the biodiesel and distil off the methanol, whilst circulating the batch - all in the reactor.

After distilling off the methanol, this biodiesel has a suspension of easy-to-filter soap particles, fragments, blobs etc.

I pump this from the reactor into the filtering tank, via a 200 micron filter and recirculate in the filter tank until no more soap is recovered in the filter bag.

Then I repeat with 10 micron.

Then I repeat with 5 micron.

By now the fuel should be crystal clear. I do a wash test on a small sample to ensure there's nothing water-soluble remaining. The water should be CRYSTAL clear, even after a violently manic shakeup, though it will need to settle for 1/2 hour or so after that.

Because I get no clouding of the water, I surmise the fuel is clean. The pH of the settled water is consistently the same as that of the water I add, so I am comfortable I've not let any NaOH through. This is consistently repeatable, a reliable process.

-------------------------------

NOW, WHAT's NEXT?

1.
Water separation from WVO should be possible, but I haven't managed it yet.

2.
Soap separation is only feasible AFTER methanol removed. The electrostatic field does nothing to the methanol concentration, so is of no use before methanol is removed. I still need to do more work on the electrostatic effect on soap in methanol-free biodiesel in production-sized batches. It looks good in jam-jars!

3.
There MAY be a good saving in time for those who water wash - the orange juice effect right after washing clears rapidly with an electric field, so this may speed drying, but I need to do more work on that too. Seems promising from 1st quick tests.

4.
There MAY also be another approach to washing, like this...

React
Electrostatically settle glycerol and drain.
Add 1% (?) water to BD and mix well.
Electrostatically separate.
The bulk of the methanol/soap/water should be heavier than the BD and clump within the field and drop out.
The remaining BD should have much reduced soap and methanol levels, to make the following stages need less energy. But I haven't tried this idea yet.

Graham,

please forgive my ignorance - possibly caused by short attention span syndrom!

But, you say that by using an electrostatic glycerine separation step that it makes the idea of a water pre-wash (insert percentage here) step obsolete. Why is this? Does the electrostatic treatment cause the drop out of similar/greater levels of soap into the glycerine phase?

If so, do you have some numbers of soap ppm remaining in the biodiesel after this treatment vs. soap levels in the biodiesel after a 5% "conventional" pre-wash?

Best regards,

Wes

Forgive me for answering on Grahams behalf...

Dogma - I believe that the primary reason for the prewash is to help the glycerine to settle out faster. With the electrostatic method, this is no longer a requirement as the glyc settles much faster using the current.

Tripitaka,

I think my understanding of what the pre-wash is doing is different.

My understanding is that a 5% pre-wash allows a big head start on methanol/soap removal from the biodiesel while mixing in the presence of glycerine (an anti-emulsification agent). Thus, when the glycerine phase is drained after the "pre-wash", it contains alot more methanol/soap than it would normally contain.

This is where my confusion is, because i didn't see how electrostatic pre-treatment would or could take the place of the 5% pre-wash. That's where I was hoping for some clarification from Graham on this.

Thanks!

Wes

Hi Wes,

I didn't word that very well.

The 5% prewash is obsolete in the process I described, but is probably still of value if you wash with water.

The soap is less of an issue when you use methanol removal to cause soap to precipitate. An emulsion is not going to happen regardless of how much soap is present.

Now, I've been fortunate to have been loaned a 4000V variable AC supply.

All experiments below at 24C liquid temperature.

I made an emulsion with 75% freshly brewed biodiesel (glycerol removed) and 25% water, vigorously mixed to make a thin mayonnaise, and can report rapid clearing of the emulsion with the voltage applied.

I didn't use plates this time, but point-source electrodes, one at the bottom of the jam-jar, one at the top, spaced 100mm.

To make it easier for me to judge the separation of water from BD, I added phenolphtalein to the alkaline water to make it pink. When thoroughly mixed, the whole mass was a slightly pink colour. As soon I switched on the 4kV, big pink blobs of water formed in the biodiesel and rapidly dropped to the bottom. I now have normal 'orange juice' on top, hazy pink water on the bottom and a narrow emulsion interface. This after 5 minutes. The control jar is still like mayonnaise, 1/2 hour later.

I also tried freshly washed biodiesel and this dropped its water within seconds, although it has not become crystal clear.

4000V makes it all happen faster. It happens at 1000V too, just a little slower.

Interestingly, when I switch off the power, a pink haze rapidly spreads thru the biodiesel, within a few seconds. As soon as I switch on, it rapidly migrates to the bottom.

I'll have a go at de-watering WVO and see if I can get any joy, with the higher voltage.

Glad to hear you're getting results with water separation. I noticed your using AC. Is there a reason and have you tried water separation with DC?

awesome,

Thanks for the clarification Graham!

quote:

Originally posted by GrahamLaming:
Hi Wes,

I didn't word that very well.

The 5% prewash is obsolete in the process I described, but is probably still of value if you wash with water.

I see your point, Graham, but it may be helpful to clarify: I suspect that those of us using KOH as a catalyst do not have as easy a time with soap removal. Just as the KOH won't gel when used in a reprocess test, I have yet to see any "solids" forming in my KOH-based biodiesel as I dry it in preparation for washing with Magnesol. So I suspect that may be one of the few benefits to using NaOH as a catalyst. This is also why I am keenly interested in using HVDC as a means to pull out soaps.

I cringe a little bit when I say this, but I hope that this all works out and magnesol goes the way of the Dodo.

-Jim

quote:

Originally posted by Jim D:
I cringe a little bit when I say this, but I hope that this all works out and magnesol goes the way of the Dodo.

-Jim

Me, too!

water is a polar molecule which may be a reason AC is used. It has a posive end and a negative end so I think would never migrate to a single dc electrode.

Am I right in thinking you would need around 33KV for a three foot gap or 11KV per foot?

I see the process as HV settle the glyc.

Distill off the methanol.

HV settle the soap and filter out the clean bio.

HV helping to settle/clump KOH soaps for easier filtering. Or possibly replace filtering.

No water involved.

The only remaining question is PH and catalyst content??

Anthother way to use the field with KOH producued bio would be to make the soap settle quickly to the bottom of the IBC and then tap off from above the bottom. only filter the high soap content bio from below the upper tap.

so can I make a safe curent limited DC supply capable of tens of KV? If higher potentials speed the process we could be looking at 50 or 100kv for an IBC. Even more might be faster.

Hi Ant,

pH / Catalyst content is truly negligible. After using the no-water process, I've taken a small sample of finished BD and washed in water, and compared pH of starting water to the residue at the bottom. Both the same, and the bottom water is crystal clear, even after a severely harsh shakeup.

I'm not so sure you'd need such a high voltage and such wide spacing.

I've noticed an interesting effect, since using phenolphthalein to make the water and wet soaps more visible, and that is that there is a lot of movement in the liquid, a flow of materials, while the field is applied. I need a magnifying glass to see it clearly.

I can see liquid which was not in the field being sufficiently stirred up by the motion to be brought into the field. It seems to clear the whole lot up OK, if taking a little longer.

May be all that's needed is to pass the bulk of the liquid thru a pair of medium voltage, fairly close-spaced electrodes, like this...



So long as the upward flow rate is set to suit the downward drop rate, separation should be swift and efficient. The whole high voltage section can be isolated into one simple external module, air free, to reduce spark ignition risk.

I've found 2kV upwards is tricky to work with, it finds leakage paths easily, so I've got spark coronas forming on the damp cables. I don't have that problem from 2kV and below. You may find 7kV just dissipates itself in a sizzling corona along the top of the liquid.

Here's the AC test setup, with a jar of just-thrashed emulsion, with phenolphthalein indicator...



I made some really nasty emulsion last night by adding a small amount of extra soap to BD and shaking up. The pink is from phenolphthalein I added to make the watery phases clearer.

The bottom dark red layer is soapy water, not glycerol. The middle pink layer is emulsion.

++++++++++++++++++++++++++++++

Here's the emulsion after 12 hours' settling under gravity, no electric field used ...




++++++++++++++++++++++++++++++



++++++++++++++++++++++++++++++

Here's exactly the same material after 20 minutes of 4kV AC field, one electrode on the bottom, one on the top a few mm below surface.

This has not settled overnight, this is 20 minutes after the emulsifying shakeup.




++++++++++++++++++++++++++++++

AC and DC seems to work the same, from what I can see, in dropping out glycerol and water. Breaking emulsion MAY favour AC, but not by much.

Soap in BD I still need more time to play with.

This message has been edited. Last edited by: GrahamLaming,

Graham did you ever getting around to testing HV DC for dropping water out of WVO?

I'd really like to see how far water content percentage could be dropped using this method.

Wes

Looking at Graham's "on the side" HV setup got me to thinking. I wonder if you could put your heating element there instead of a pump. Let it thermosiphon. It might flow slow enough to let the separation happen without stirring things up too much. Does temperature seem to have any effect on the separation? I wonder if this form of separation would lend itself to a container that it taller than it is around. With the electrodes not actually in the mix you might have some corners of unprecipited <sp?> glycerol/soap in the main container.

Keith