Biodiesel and SVO Forums Home
World Famous Dr Pepper Technique (Pat Pend)

This topic can be found at:

November 20, 2001, 06:52 PM
World Famous Dr Pepper Technique (Pat Pend)
World Famous DR Pepper Technique (Pat Pend)
The idea for the Dr Pepper Technique came about one day many years ago (it seems) when I was trying to figure out an inexpensive way of making my first few test batches of biodiesel.

I will assume that if you are reading this for the information then you are new to Biodiesel making, so just a few words of caution:



SODIUM HYDROXIDE (Caustic soda, NaOH, lye) can cause severe burns and death. wash off with water.

Long-sleve shirt, full shoes and trousers are recommended, no shorts or sandals.
Wear chemical proof gloves, apron, and eye protection such as goggles and a full face shield are mandatory.
Always have running water available to wash off any splashes.

Now that I have managed to scare you, just realize that Methanol is the fuel used in most Model airplanes. In the USA methanol is available in small quantities as HEET brand fuel line antifreeze (Yellow bottle)
Lye is an every-day drain cleaner.
Both are freely available in most large shopping centres.

Now find an area where it is safe to make the biodiesel away from children and other distractions. Outside is best. If you are inside it should be well ventelated. The Kitchen is not a recommended work area.

1 litre oil- new or used
NaOH (lye/ caustic soda), at least 6g. Used as a drain cleaner and can often be found next to the Drano.
Methanol at least 250ml. HEET® Gas- Line Antifreeze in the yellow bottle is methanol and readly available in most auto supply stores in the USA.

1- 2 litre (1.9 in US) Dr Pepper Bottle in sound condition with tight fitting top and dry inside. As NaOH/KOH will degrade the plastic used to make soft drink bottles over time, these bottles should never be used to make more than one batch of biodiesel.

1-measuring cup to measure out 250ml methanol
1-scales to measure 6- 7g NaOH
1 teaspoon measure metric or imperial.

1- container to mix the methanol and NaOH in which makes methoxide.

1- Funnel



If using waste oil, take one litre and heat to at least 120 deg c to remove all water. If water is present the oil will spit and pop and carry on. If there is a lot of water this could get very violent, so be careful.
Once the water is gone (Oil becomes calm and there is no more spitting and popping) let the oil cool.

If you are using new oil from the bottle it should have no water in it, so in this case just heat to 55deg c when you are ready to mix.


While the oil is cooling mix your methanol and NaOH(lye) to form the methoxide.
Use 250ml of methanol.
This is more methanol than most people use but will help insure a successful first batch.
If you are using new oil this will require 5g (about one level Teaspoon) NaOH.
For used oil, you should do a titration to determine the correct amount of NaOH to use.
However, if you do not have the materials to do a titration, just use the empirical quantity of about 7g NaOH as this amount almost always achieves separation. If you do not have a scales, this is about 1 slightly rounded teaspoon measure (metric or imperial).

NaOH and Methanol do not readily mix, so if you are doing it by hand a bit of time and patience is required. Don't sniff the fumes. CARTRIDGE RESPIRATORS DO NOT WORK WITH METHANOL
For quickest mixing, start with the methanol at JUST body temp (not warm). As you mix, the temp will increase substantially. This is normal. Make sure ALL the NaOH (lye) is dissolved. This may take 10 minutes or more.
Hand mixing can be accomplished using the back of a spoon to stir/crush the NaOH granules in a coffee cup or glass; OR placing methanol/NaOH in a glass bottle with a tight fitting top and shaking/ swirling until ALL NaOH is dissolved.
After ALL the NaOH has dissolved, top up to 250ml with fresh methanol, as there may be some evaporation during mixing.

When the Oil's temp has dropped to 55 deg c or a bit less, using a funnel, pour the litre of oil into a DRY 2 litre Dr Pepper bottle (in a pinch any other brand of bottle will do).
Take the mixture of methanol/NaOH (commonly called methoxide on this forum) and pour on top of the oil using the same funnel.
Remove funnel.
Screw the top down TIGHT onto the bottle.
Shake vigorously for about ten seconds/ 40 good shakes.
NO appreciable pressure is generated during this mixing.

Now place the bottle on a table.
If you were using WVO you will probably observe the oil change colour from a "Light Chocolate milk to a rich, darker brown."
Then, as if by magic, within 10 minutes the by-product (commonly refered to as glycerine on this forum) starts to settle out and form an increasing layer on the bottom of the bottle.
Be sure to notice that you can see a very definite, slowly sinking line towards the top of the Biodiesel as the glycerine slowly settles.
Within an hour, most of the glycerine will be settled out.
This is referred to as separation.
NOTE: Neutral has shown through GC tests that although one 10 second shake produces very useable Biodiesel it is in fact not ASTM conversion standard.
To produce ASTM conversion biodiesel requires titration and a further four- 5 second shakes over the space of an hour.

If you used New Oil, the colour change is not nearly as great and you may have problems seeing much of a colour change at all.

You should now have a bottle containing lighter coloured biodiesel on top of a layer of darker glycerine.
If you reacted New Oil the glycerine layer on the bottom will not be much darker than the oil you started with and the biodiesel will usually be very light in colour.
The biodiesel will be very cloudy, and it will take a day or two more for it to clear.

If it does not work out come onto the forum and see if anyone can figure out what went wrong.

And that is all there is to it. Only a little magic is involved.

Remember, after producing in excess of 53,459 litres of biodiesel in any calendar year using this process, please make royalty check out to Tilly From Paradise and send it to me through your favourite Numbered Swiss bank account.


A few cautions: DON'T mix the methanol and NaOH (lye) in a plastic bottle as NaOH attacks some types of plastic. Once mixed it is quite acceptable to mix your biodiesel in a Dr Pepper plastic bottle.
Do NOT store unused methoxide in plastic bottles. Some plastic will degrade over time when in contact with methoxide.

DO NOT allow any WATER into any steps of this procedure.

Methanol boils at about 65 deg c. DO NOT mix until the oil is below 60 deg c.

Again, these are dangerous chemicals and care is necessary, especially if there are children in your home.


I have Up-Sized the Dr Pepper technique by using a 20 litre plastic drum (longish and skinny) and make 14 litre batches. I simply lay the drum on a wooden "cradle" and place the center of the cradle on a pipe and rock back and forth for 3 or 4 minutes.
I then decant the mixture into a "finishing" container and so am ready to process another 14 litres straight away.

Now that you have a Litre of Finest biodiesel you will probably wish to wash it and this is quite easily accomplished using the World Famous Dr Pepper Wash Technique (Pat Pend)

Tilly SBC/IBA Dr Pepper Div

"No job is so simple that it can't be done wrong".

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

November 20, 2001, 09:41 PM

It is good to see a simple explanation of BD preparation that will work. One small point however:

It has been pointed out in several places on this forum that the reaction will go further if the stirring period is lengthened.

At about 55 degrees maximum yield is obtained if the material is kept suspended for an hour. Gentle stirring for an hour will do this, but if doing it by hand it is easier to periodically shake briefly over a couple of hours.

"Make everything as simple as possible, but not simpler". Albert Einstein.

007 at it again
November 21, 2001, 08:27 AM
My tests seem to indicate that at 55 deg c (or at most any other temp) 3.5 seconds VIOLENT mixing is sufficent.

Tilly SBC/IBA Relativity Div Cool Cool

"If the facts don't fit the theory, change the facts." Albert Eienstein.

This message has been edited. Last edited by: Tilly,
November 21, 2001, 10:22 AM

My original tests were done using gas chromatography (gc). They confirmed what other workers had reported, namely that an hour was needed when stirred hot for maximum conversion. The gc measures unreacted starting material and intermediates directly and can detect very small amounts.

When done at room temperature the reaction went just as far but took 4 or 5 hours till no further change in composition could be detected.

Of course this doesn't mean that once-shaken BD is no good. There is a glaring lack of information needing careful research into the question of how much unreacted starting material can be tolerated in various types of motor. If anyone comes across information on this topic it would be good to see it posted it in this forum.
November 21, 2001, 03:43 PM
Tilly and Neutral,
Perhaps we should be considering the rate of mixing when assessing reaction time.
With the Dr. Pepper technique the mixing is very quick and involves a lot of agitation for a small ammount of liquid. Evidence of the reaction is apparent very quickly. Solution of methoxide in wvo, uniform liquid, lowering of viscosity and characteristic ester smell. Using a hotplate stirrer and the same quantity of reactants it took much longer to reach the same degree of reaction. Even with a large stirrer pellet in 1L beaker, with pellet at max speed and vortex in liquid ten minutes or more would elapse before reaction point occured. Moving up to 4L of reactants in a 5L beaker with the same agitation extended the time even further. I resorted to stirring the mix vigorously with a large spoon, indication of reaction dropped to 3 to 4 minutes.
Obviously these observations relate only to an initial stage in the reaction. Intend to do a few tests on this one. Unfortunately I don't have access to GLC so will have to rely on SG of raw BD which could compromise results.

Regards squarepeg.
November 21, 2001, 06:24 PM
In a number of tests that I have performed, the results indicate that while a short vigerous mix is all that is required, it does take some additional time for the reaction to reach completion.

If you mix the oil at 55 deg c or warmer it seems to take about an hour for the process to reach completion as determined using the Dr Pepper Viscosometer (Pat Pend).

When I did comparison viscosity tests of identical batches with the only variable being the amount of mixing, there was no difference in viscosity. This seems to indicate that for all practical purposes the batches that were shook once for 3.5 seconds were identical to the batches shook a number of times over the hour.

Tilly SBC/IBA Cool Cool

"So many Christians. So few lions.
November 21, 2001, 09:26 PM

The difficulty of getting the reaction started in typical glassware is certainly a variable to be reckoned with. One way to reduce this variablity is to only start measuring the time after the viscosity drops. Once the viscosity drops, indicating the bulk of the reaction has occurred, it is easy to keep the material fully suspended. From that point on it should be just the inherent rate of reaction that is observed.
November 22, 2001, 05:52 AM
the mixing time for a good reaction would have to be more for a larger batch , say about 20l or more.I wouldn't mix a large batch of say 50l or more for less than half an hour.A 1 litre mini batch is no comparrison as far as mixing goes.
November 22, 2001, 07:23 AM
It has been my experience that WVO and methanol/NaOH (commonly called methoxide on this forum) do not readily mix together.

If you were to do a few Dr Pepper mini-batches you will notice that the methoxide will sit on top of the WVO almost indefinately with no intermixing. Even if you rotate the bottle end for end, the WVO and methoxide will not mix to any appreciable amount.

The best and quickest way to get them to mix together seems to be to make the mixing violent. Make the WVO and methoxide "crash" into each other over and over.

It appears that the "Quality" of the mixing is what is really important, and the Dr Pepper Technique provides the violent mixing that is required.

It appears that once this violent, intimate crashing mixing has taken place no further mixing is needed. The reaction seems to just go ahead and go to completion with no further mixing. At least that is certainly what my tests suggest happens.

I suspect the size of the batch has very little to do with it as long as you can get the high quality, violent crashing together mixing to take place.

As I remember, Dhill mixed batches of around 50 litres in a cement mixer for 15 minutes and gets good result.

BioCruiser Mixes 30 litre batches in 7 min using a drill.

Tilly SBC/IBA Violent mixing div Cool Cool

"The important thing is to not stop questioning. Curiosity has it's own reason for existing." Albert Eienstein.

[This message was edited by Tilly on 23 November 2001 at 02:41 AM.]
November 23, 2001, 11:33 PM
<James From Perth>
The secret of mixing seems to be in getting it happening as much as possible on a wide as front as can be.
ENERGEA use this idea in their continuous process, a large vessel filled with steel balls to increase the contact/interface area, and pumped through at a huge rate of knots. They claim that the reaction takes place in 3.5 seconds, though the mixture travels over several squillion square metres to get there. Heat and pressure are probably involved too.
The nearest we can get to this would be a blender....or a suitably shaped soft-drink bottle.
Or, if we have time on our hands, a low-powered blade mixer and a few hours.

James From Perth. Smile
November 24, 2001, 07:25 AM
Iv just gotta have my say ear

Tilly has proven time and again his argument on this forum of course this was thanks to James from PERTH who first bought up the idea (even if it was stolen from that OTHER mob)

after considerable foolin round following up this mixin idea I applied the age old KISS system to all the mixing methods I was lookin at and the winner is NOT pumps IE recirculation mixing
NOT in line pipe mixers

BUT yes the good old blade mixer in a barrel simple effective inexpensive and unlike a pump u do NOT have to drain the pump or lines

NOW how to get the oil methoxide to crash to gether well it really is simple and IF I can get my mixes down to 7 to 10 minutes with 100% conversion using TALLOW then why should I go BACKWARD and employing expensive pumps/vaccume units/ECT,ECT when I am achieving such good results with really inexpensive equipment...

how do I do it THINK "KISS"

BC SBC Perth Smile
November 24, 2001, 09:02 PM
The end result does not depend on how hard the mixing is done. As long as the surface, which initially has the methanol, is mixed into the oil the reaction will go. And that is simple to achieve as Biocruiser says.

The size of the mix is not a factor provided the mixing effort is scaled up in proportion to the size. It would of course be very expensive to design a 20 tonne vessel strong enough to be shaken so hard that the contents were fully dispersed in a few seconds, but it doesn't matter because the same result will be achieved in minutes rather than seconds with a suitable agitator. The harder the mix the faster the initial reduction in viscosity, that is all.

But nobody gets 100% conversion no matter how they do it or how much methanol or NaOH they use. It just is not possible in the nature of things. But most people appear to get enough conversion judging by the few reports of engine problems.

How much conversion is needed is a question not properly answered yet. From studying the information on this forum it seems that skimping on the methanol, which certainly lowers conversion, may cause problems with the "dreaded white stuff", emulsions, yield and separation, but as long as separation is achieved it seems the BD is good enough for the motor.

Of course most users of BD are driving vehicles with precombustion chambers. Few are using direct injection motors and, of those that are, many are in Europe and using commercial BD which has a high conversion. In the future more direct injection diesels will be built because they are more efficient.

I am keen to hear reports from people using direct injection motors and making their own BD. How much methanol are you using, and how far have you driven?

[This message was edited by neutral on 24 November 2001 at 09:52 PM.]
November 25, 2001, 12:29 AM
I've come across some oil that is a little strange. It tries to gel when you put the methoxide in. In a blender, you can see the oil thicken after the initial reduction in viscosity. If you're lucky, it gets past this thick stage in about a minute and a half. I wonder, therefore, if the doctor pepper method would work for this type of oil (used, from a restaurant).
November 25, 2001, 02:45 AM
Blanerobinson, the Dr Pepper Tecnique is definitely worth a try.
One test I ran involved comparing mixing times. Well used oil was the feedstock. This oil looked and smelt more like differential oil than veg oil. A vigorous stir with spoon for seven minutes gave reasonable results with a poor BD yield.
For the second part of the test a laboratory heater/stirrer/pump was used (centrifugal pump passing liquid over heating element, thermostatically controlled). Set the temp controller for 55 deg C. After one hour the previously liquid BD/glycerine mix had set to a slurry. Dr. Pepper tests on the same oil with varying ratios of reactants gave good separation.
Regards, Squarepeg
November 25, 2001, 04:13 AM
By all means give the Dr Pepper method a go.

Have you done a titration of your oil? Or are you just doing minibatches to find the correct amount?

Please keep us informed as to the outcome, as every bit of info is another step towards "Total Understanding".

No royalties due for private use until you make in excess of 53,246 litres of biodiesel in one calander year.

If you excede this amount just make out the royality check to Tilly From Paradise and diposit it in your favourite numbered Swiss Bank Account.

Tilly SBC/IBA counting house div Cool Cool

"People who say you can not buy happiness don't know where to shop."
November 25, 2001, 07:58 PM
this is very strange there are those on this board that have said to have achieved 103% conversion Personally I go by the amount of BD I get from the amount of oil I put in and no most occasions I end up with the same MEASURED amount

HOWEVER I admit this will drop down somewhat during and after the wash phase especially with Tallo , having said that it is not very much of a loss but it does depend on how much floaty white stuff I get from the Tallow..


November 25, 2001, 08:37 PM
The conversion % looks at how much of the oil actually was transesterfied and how much remained un-transesterfied oil that you started with.

If you use the correct amount of NaOH and 25% Methanol, it will make biodiesel that will be around 98% esters and 2% still un transesterfied oil. This has nothing to do with how much by-product (commonly called glycerine on this forum) is produced.

Glycerine (By-product) production depends on how used the oil is. The more used the oil, the more glycerine (ByProduct) is produced and therefor less biodiesel is recovered per litre WVO.

Remember, when transesterfying a litre of WVO we actually have a total of 1.25 litres of liquid in the reactor- 1 litre of oil and 250ml methanol.

Lightly used oil may produce more biodiesel than the starting amount of oil, BUT this BD will be only about 98% esters with the remainder of the biodiesel untransesterfied starting oil.

You cannot really tell by looking at the BD how much untransesterfied oil is present, Although the more untransesterfied oil that is present the darker the BD usually is.
BUT different oils naturally produce different colours of BD, so colour isn't a very accurate way of checking.

There is one sure fire way of checking what % of your BD is esters and how much is still un-transesterfied oil and that is using GC which is very expensive (so I am Told).

The next best way of getting a good idea of the quality of the BD is to check it's viscosity. At a total cost of about 10 cents Aus, the Dr Pepper Viscosometer is a very cost effective device as it can detect less than 1% difference of untransesterfied oil in similar batches of BD. But again different types of WVO (canola or tallow for example) naturaly produce different viscosity BD, so you have to know what the BD is made from.

Tilly SBC/IBA transesterfication div Cool Cool

"3 out of 4 people are 75% of the population."

[This message was edited by Tilly on 25 November 2001 at 09:36 PM.]
November 26, 2001, 01:18 AM
Arne N

You are speaking of 98% conversion at a maximum.
Can this limit be overcome with a 2-stage process, i.e. removing by-product and introducing more meth?

Is it the unreacted oil that screws up in direct-injection engines, and why?

November 26, 2001, 01:34 AM
Tests that I have done indicates that the two stage process DOES NOT produce better biodiesel than the single stage. Using more methanol does.

Possibly you could get a more complete transesterfication if you used 500ml methanol/litre WVO.

I think the chemists on this forum seem to agree that once the By-product precipitates it is removed from the reaction.

I am not aware of what the problem is with direct injection engines or what causes it.

But I am sure someone will provide the answer.

Tilly SBC/IBA Humming bird div Cool Cool

"Can I trade this job for what is behind door #2?"
November 26, 2001, 10:44 PM
Try heating your sample batch slightly. Fill the sink with HOT water and place the sample container in it.

If it starts to work properly, you will find that the mix is most likely made from a great deal of Tallow ( animal fat ). This happened to me and that was how I got it to go further.

That is one of the main reasons for sample batches. You don't stuff up great quantities at any one sitting.

See how you go.