Biodiesel and SVO Forums Home
GL's Eco-System processor

This topic can be found at:

February 28, 2010, 03:04 AM
J Long
GL's Eco-System processor
Originally posted by Hannes:
Good grief, I have been making bio for 6 years now with more than 250k km done on my own B100. Only recently came to this site and had the revelation that I am still brewing with dinosaurs. I have read and read and read just to catch up.

Q. The venturi is used (a) to mix in the methoxide and (b) to mix in air for drying and demeth. Once the methoxide has been mixed in (the venturi will then start to mix air in), do you bypass the venturi for the rest of the mixing/reaction period or do we continue with the venturi?

Q. Electricity has become very expensive here in the good old RSA, so I am not iterested in methanol recovery. I do however treat the next batch wvo with frist stage glycerol and discard the 2nd stage glycerol. Therefore not interested in whole batch demething. Am I correct to asume that demeth of the remaining bio in the reactor is done via the venturi mixing air in?

Answer 1.
The venturi is to get the methoxide into the reactor/processor. Once that is done, the flow is redirected to just mixing. Most of the GL users have installed a eductor to make the mixing more efficient. This is an extra gadget they have added to make the processor more efficient.

Answer 2.
Understandable about the electricity. There are ways to cut the methanol recovery, and mixing. This is the whole purpose of the AAF method. It will reduce processing time down to minutes, not hours (by reports....I have not tried it yet). This is based on using the flash vapor recovery, which hasn't been published yet. This method is used to recover the methanol quickly, but the procedure is very critical. Methanol is the largest cost of biodiesel (at least in the USA). If you can recover almost half of what is used (7% of the total 20%) then you can save a lot of money with methanol recovery.

With the AAF method, there is not a dual process. It is a single process to perform the reaction. The AAF method puts the methoxide in first, then mixes preheated oil into the methoxide. This has been reported to allow the reaction to mix more thoroughly. The better mixing ability allows the reaction to happen faster.

The venturi is mainly to remove the methanol. If you are not going to remove the methanol, the venturi will do little good.

These ideas are all about efficiency. To do them, the design has been significantly changed. I believe I have figured out the design changes, but will not say anything about the flash evaporation. I will let the designer disclose that information when he feels all the safety issues have been addressed with his design.

James L
March 25, 2010, 07:47 PM
Hi all,

I am interested implementing the venturi into my biodiesel process. I am using an in-line heater between the pump and the reactor. My question is if anyone is cognizant to the fact that the venturi will be pulling MeOH vapors (from the outlet of the condenser) as well as air (from the vent) together which will be going across my heating element. This would cause a situation of potentially explosive conditions, no?

If so, should I then consider putting the venturi after my in-line heater?

Is anyone carrying out the setup that I just previously mentioned??

Chemical & Energy Engineering -- The Pennsylvania State University
March 26, 2010, 05:45 AM
Hi Tokashila

The top of the tank should be on the outlet side of the venturi.
The venturi should be on the outlet side of the heater.
The heater should be on the outlet side of the pump.

You should not allow air to enter the inline heater, and the heater should not be able to drain dry.

Hope that helps,


Rover 75 + Skoda Fabia on B100

Bicycle on G100 12,000 miles p.a. ( )
March 26, 2010, 07:51 PM
Yes very much. That does sound like the safest way to go about using the in-line heater and venturi in series. Thanks for you quick reply.

~Ryan J.

Chemical & Energy Engineering -- The Pennsylvania State University
September 20, 2010, 12:50 PM
A word of caution from a pump guy. If you have a vortex at the bottom of your tank, you are most likely getting air in your pump.
September 26, 2010, 04:40 PM
Originally posted by J Long:
I believe I have figured out the design changes, but will not say anything about the flash evaporation. I will let the designer disclose that information when he feels all the safety issues have been addressed with his design.

James L

As a matter of interest the Fash Evaporator has been running fo several months now with great success.
On my pilot plant I can demeth, Bio only, in 35mins, Whole Batch Demeth in 75mins. We have a brewer with a FE on a 180ltr (WVO volume) reactor WBD in 2hrs. We are recovering methanol at 98%+ pure. The use of reclaimed methanol means we only use approx 13% of virgin meth per batch, that's one hell of a saving over a 12 month period.

To cope with the massive rate of recovery I've moved on to a Quad PD and incorperated a Vapor to vapor heat exchanger to strip off as much heat as possible and retun it to the reactor thus reducing the energy cost. To reduce cost even further the demething is controled by a PID.
September 26, 2010, 05:20 PM
Hi James, where do we read more about this please?
September 26, 2010, 05:47 PM
Originally posted by Paulus:
Hi James, where do we read more about this please?

Hi Paulus,

Long time no speak. Here y'go

Flash Evaporator
September 27, 2010, 06:01 PM
Hi All,
First post - Been making bio for 6 months now (newbie)in home made rig with 205 litre oil drum with reasonable results. I follow the forums to widen my knowledge and try to keep my cock-ups to a minimum. Found a 46 kg propane cylinder and so I am planning to make a GL hybrid processor with a few of my own mods (experimental) to see if I can fine-tune the process to suit my day job. Will post if anything concrete becomes apparent. One thought occurs though. Graham's processors are pressure vessels to a certain degree. If we are reducing pressure to recover methanol and dewater quicker and at lower temps, is there any advantage to reduce processing times by increasing the pressure when processing? Obviously there would have to be controls over pressure increase. Increased pressure would prevent methanol vaporisation and hence process temperature could rise perhaps shortening transesterification time. I am not a chemist so would welcome expert knowledge on this subject. Just trying to find ways to reduce amount of time spent in workshop without spending millions on full automation with PIDS, sequence valves etc. Fed up with water washing and all associated problems so trying for whole batch demeth with acid neutralisation (H2SO4) when I build new rig. Regards to Graham for his valued contribution and all you others who freely give your test results for idiots like me to follow easily.

'There are more questions than answers - the more I find out the less I know!'
September 28, 2010, 07:37 AM

You would be ill advised to pressurise your plant in a domestic situation.
As for speeding things up you obviously haven't been following the vegetableioldiesel forum here in the uk.

I've taken Graham's Eco system to the point where I can convert a batch in as little as 15mins, demeth bio only in 35mins and WBD in 75mins returning as much as 40% of the methanol used at 98% pure.

What more would you need, all the details have been published and quite a few of the brewers over here are using my system with great success.

NO pressure, NO vacuum, NO need to re-invent the wheel.

You have a lot to learn young Luke.
September 30, 2010, 09:59 AM
Ok James,
Thanks for the heads-up on that one. Better I come and see you and discuss the finer details before making a balls-up. Might save me trawling through tons of posts and just confusing myself and asking stupid questions when in reality the answers are already out there.

'There are more questions than answers - the more I find out the less I know!'
April 15, 2011, 12:56 AM
would the plans remain the same with the exception of size of plumbing and pumps if i where to use a 500 or 1000 gallon tank?
April 15, 2011, 04:04 AM

Just in case you weren't planning for this: your going to have a lot of glycerol separation and thus will be pumping a lot of glycerol near the end of the rxn

I believe this problem was brought up in GL's process to place the pump inlet abt 1/5 the way up the reactor or higher (imo right at the phase separation b/c that's where monoglycerides collect)

Would also use static mixing, but switch to no static mixing for near end of reaction and for water pre-washes

Chemical & Energy Engineering -- The Pennsylvania State University
July 08, 2012, 03:33 AM
Hi All, as you can see by my join date I've been lurking this forum for quite some time. I've always been keen on eventually building a processor and making biodiesel. Now that I have time and a bit of money I've gone for it. I've scoured all the helpful ideas, suggestions and hints here and elsewhere trying to adopt "best practices" for quick production turnaround with an eye towards hopefully profitable commercial production capability.

With my pilot small 10 litre capacity practice rig I originally I was going to make a GL push-pull, but ended up with essentially a GL eco-processor with many of the push-pull features. I've incorporated an inline heater as with the GL push-pull, a foam trap as with the push-pull, and an external venturi vacuum system also as with the push-pull in addition to the venturi vacuum found on the eco-processor. I've made a plumber's delight condensor and I've also incorporated a flash(y) evaporator that tee's off after the heater with it's own separate pump and pressure switch for quicker de-watering and de-meth'ing.

Everything works as it should independently and in isolation. Right now I'm trying to refine the process so that I can go from raw feedstock to a finished ASTM-worthy product in less than 24hrs (quicker the better). In order to maximize yield I've chosen to go with the acid esterification (if need be) and a two-stage base transesterification. The procedure is essentially the eco-system process (with two stages of transesterification instead of one), with the addition of flash(y) evaporative de-mething for the sake of speed.

This is where I'm having some trouble. After increasing the temperature to 83 degrees or so, I start directing the flow through the flash(y) evaporator. I get a really good rate of methanol recovery for about 15 minutes and then suddenly, for no reason that I can discern, the vessel undergoes a fairly sudden change of positive pressure and the batch starts to fill up the foam trap and vacuum lines, into the separate auxiliary vacuum system, through the condensor and into the distillate reservoir. Basically everywhere. Can anyone explain what is going on here?
July 31, 2012, 11:12 AM
Hi Beernini,

I have seen this also, I think it happens when there is not enough methanol to keep the soap in solution, and you reach a point where the turbulence helps the soap to foam.

I reduced the foaming dramatically by adding a little silicone oil to the batch.

Experiment to see how much you need, because silicone oil is not cheep, but you shouldn't need much to keep the foaming under control.

Hope that helps,

Best regards,


Rover 75 + Skoda Fabia on B100

Bicycle on G100 12,000 miles p.a. ( )
July 31, 2012, 02:22 PM
I too had this problem with the GL1 and WBD and Flash evaporation.
My fixes were:
1. A PID regulated temperature control
2. Leaving a larger headspace when WBD so as to leave more room for the foam to break before being sucked into the condensing system.
3. As you did, added a foam-over trap, actually a translucent oil filter housing, so I could catch the process before it ruined my recovered methanol.
4. I have a separate tank port for the sight tube which connects into the top of the reflux before the foam over trap and condensor. By watching this sight tube it would indicate when tank pressure was increasing over the methanol vapor recycling pressure, see when foam over was about to happen and adjust what i was doing, like cutting down on the venturi flow to only what was necessary to move methanol vapor. I think one of Grahams earlier posts on his site shows a manometer type setup to watch pressure changes.

Is your flash(y) evaporator a nozzle sprayed into the headspace at 2.3bar?

TDI 50,000 mi on Bio100, when it's not below freezing here