GEK Wiki / Luc Gosselin in BC, Canada
  • If you are citizen of an European Union member nation, you may not use this service unless you are at least 16 years old.

  • You already know Dokkio is an AI-powered assistant to organize & manage your digital files & messages. Very soon, Dokkio will support Outlook as well as One Drive. Check it out today!


Luc Gosselin in BC, Canada

Page history last edited by Luc Gosselin 15 years, 1 month ago



     Last year, the price of gasoline here in BC soared to over $1.40/liter. This had a huge impact on my disposable income. It aslo made me realise how dependant we have become on large corporations for our power/energy requirements. As a result of this I started looking for an alternative energy source.


     I discovered what I think may be a solution for myself as well as many others looking to be freed form the power grid. The solution......wood gas. Wood gas has been around for almost a century and has been all but forgotten by the masses. 


     My intention is to inform the public that there is an alternative, and with the help of Jim Mason's GEK and a bit of knowledge, you can provide yourself with all of the power you need, regardless of your location.


The GEK is an excellent starting point for taking biomass and turning it into gaseous fuel that can run Internal combusiton engines for power generation and mobility, or gas furnaces. Unfortunately, the GEK is not a complete turn-key system for power generation. It is a work in progress and is constantly being improved on by Jim, myself and a growing community that are providing open source information to make the GEK better.


My goal is to help make the GEK into something that a lay-person can use without needing to learn about the technical aspects of gasification, but at the same time provide them the information, plans, materials should they decide to want to build what I am working on.


Knowledge is power.......and so is biomass.





The picture above shows my insulated hourglass design that I constructed. My hope is to sufficiently raise the core temperature in order to crack the tar. Although you can't see it this photo, the unit is completely filled with high temperature refractory castable. There are several holes underneath that I used for the filling/venting process. Once filled, I placed the unit in an oven and gradually increased the temperature to 250 degrees F over a period of 14hours. I have included drawings for this unit in dxf format as well.


I will be testing this unit shortly and will be posting results shortly.


Your input is welcomed.


Feb 17th


I'm up and running again. It's been a long process getting everything ready but here I am. I've taken video of the gas I'm producing. I think it looks pretty good. You'll have to let me know what your thoughts are. The true test is the tar test. I didn't have time to make a filter as it was getting very late and as I still hold a day job, I figured it was high time to shut it down for the night.


Luc modified GEK test run #2.MP4


This is a snap shot of my flame running the GEK at low speed.


I've been burning wood pellets made from pine. I'm not sure what the moisture content, but I think it is typical of regular wood pellets. I haven't had a chance to do a filter test, but I did have a quick look inside the fan and everything looks good with very little residue inside.




Luc auger design video.MOV


This is the auger feed that I came up with. The unit will fit under a large container that will allow the GEK to run for long periods so that I can tweak performance of the GEK. The auger will be controlled via microprocessor.


The container isn't shown here, but it is a sealed unit. You're right Jake. The auger does crush some of the pellets. Do you think this is going to have a big impact on my burn? Ideally, I would like to make a finer auger similar to the one that Jim fabricated, but I haven't had time to sit down and work out the details yet. I just needed something to allow me to have a long, controlled burn so that I can perform my testing. The nice thing with this is that it is modular and easily removeable.


Thank for the input. Keep it coming. I need your help to make this GEK better.






Comments (14)

Bruce Chovnick said

at 6:52 am on Feb 18, 2009


This looks very promising. Is the flame as good in person as it is on the video? Can't wait to hear how the filter test goes.


Daniel Chisholm said

at 3:17 pm on Feb 18, 2009

Hi Luc, nice video, thanks for putting it up!

What fuel were you using? Were you using the standard burner, or have you modified it (e.g. with an ejector)? I'm wondering why the flame was small, and also why it was fully formed and burned inside the burner can.

(account deleted) said

at 9:03 pm on Feb 18, 2009

Hello Luc
Could you possibly post a still picture of your burner flame. Many of us GEK owner live in parts of the world with dial up only. To view your vidio would be a 55 minute download that can only be done after midnight in my case.
Thank you

(account deleted) said

at 12:54 am on Feb 19, 2009

Thank you Luc. The Gek flame test run #2.bmp picture shows up fine. Very nice flame color-looks just like a clean natural/propane gas flame. So as I understand it, you are using the cast refractory to SHAPE your hearth FORM in addition to retaining and focusing the heat and provide probable thermal mass assistance? SteveU.

jake said

at 8:08 pm on Mar 3, 2009

Hi Luc

I was just looking at the auger. It looks really good and will have good control.
Just a few comments you will have to make sure your hopper is sealed up otherwise you will get CO leaking out.
If you want an open hopper there are a few ways to get around it. Best solution is a fuel lock with two valves one on the top can be a slide valve and the bottom one can be a flap sort of assembly anything else will get clogged up with tar.
I can see the pellets getting a little bit crushed in the auger maybe a finer auger could solve this.

Cheers Jake Martin

jake said

at 9:41 pm on Mar 3, 2009

Hi Luc

Pellets are not an ideal gasification fuel mainly because of the lack of structure.

What I mean by this is that wood blocks have cell walls and most wood will break along these lines think of a piece of charcoal in the fire it breaks down into uniform blocks. This keeps the bed of char porous and mean you can get the most out of your. When you make a pellet you break all those cell wall structural bonds. So when the pellets gets blasted with air they just fall apart and if they are thermally bonded then they will expand out as well. This causes a lot of problems as the fines clog the bed very quickly. The way around this is to take the fines out quickly (more turns of the grate) but this causes the inefficient use of the resource. I guess gasification is really fuel specific.

In your situation I would suggest a reduction tube that gets larger at the bottom to try and combat the fines clogging it up I would also make sure you turn the grate a lot. Your unit will probably produce more tar so use charcoal as the filter media. If it was me I would mix the pellets with wood chips and increase the size of the auger try an auger like this it seems ideal.

Cheers Jake Martin

jim mason said

at 10:11 pm on Mar 3, 2009

well luc, you win the prize this month for elegant simplicity. a pellet auger feed out of $20 of plumbing fittings and a bore drill is very impressive. congratulations.

here's a couple things we learned about augers that might be of help. (after the "don't attempt augers, they are a giant hassle" advice. advice we unfortuantely ignore regularly)

you can see the "proper" way to do this on the pellet stove sites. here is one with some pix of a pellet auger. you will see the size they use is about 2.5" total diameter. 1" pipe in center and flights of about 3/4" around. then in an appropriate pipe. they turn at 1-4rpm. much slower than yours. but you can pulse yours in the end.

we found the easy way to make an auger is is shaftless type, simply wound from rod. you can wind this on a twisted bar of appropriate size. or you can make them out of coil springs from motorcyles/bicycles/hardware springs. for a horizontal auger, the standard is that the diameter is equal to the flight pitch. if you look here you can see how we wound them. this one is an auger inside an auger. you would really only need the little inside one. the auger inside an auger is a reasonable way to get better traction at larger diameters, while still making these simply out of rod. making the proper auger flights is a dedicated machine. they are very difficult to diy.

also, these rod shaftless augers are much more forgiving than the solid flight ones. they bind less and generally move around problems. the biggest problems are always between teh wall and the auger at the entrance. in general, if you have less clearance from teh auger to the wall than teh size of your fuel, you will get binding at some point. you will get extra here due to the wide sides of your drill screw.

jim mason said

at 10:12 pm on Mar 3, 2009

(conntinued due to excessive length)

you can also fix the binding at the auger into tube transition by putting a turning tube at the pickup up end before you enter the static tube area. attach a section of auger tube housing directly to the auger, so it turns with the auger. this provides a transition area to get the material inside the tube without exposing it to an pinch point between the auger and tube. see how ice maker augers are done to understand. i just uploaded some pix of refrigerator ice box auger tech. see here

as for sizes, there is a claim from the materials handling world that you need a 5:1 or 8:1 (i always forget which one) ratio or higher between you narrowest opening and your biggest fuel dimension to ensure good flow without binding. certainly pellets will go through smaller than 5" feed holes. but i think you will find your current 2" is too small. you will likely get some binding on the fall down side. or maybe not. the biggest problem will be on new start up when the pellets have fallen apart from wetness while sitting. either way, you would benefit from a larger in and out if possible. i'd cut in a 3 or 4 inch hole for your lid entrance. you can use black sewer pipe for this. or just regular tube. then similarly do a open cradle at the intake, not the constricted 2" tee opening. more like an ice box, of course ;-) but again, all this is relatively minor and likely easy to fix. and it will likely work reasonably just as is.

jim mason said

at 10:13 pm on Mar 3, 2009

(and continued once again due to excessive length.)

on your auger connection, note there is a very easy way to cheaply do a lovejoy. you can simply use a square tube as the drive inside a slightly larger square that is the driven. the result is a loose self-correcting low rpm couple. see here: we call it a peefjoy, after the guy who came up with it in our shop.

aside from all those details, you just hugely increased the range of performance of your gasifier. being able to vary fuel rate in to make the gas flow needs allows you to keep your pyrolyiss rate in synce with your needs. no giant pile of fuel pyrolyizing from latent thermal mass, when you do not need all that tar for combustion. start up and shut down too can be much cleaner, as you can do both without raw fuel available for making pyrolysis messes.


Luc Gosselin said

at 11:36 pm on Mar 3, 2009

Jake, Jim, thank you for your help.

Jake, I didn't realise that wood pellets were not a very good fuel to use. I figured that pellets would have given me better performance by having the fuel size and moisture content consistent. You're right though, I have been struggling to keep tar down and have had to run my GEK at low speeds. I'm going to try wood chips now that I know better.

Jim, thanks for the advice on augers. I wish I had this knowledge prior to making mine. Oh well, I will use this for the time being and will work at building something like what you've done a little later. I didn't understand how your auger worked when I first saw it. That is partly why I chose this design. Intuitively, it seemed like a good way to build it at the time.



jim mason said

at 9:32 am on Mar 6, 2009


wood pellets are not the best fuel, but we also need to figure out how to make them work. they one of the very few consistent fuels that we can use to compare performance between units, as well as somewhat guarantee performance particulars out of the box. wood pellets are a raw standardized solid fuel, thus we need to find a solution.

the problems follow from them being compacted, and then can uncompact in steam. pellets when left a reactor/hopper after shut down will tend to decompose to sawdust again. the resulting mess just packs up on the next start. but an auger feed can solve this, as you stop the auger long before shut down, so there isn't any raw pellets left in the reactor to go do mush. once they are pyrolysized, they do much better staying together.

jim mason said

at 9:40 am on Mar 6, 2009


the bigger problems come from their small size and lack of regular woodgrain. the small size makes it more difficult for gas to penetrate the bed. the solution to this is likely smaller nozzles and generally smaller dimensions in the reactor.

the gek nozzles are designed so that you can put 3/8" plumbing caps over the street 90s that make them, and drill a hole in the cap to your desired nozzle size. you can also weld a small pipe in the nozzle ends and drill it out to what you want.

i would encourage anyone working with low gas flow needs, or trying to run pellets, to do this. the nozzles are default a 3/8" opening. if you read the imbert sizing charts, you'll see the motivation for this. remember also that we need somewhat larger nozzles than typical, as the air is preheated, thus larger volume. but for pellets, you should likely bring this back down. i suggest 1/4" for starters.

one way to figure this out might be to decide a max pressure drop across the reactor that you want to see, then keep dropping the nozzle size until the vac goes up accordingly. i often struggle to get the drop across the reactor even up to 1" h2o. this is on the low end. i believe one tried to see 1-2" h2o drop across the bed. this is mostly coming from the restriction at the nozzles, assuming you do not have a sawdust packed reduction bell.

so try to reduce the nozzle size until you see vac rise up to this range.

jim mason said

at 9:48 am on Mar 6, 2009


and finally, the lack of woodgrain in pellets is likely an issue. the lack of grain passage through the material likely changes pyrolysis performance, as well as reduction performance. the specifics of this i have not seen documented, but anecdotally we know you will get worse performance with pellets of x size incomparison to wood chunks of the same size. maybe pyrolysis is not proceeding as quickly and we are ending up iwth not fully pyrolysized chunks in the reduction zone. maybe reduction is proceeding slower as the surface area for reactivity is lower in pellets than open grain wood.

the proposed tar recylcing ejector venturi nozzles will significantly extend pyrolysis time, and thus likely help with the first problem. remember one of the likely big wins of the internal tar recycling scenario is it will create an updraft flow in the center of the reactor, somewaht in the manner of the beijer/susanto gasifier. fully finishing pyrolysis before the combustion/reduction zone is critial. and doing pyrolysis at lower temps, and thus over longer times, gives you a less refractory tar to crack, and thus better conversion when you have to finally crack it.

typical downdrafts end up having a very short and very hot layer right on top of combustion where the pyrolysis is happening. this is non-ideal. stretching this out will be good, especially for pellets.


Luc Gosselin said

at 12:26 am on Mar 9, 2009

Jim, I must be missing something here. Why does it matter if you can't pull more then 1" of water? For instance, let's say we have no blockage in the bed and the only difference being the sizing of the tures. Assuming that the fan is adjusted for the same gas flow, what effect does a change in vaccum mean to gas production? I can see that the initial velocity of the air comming through the tures would be higher, but this would slow down once inside and you would still have the same amount of air flowing through the bed.


You don't have permission to comment on this page.