GEK Wiki / Louis-Martin Dion @ McGill University (Montreal, QC, 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.

  • Stop wasting time looking for files and revisions. Connect your Gmail, DriveDropbox, and Slack accounts and in less than 2 minutes, Dokkio will automatically organize all your file attachments. Learn more and claim your free account.


Louis-Martin Dion @ McGill University (Montreal, QC, Canada)

Page history last edited by Louis-Martin Dion 10 years, 2 months ago


I am pursuing a Master's Degree in Bioresource Engineering at McGill University in the Macdonald Campus. I'm part of the Biomass Production and Controlled Environment Laboratory under supervision of Dr. Mark Lefsrud.


This research began as an undergraduate design project on Biomass heating of Agricultural Greenhouses. Common heating systems of greenhouses use Natural Gas and can also use the CO2 emissions from the system's exhaust gas in the chimney to provide CO2 to the growing plants (flowers, vegetables) of the greenhouse. CO2 enrichment is quite common and well proven in the greenhouse industry. By adding CO2, photosynthesis is optimised leading to higher yields. However, CO2 enrichment is mostly done with fossil fuels or liquid CO2. We wanted to explore the possibility of doing CO2 enrichment form the emissions of a Biomass Heating system.


Fine particulates is one of the biggest issues of Biomass heating system requiring expensive Electrostatic Precipitators to reduce PM to acceptable levels for the workplace. This is one of the reasons why we wanted to explore Gasification technology. The syngas produced in the reactor is combusted to produce the required heat and CO2 for the greenhouse. Since it is a filtered syngas that is combusted instead of the raw biomass, much less PM is emitted in the chimney.

Therefore, for CO2 enrichment applications, we are really interested in the Emissions of Syngas Combustion. We are planning to take samples and measurement of Flue Gas as well as measurements of the syngas (for H2 and CO content). We will also be testing different kinds of Wood Pellets.




We have installed and run our Gasifier a few times. For our tests, we put the GEK flare inside the Combustion Chamber of a Biomass Direct Combustion Furnace. This combustion furnace has a air-to-air heat exchanger to distribute heat to the greenhouse and a 6inch chimney for exhaust gases.


Here is an update of the instrumentation for the GEK. We have a total of 10 thermocouples, 4 pressure sensors, one outlet with valve for syngas sampling:

- Additional Thermocouples to analyze reactor's operating Temperatures (Top thermocouple seems very important: If Hourglass is clogged or if reactor is depleted, top temperature and pressure can rise considerably, can deform the top lid (it happened once) and be dangerous. Top thermocouple becomes one of  signs to know when to shut down the reactor.)

- GCU                                                                                                 - Syngas sampling port on top of Charcoal Filter

- Real time recording of data                                                            - Chimney port for on site flue gas measurements with portable analyzer.

- Set up                                                                       - Flame during early tests before modification of Burner. (CO emissions : around 20,000ppm)


- This is our Syngas sampling apparatus. Syngas was originally passed through two glass tar condensate trap in an ice bath. Syngas is captured in teflon bags to be analyzed in a GC. For exhaust gas sampling in the chimney, gas is passed through a particulate filter and a moisture trap and then captured in teflon bags for GC/MS analysis.


Burner Modifications:


Our goal was to achieve stoichiometric ratio for cleaner Syngas Combustion and reduce our undesirable emissions. The air intup that maintains the necessary 5inH2O vacuum in the reactor does supply enough oxygen for complete combustion of Syngas. Considering that we would be running our burner for over an hour, we wanted to avoid having a pre-mix flame to reduce any risks of flashback. We also wanted to monitor the additional amount of air (O2) to the burner (with pressure and flow meters). Adding the option of propane combustion  was also necessary to reduce toxic emissions during start-up, shut down and purging.


- The most recent version has 2 pressure sensors and 2 flow meters                                        - The recent version has more copper tubbing for cooling

The modification we did consist in splitting the compressed air inlet: one way maintains constant GEK internal pressures while the other way (yellow pipe) allows variable additional air input. During early tests, we tried to maintain the flame in the middle of the burner on top of a mesh. The mesh (steel, mineral wool, alumina spheres) would help cooling down the flame to avoid propagation at the bottom of the burner. However, these trials were only successful for the first 15min: without water cooling, the heat would inevitably propagate to the bottom of the burner.


- Burner with a mesh of Alumina Spheres. It worked very well for 10min (requires better cooling system + better air/fuel injection adapted to this application: complete syngas combustion for CO2 Enrichment in Greenhouses)

- Flame with current conditions.


Propane combustion: We moved the propane inlet closer to the location where we wanted it to burn (at the Burner: blue ball-valve) in order to make it safer and avoid flashbacks. The original input (tee pipe fitting close to the compressed air inlet) was under too much pressure and would have flown air towards the propane tank (we are using a standard BBQ propane tank).


For current experiments, no mesh is used. So far, the simple addition of an extra air input has lowered our CO emission to an average of 40 ppm (ethylene C2H4 is in average at 0.1ppm). We are still doing experiments in order to have more rigorous numbers.


Current and Future Activities:


- Perform mutliple runs under different air input conditions

- Better modification of the Burner: Stainless steel or Pyrex + Better air/fuel injection design.



Experience with full GCU:

The GCU has been really helpful to record temperatures of the Gasifier. It provides a real time view of the Reactor's core temperature in order to adjust the Gasifier's vacuum pressure and keep the reactor over 850oC. It also helps viewing temperatures at the top of the Gasifier which provides helpful clues for when to shutdown the GEK. We are currently trying to improve our skills with the GCU:


- We are having difficulties to record data from the pressure sensors (the LCD screen shows data that make sense when comparing with the Water Manometer. However, when transfering into a computer, the pressure values are completely different. It is confusing since Temperature data transfers very well but not the pressure values.)

- We are looking to record data on an SD card.



I will be very interested in comments, suggestions, opinions, etc. Thank you for your interest!


Comments (2)

bk said

at 9:56 am on Aug 9, 2010

Hi Louis-Martin,

Congratulations on your setup, it is very impressive.
It will be very interesting to see the data and results you achieve with this. We as of yet haven't had many opportunities to do gas analysis or emissions testing, but of course will be into the future.

Quick thoughts/comments:
Flare emissions: The photos of the flare show it running in a very rich state (when the flame is outside the flare it indicates that more oxygen is needed to fully combust the fuel). Have you run the flare with a stoichiometric/lean mixture (combustion fully inside the flare)? We adjust the mixture with the brass ball valve on top of the 1.5" NPT cross. I'd be curious to know the emissions characteristics when run in that state.

Fuel: I see you are running wood pellets, at least in one photo. Pellets are a somewhat tricky fuel. They'll likely require a high reactor vacuum to pull in enough air to maintain proper temperatures at the top of reduction (constriction) to effectively crack tars. What temperatures are you achieving there? Temperatures there should be >800°C, ideally above 900°C.

Reactor Lid Temp: Yes, good idea on alarm condition/low fuel detection with the top TC. I've started implementing some state checking to produce alarms if something is wrong. I've also thought that a simple thermostat on the lid (available from Mouser) could be the low tech (non-GCU/TC) method to check for low fuel. Also see that if pressure at combustion and reactor are ~equal, the fuel/reduction bell have burned out and is now void space.

Piezo Ignitor: A simpler device is the new silicon nitride ignitor elements that are available. Search for silicon nitride/hot surface ignitor. There are vendors that make them in 12V, 120V is more common.

Great work on this.


bk said

at 1:42 pm on Sep 19, 2010

Louis-Martin, great to keep hearing back with your progress.
Re: GCU:
- Pressures - the display and datalogged values are in different units:
on screen with current libraries: in inH2O*10
datalogged values: in Pascals
values are handled internally as Pascals. 1000 Pascals ~= 4 inH2O
note that unfortunately, different GCUs had pressure sensors installed in different orders, there's a note on this at (see the note on GCU_Setup()), if not correctly set pressures will be reported at the wrong scaling factor.
- Firmware for the SD card isn't yet adapted to the GCU. Something we'll be working on.
- I'm in the process of learning Git to do proper version control on the sketches/libraries, etc so we can all collaborate/share code.
- See, for example:


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