With my original plans shot down in flames, I’ve had to regroup and head in a new direction. I’ve been looking at all the information I can about Imbert based gasifiers for this time around. Along with changing from FEMA to Imbert, I think I’m also going to start looking for a new engine.
I had thought my original engine plan would be adequate, but the more I read the more I find that is less than desired. My Wisconsin-Robin is designed to give a maximum output of 16HP at 1800RPM. I never really gave it much thought before but part of the engine is a 2:1 gear reduction housing, so the engine is actually spinning 3600RPM internally. I know that some people have successfully run small engines on woodgas at 3600RPM, but my understanding is that with the long burn cycle of woodgas, a slower moving cycle of the pistons is more desireable. The Wisconsin-Robin also runs a 6:1 compression ratio. I read that higher compression engines are more desireable - I haven’t actually found the reason in print yet, but I’m thinking that drastically increased timing coupled with low compression makes for very low power? If I’m wrong, could someone help me understand the issue with low compression ratio? Anyways, it looks like the Wisconsin-Robin will probably be tethered to gasoline for the time being. Since my original goal was a 5000-8000 watt generator, I’m going to start looking for a slightly larger engine to run on woodgas - probably something around the 30HP range. It’s hard to say whether it will take me longer to build the gasifier or find the engine I want, but I doubt either will happen overnight. To me, owning a working gasifier is no more important than having the ability to design and fabricate it, so time isn’t critical.
Even before I started looking at Imbert gasifier designs I came across the All Power Labs site and their GEK design with the free plans. I’m sure this is a great setup, and may well be on my list of things to do in the future, but at the current time I think it’s a little too much of an undertaking since I’m doing this one all from scratch. There is a lot of custom fab work and special hardware to it, and for my first build I am trying to go a little simpler and cheaper. This isn’t to say that I can’t take a few lessons from the GEK: The first thing I noticed were the stainless steel tubes wrapped around the reactor tube. After learning that these were to help preheat the incoming air, but would be a rather costly addition to my first attempt I thought about how I could do something to preheat the air a little, and decided that if I couldn’t easily implement the wrap-around tubes, I could still run them vertically inside the gas jacket to at least absorb some heat.
It wasn’t too long after I sketched up my first design with the air preheat tubes that I came across the drawings of Richard “Pepe” Lemieux. His sketches were pretty much identical to what I had started planning. The fact that he has taken a build-it-yourself approach to everything including gasket making mirrors my philosophy on many things. Many thanks, Richard. This has given me something to look toward when I don’t know what to do, but even though I can see the answer I sometimes still have questions about how to get to the answer. I’ve got a tentative sketch in progress, but there are a few spots where I am unsure.
I am tentatively basing my design on the table 5.2 Imbert dimensions, with a hearth opening of 4" and a reactor diameter of 10.5", and I have the following questions, most of which are from table 5.2 in the Handbook of Biomass Downdraft Gasifier Engine Systems.
-
Table 5.2 shows a simple inverted V hearth. I assume that if you use this, you need to close off the area around the hearth opening to direct the fuel toward the hearth opening? I also make the assumption that this is why people use the hourglass shaped hearth? If so, are there any special dimensions, angles, and clearances required for this?
-
Table 5.2 shows the bottom of the inverted V flush with the walls of the reactor. Is this critical? I’ve seen another sketch on the GEK site that says the diameter of the bottom of the cone should be 2x the top diameter, or 8" in my case. Is anything in this range acceptable?
-
Table 5.2 shows dimension “H” as 10.125" from the bottom of the hearth to some point slightly above the nozzles. What is this dimension for, and where does it go to?
-
Dimension “R” is from the bottom of the hearth to what? Is that the dimension to the grate, or the bottom of the container? if not to the grate, what is a good dimension from the bottom of the hearth to the grate?
-
This is somewhat related to question 3. Is there a minimum and maximum dimension from the center of the air nozzles to the top of the reactor? I assume there is not any hard number for this, as it is just the fuel storage hopper, or am I not understanding something?
This is all I can think of at the moment, and I’m sure I am probably overthinking some of this, but it’s always easiest to do the measure twice cut once approach, especially with metal.