No replies to this topic

[jimmason]

the goal in tar cracking in an imbert type gasifier is to maintain an adequately high temp and residence time combustion lobe across the full hearth, all the way to the restriction. the goal is to spread adequately hot cracking temps across the full the upper hearth area, all the way down to the restriction, and not let any tar pass around the edges, center, or other cooler spots.

thus tar cracking is an issue of combustion zone performance- the area from the nozzles to the restriction.  tar cracking is not really an issue of reduction zone performance.  this is often confused, and important to understand at the outset.  

but how might we measure this in some easy real time manner, which we can monitor to predict our real time tar production/conversion?  can we find some real time data point like this, attach it to an "idiot light", and help all of us driver our gasifiers?  such is the topic of the note below.


------------------------------------------------------------------------

from what we can determine through observation, and what we can find in the literature, raw temp is a much more powerful lever towards tar cracking/conversion than the catalytic effects of char on cracking tar.  of course having them both together might be better, but such begins the problems of trying to establish a full combustion/cracking lobe in the presence of char that is constantly trying to suck out your heat for the endothermic reduction reactions.  in a fixed bed downdraft, we're always trying to do our real thermal work under the equivalent of a water bucket brigade.

this is why we need the void space in the fuel.  we need open volume where the gasses can exist at high temp for enough time to complete the cracking before they start interacting with the char and temps drop.  as fuel/char sizes get smaller, there is more surface area for reduction reactions, thus it gets harder to maintain the high temps for adequate lengths of time.  this is why larger chunk fuels are easier to complete tar cracking.  this is why small wood chips or pellets are so difficult to get to work.


thus to make things work in an imbert, we have to find some combination of void space, penetrating nozzle blast, hearth geometry and pull rate so that a sustained combustion and cracking lobe is created across the entire upper hearth area, with propagation all the way down to the restriction.  the most important thing here is full and complete temp PROPAGATION, not just high spot temps in front of the nozzles.  figuring out how to measure this propagation and resulting tar conversion in real time, with simple data signals, is what we've been trying to achieve in our recent tests.

our hypothesis is that the best way to measure this fullness of propagation is the temps achieved at the restriction, NOT the temps right in front of the nozzle.  if the nozzles are sized reasonably, and we find some X adequate tar cracking temp still present at therestriction, we can reasonably conclude we had even higher temps and lobe fill above this, which were the conditions actually responsible for the tar conversion.

or said another way: are there indicator temps at the restriction that we can correlate meaningfully with tar production?  if there is, this is would be a very discrete and easy to measure data point that gives us the most important info: are we making tar, or are we not making tar.

we have been very pleased to find there is in fact a very strong correlation with temp at the restriction and tar in the output gas.  we've found that if we have 900c or above at the restriction, tar will disappear in the output gas.  see here for graph of constriction temp vs tar produced  (and thank you again to bear kaufmann for running all these tests and producing all this data).

http://gekgasifier.p..._s2_v_tredb.png

you will see the line of tar quantity zeroes at about 900C.  this suggests the temp above the restriction was actually somewhat higher, but 900c at the restriction related to some other higher temp above that was adequate for the conversion.  

this is not terribly surprising in some ways, as we've long had the rule of thumb that you need to get tars above 1000C to fully crack them.  however, this means across all the hearth, and at some adequate residence time, thus finding 1000C right in front of the nozzles is not terribly meaningful.  the tests often had 1100-1250c or more temps at the nozzles, but note how poorly the nozzle combustion temps correlate with tar conversion.  see the very messy correlation here: http://gekgasifier.p..._s2_v_combb.png

if you want all the multifuel tests and all the graphs and ingest all the details, see here:
http://gekgasifier.p... Run Comparison

we were very happy to find such a strong signal of restriction temp with tar convesrion.  there is a very clear and and rather linear relationship between temp and tar cracking success.  this gives us a very simple probe point we can use for instructing newbies how to drive a gasifier.  

"put a TC probe at the restriction in the hearth.  if you maintain that point above 900c, you will likely have tar free gas.  you will likely have tolerable gas down to about 800c of restriction temp.  800c is about the bottom of where the corresponding tar samples seem tolerable to us."

thus we might have a new rule for driving a gasifier:

"always keep the restriction temp above 800C!"

one can monitor this manually.  or electronics can automate reactor conditions based on this metric.  we can also move this threshold temp around depending on needed gas purity, or other fuel variables we learn with more testing.  the main point is not really the exact threshold temp, but rather that we do in fact have a strong and measurable temp indicator for tar production.  this measurement is fortunately also at a lower temp where one is not constantly melting TC probes, as is the case in front of the nozzles.

i'm very excited about about the reduction in black magic needed for successful gasification suggested by the above.

jim