3 replies to this topic

[bear@apl]

This is a quick report on testing of an idea to allow reasonable recovery from a packed reduction bell condition.  

The goal is to automate the recovery from a packed bell, sparing you the work and mess of manually digging it out.  This is done by reversing the flow through the system, running the reactor in updraft mode, and thus burning out the reactor from bottom to top.  This reverse flow can also be used to generate char from raw biomass for the first start up.

A packed reduction bell occurs when fine material (fine char and ash) has accumulated within the reduction zone, leading to a high resistance to gas flow and poor operation. We detect this condition by looking at the reactor "pressure ratio" (P_comb/P_reactor: pressure at combustion (nozzle level)/pressure at the grate (after reduction)). We've been using a threshold of 0.3, where a ratio greater than 0.3 is fine, and below 0.3 "bad". Values can go down into the 0.2, 0.15 range, and we've seen significantly lower as well.

The purge operates by essentially driving the reactor in reverse, running the GEK as an updraft instead of a downdraft. Air is brought in below the reduction bell and gas is drawn off the top of the reactor and flared. The air passing through the reduction bell should oxidize the fines/char to ash, allowing them to be removed from the bell. New char is generated above the reduction zone, which can re-fill the reduction bell.

This system was implemented by connecting tubing to the top of the reactor lid and T-ing after the flare/ejector side of the 1 1/2" NPT cross. A valve was placed on the tubing to only allow flow during the purge cycle. Air was introduced through a 1/4" NPT hole in the 1 1/2" NPT bung above the GEK reactor grate flange.

Results:
The first purge yielded a change in pressure ratio from 0.2 -> 0.5.
The second yielded a change from 0.2 -> 0.25, but was run for a shorter period. The top of reduction thermocouple was not reading, but should provide an indicator of when the purge is complete (e.g. combustion temperatures > 1000°C are seen at the constriction).
The switch over between modes could be accomplished with the flare combusting gas nearly the whole time. The flare color during the purge (updraft mode) was very yellow indicating the high tar content of the gas.

[bear@apl]

Purge circuit from pyrocoil lid to flare.  It is the flex ss line.




New 12V Hot Surface Ignitor from CrystalTechnica on flare

[Tom deSabla]

I've never seen what happens when the bell gets packed. Normally, if the bell gets packed, the run is over, the GEK has to be opened up, and the bell cleared, is that right Bear? Is that typically done coming up from underneath (the grate area) or does it have to be done from the top of the reactor?

So, this purge cycle makes it so you don't even have to stop running at all? Even with a packed bell?

Amazing. I hope it will also work with the new blower motor instead of the compressed air.

I appreciate how dedicated you all are to making the GEK better and advancing the art of gasification. It's really awesome.

(By the way, I can't see the pics)

[JayAlchemAPL]

Yes, so this is a way to clear the bell in situ with out having to shut down, waiting for the gasifier to cool, opening the gasifier and manually raking out the packed grate of char.
Just oxidize the char that is packing up the grate and it will turn to gas to then flow through the flare stack.
Great job Bear.
What was your bottom of reduction temperature reading during the purge?
-Jay