a while ago mr. wayne keith mentioned he was measuring vac/pressure in his hopper and comparing this with the vac/pressure drop across the full reactor. he was using this to see when his bed was getting clogged and the grate needed shaking to purge things. if the hopper gauge dropped, but the full reactor gauge read high, there was new resistance in the bed and grate shaking was needed.
i thought this was very smart and we adopted it for our datalogged test runs. we changed it a bit and instead took this "new" vac/pressure reading at the nozzle level in the hearth, so as to eliminate any effects of the gas flow through the upper bed and gas volume production in pyrolysis, both of which might effect a reading at the top of the hopper. thus we went to just taking this second reading right in the combustion zone, behind or to the side of the nozzles a bit.
we originally were using this to know when to shake the grate to maintain "steady state" for the tests. but while doing so, we learned some unexpected things about the relative contribution of the nozzles and fuel bed to the total pressure drop across the reactor. the results have suggested how we can use these two vac/pressure readings to tune the size of nozzles in relation to different fuels, as well as see when the bed is plugging and its time to shake the grate, as wayne originally showed.
here's the emerging nozzle size tuning guidline:
we're seeing that when nozzles are in the realm of properly sized and creating the needed blast velocity to penetrate the bed, the total pressure drop across the reactor is about half from the nozzles and half from the gas passage through the bed. the new pressure/vac tap in the hearth is measuring the pressure difference across the nozzles, or from atm to the point here the air is entering the bed. using this new measurement, we are seeing a 1:1 ratio between the pressure drop across the nozzles and the drop from the resulting gas flow through the hearth and out, seems about right to create the combustion lobe spread we need to fully convert tars.
this "new" pressure at the nozzle level inside is easy to take on the gek by using the lighting tube as your sample tap. put a barb on the end of the lighting tube (once done lighting) and attach it to the standard manometer. put the other channel of the manometer on the usual end of the reactor tap. now you have the total reactor pressure drop and just the nozzle pressure drop. the difference between the two readings is the bed drop. or read it as the nozzle reading should be 50% of the total reactor reading.
if the the nozzle drop is less than 50% of the total, your blast rate is too low and you don't get the proper bed penetrating lobe happening. if your nozzle drop is more than 50%, you are not going to be able to achieve the full bed flow rate without extremely high total reactor pressure drops, which will impact the engine pulling the gas. thus a guideline for actually measuring nozzle sizing in situ, and having some motivation to adjust it accordingly (as well as compare it to other machines and results)
the nice thing about watching this ratio is that it can be used to resize nozzles in relation to what is needed for different fuels. if you go to a larger fuel with larger void spaces, less blast rate is needed to penetrate the bed, and there is less resistance to gas flow through the rest of the hearth. the nozzles will need to go larger to maintain the same 1:1 ratio between nozzles and bed pressure losses.
similarly if you go to a smaller fuel. there will be more resistance to flow through the fuel and smaller nozzles will be needed to penetrate the bed. nozzle size will need get smaller to up the pressure drop across the nozzles to increase blast rate, in proportion to the increased pressure drop across the bed.
whether these relationships really vary linearly, i do not know. whether the proper relationship is really 1:1, i can only say that is emerging for us to be a useful rule of thumb. whatever these numbers refine to be, what is clear is that we can use this "new" pressure reading point to start to develop a formal method to size nozzles to create the proper blast rate conditions. there seems to be a range of proper relationships here that we can use to define where good performance will be found.
previously most of us just use the traditional charts, but had no real feedback from the real system running with real fuels. what is often needed varies significantly from the charts in the end. but knowing how is only available to those long deep in the black goo and well paid in gasifier hazing dues.
i'm seeing that we can make this an easy formal measurement. like a timing gun or vacuum gauge for tuning an engine. we can similarly set up formal measurement points and related numbers to diagnose and adjust the particulars to success.
we've also seeing from recent tests we can do the same on the temp side. with two simple measure points -- the combustion constriction and the base of the reduction bell-- we can diagnose most of the temps that relate to min and max pull rate for good tar conversion, minimized soot production, and the generally usable gas flow range. i'll write that out soon too.
(Addendum: these speculations have now been formalized into a complete gasifier operation and tuning method call "The Masonic Method". See here for the full description.)
How to size nozzles for an Imbert type gasifier
No replies to this topic
0 user(s) are reading this topic
0 members, 0 guests, 0 anonymous users