dual fuel concept

[mobile_bob]

with the link provide by

post #42

http://www.microcogen.info/index.php?topic=430.msg11065#msg11065

got me to thinking about something that was brought up a few weeks ago, that being the addition of gasoline
into the inlet of a diesel engine and using the oem injection system to provide for an ignition source.

i had always assumed that gasoline would detonate severely in a diesel engine, particularly those that are high compression
such as my changfa idi s195, which is 20:1

what i found out was the engine does not detonate with the admission of gasoline into the intake, i checked that at part load
using a crude dribble method. (don't do this at home, i was prepared for fire and damage)

apparently the researchers at the university of wisconsin are using a standard gas injection setup from gas engine's to dual fuel
so it does not take specialized high pressure system to get the gas into the cylinder.

post 42 above relates to a computer system for control of ignition and fuel injection of a single cylinder engine, so...

if i were to use a throttle body from a ford small 4 cylinder car, and replace the injector with a unit from a small v8 with port injection
i would then have an injection system approx sized to the 195 running at 1800rpm,  i could control the unit with the computer linked to

now i don't need the ignition/spark for the cylinder, but it could be used to relight in the after treatment possibly, and
i don't need the throttle plate as it would be set to wide open throttle anyway, but

it could be very useful as an air cutoff device for emergency shutdown if attached to a suitable solenoid.

the interesting possibility to me would be that in dual fuel mode the exhaust emissions could be very clean, and compatible with
a cat converter, also the puter should provide for control of an EGR valve as well.

the end result could be a very clean running engine, at a set rpm (in my case 1800rpm) and at a very narrow load range (approx
9kwatt mechanical in my case), having very low CO, unburnt hydrocarbons,and low nitrogeous oxides (possibly), and have very low PM to contend with in the PM filter stage.

with some good programming and some thought, it might be that the engine could be made to run cleanly using a variety of ignition fuels
and gasoline and maintain the very clean emissions (relative to the stock changfa) and if the University of Wisc. report is correct the engine
thermal efficiency might also be significantly improved.

seems like a worthwhile research project to me

comments?

bob g

ps. lest i forget

we also get the ability to use a "real" air filter assembly, and also resonator if we pick the right donor car
along with all the other needed sensors, such as intake air temp, coolant temp, map or maf sensors
cam or crank sensors.

[Crofter]

If the motive is to burn up the diesel particulates it might work. I dont think you can extrapolate a lot from as you describe the crude mixture control of the experiment. It wont detonate if it is exceedingly rich or if it is exceedingly lean (NOX though) It also will not detonate under light load in a throttled engine with high static compression but in the unthrottled diesel engine I think it might be happy only under narrow load conditions.

Igniting the gas mixture with the diesel injection gives up almost all control or variable timing but it will force ignition of a mixture too lean to spark ignite at conventional gasoline compression ratios. Might be a good way to utilize some low quality wood or bio gas mixes but I dont think it would be an efficient way to use up top quality gasoline.

[mobile_bob]

Frank:

i would agree with you in total had i not read the reported results from the university of wisconsin

what they are reporting is so significant an increase in efficiency, running in dual fuel mode, that it just cries out
"try me"

i am thinking it might be fairly easy to find out, in that my application is a fixed speed/fixed load condition unlike
their application to a truck cat engine that is under variable speed and load.

it would appear to me that it would be infinitely easier to "dial" in a fixed speed/fixed load than it ever would be
a variable speed/variable load application

they are reporting over 50% thermal efficiency, that is a huge improvement over anything in class when we are
talking about a truck engine.

as for variable timing i don't think we need it, we set the timing for the rpm and load we want to operate under
and there then is no need for changes.

looking back, had i not been able to admit "any" quantity of gasoline to my changfa's intake air flow without setting off
detonation, or in other words had i found it just to finicky to admit any without the thing starting to knock or ping, i would not
be suggesting this concept now.

because i was able to admit a variable amount of gasoline, without pinging or knocking, i am led to believe it may well be possible
to use gasoline in dual fuel mode.  if this turns out to be the case, and i can use and oem gasoline intake injection system as suggested by
the university researchers, then it stands to reason the quantity of gas can be tailored very accurately and then the BSFC can be tested
against load to determine if what they are reporting as an increase in thermal efficiency is replicable.

"if" it is replicable, my sense is that would be the single most important development for these engine's i know of, and i would be all over using
the technology.

"if" it worked out that i did not have to alter the injection timing, then the engine could be used as a standard diesel burning straight diesel, and then operate as a highly efficient gas engine running in dual fuel mode.

it also opens the door for burning alcohol based fuels as well.

i am thinking for a simple test of concept, perhaps a throttle body could be mounted to the intake (not a problem) and the injector fired
with a simple pwm unit so that one could dial in the amount of fuel injected while the engine is running under full rated load. then a series of
BSFC tests could be run to determine the optimum pulse width to get the best economy at that power level. if it turns out that the efficiency of the engine shows significant improvement, maybe then it might be something to further explore with the system under computer control.

bob g

[vdubnut62]

For some reason I get the feeling in my gut that I am overlooking something, BUT.....
Since you will be running in such  narrow parameters, why couldn't you just use a venturi/ jet/ needle valve assembly in the intake manifold to introduce a the small amount of gasoline. I'm probably wrong, but in your particular application I suspect that the ratio is not going to be that
persnickety.
I'm probably way off, I'm not sure I understand the purpose of all this. Is it supposed to increase efficiency, burn cleaner, decrease overall fuel 
demand, or all three. I got kinda confused by the link.
Ron

[mobile_bob]

they report a dramatic increase in efficiency, which should relate to lower fuel consumption, and because
the fuel is mainly gasoline the cat converter has an easier time processing the waste gasses, and because
diesel is used as only a pilot ignition source the particulate matter should be dramatically lower as well.

if any, all or any combination turned out to be replicable, i know i would be interested in finding out.

when i think about it, i have to ask

"why are we married to diesel or oil as a fuel source?"

to me i look to a diesel for higher efficiency, lower fuel consumption, and ease of maintenance afforded by the
engine's simple design (to a lessor degree).  Therefore i am very interested in any report that would support significant
increase in efficiency, economy, and clean burning, most especially if it can be done with "obtanium" parts.

with the EPA reg's getting nothing but tighter, and they will get much tighter as time goes along, it just seems like we ought to
at least be forward thinking and utilize any technologies that are available to us.

i suppose we could sit around nursing a 6/1 on waste veggie oil for a while longer, but how long does anyone really think that fuel
source is going to be "legally" available?

there are those that are working on converting the 6/1 to burn woodgas, methane, frog farts etc by means of an electronic ignition
and fitting a spark plug, why not explore duel fueling these engines?

perhaps at the relative low power density of these engines, they don't follow popular belief or thinking when it comes to what is possible.

hell, if someone would have asked me a month ago whether i could inject gasoline into the intake of my changfa running at 1800rpm
and at any load level, i would have scoffed too!  i would have expected the engine to protest mightily, yet it didn't!

so something is in play here that runs counter to what i have been trained to believe. i have been trained to believe "any" gasoline either in the
diesel or injected into the intake was a path to destruction.

apparently what kills those diesels i was trained on, does not necessarily kill a changfa? and apparently according to the university of wisc it won't kill a cat engine in a truck either. quite the contrary it would appear that it is beneficial in a number of ways, at least with the cat engine.

bob g

[Westcliffe01]

Bob, in your situation, I would look for the smallest gasoline injecor I could find.  The injector off the 1000cc 3 cylinder Geo Metro engine or a Mercedes Smart would be good examples. It is highly unlikely that you would ever be running the thing flat out.  Bigger injectors cannot be turned down beyond a certain point, due to the response time and the large size of the orifice, which has poor atomization below a certain flow rate.   For a dual fuel application, I would imagine that you would want very fine control of the dosing down to very small amounts.  Hence the recomendation for the smallest injector you can find.

For your experiment, a simple signal generator with adjustable pulse width running at a frequency of 20-30hz would be all you need.  You should do some research into injector driver circuits, since the coil generates pretty severe back EMF when the driver transistor switches.   My tip in this regard is this special purpose chip and circuit:
www.national.com/ds/LM/LM1949.pdf  It is purpose made do do this job and the circuit works perfectly if you follow their recomendations.

[Crofter]

If they are getting 50% thermal efficiency into the crankshaft that has indeed got promise. I did not read the particular article. I did watch the developments a bit a number of years ago when Honda? was experimenting with the stratified charge engine which would initiate ignition in a small portion of conventional mixture which would then burn the major portion of a much leaner 17 or 18 to one charge(memory?) I think compression was quite high and good engine controls were necessary to prevent detonation. I think you have that covered with the idea of throttle body injection and electronics.

I think that a lot of us in our youth have messed with upping compression, advancing timing and carb twiddling to get a bit more performance out of the cars; I am sure we could all identify with ping! I cant force myself to believe that a full throttle well atomized mixture of gasoline and air could not be found that would make your changfa ping! I would guess it might have been too coarsely mixed. Anyways that is not what you are proposing for a working experiment so I am only quibbling on that point. There could be some kind of a symbiosis in the two fuels burning together that would give an extremely efficient operation under the right conditions.

If a heavy load came on, the mixed charge could be cut and the normal diesel function would respond. Under light loads the gaseous mixture could be just blessed enough by the diesel flame and high compression to carry a light load ( something the diesel is not good at by itself!) I think that systems virtue well might be for small cogens with fuels such as alcohol from algae etc,  with solar powered extraction.

For larger installations I dont think it will come near the turbine cogens for efficiency though. I have worked on several installations that uses a jet aircraft engine as the prime mover. Dual fuel too; natural gas or diesel but I dont think they mixed except at transition. Dont let that sound like I know anything about a jet turbine though, as I was just running piping according to the drawings!

Could be good for some experimenting for

[mobile_bob]

Westcliffe:

the need for a very small injector was the thinking behind using the ford throttle body for a small 4 cylinder engine,
the reason being ford uses a bosch style injector, which could be swapped out for an even smaller injector from
a port injected v8 which uses 8 injectors, 1 per cylinder which should be much closer to the proper size needed for the changfa.

thanks for the link for the driver chip, i was aware of their existence but would have had to go looking for them, so you saved some effort
there.

if i get a little time, i guess it is time to go kick around my favorite pic/a/part yard and see what is available, parts there are very cheap
a throttle body is about $7.50 and the injectors go for about 5 bucks each.

so it wouldn't cost much to put together enough stuff to experiment with.

bob g

[jimmason]

this is a very interesting scheme bob.  i had not ever heard of someone trying this and i too am surprised that it works.

my guess is that the lack of detonation is due to the partial load you are running and the lack of mixture within the flammability limits of the gasoline.  gasoline has surprisingly narrow flammability limits, so at a significant partial power your gasoline mixture is going to be very lean, maybe so lean that it is out of range for detonation.  once the diesel injects and starts the combustion, out of limit mixtures will also ignite.

here's an interesting chart of flammability limits of common fuels.  note the extremely wide flammability ratios for hydrogen and carbon monoxide, which is what makes wood gas as forgiving as it is.

http://gekgasifier.pbworks.com/Combustion-Characteristics-of-Common-Fuels

also it would be interesting to repeat your experiment in an indirect inject diesel, like the common lister.  we often speculate that these will be more sensitive to detonation due to the edges of the precombustion chamber than can run hotter than the general cylinder and piston.  the DI engine has fewer edges to worry about and thus can usually take higher compressions in dual fuel.  or at least this is what we like to tell ourselves with wood gas.  i've let to test them side by side, but such is the common understanding.

jim mason (newbie)

[mobile_bob]

welcome to the horde Jim, hope you find our little corner of the world interesting.

i have been telling myself i was going to quit testing and do the complete tear down of my machine
for final welding, sandblast and powdercoat, but damn... there is something about the dual fuel with gasoline
that just keeps gnawing at me.

the question about the system being load sensitive has me thinking that i might go back and retry the crude experiment
with the unit running at full rated load and see if i can get it to detonate with the admission of gasoline into the intake air stream.

i suppose if the thing does not detonate, i will be left with no other choice but to go ahead and fit the engine with a TBI system
and give it a whirl.

i am betting that because the engine does not detonate a part load it will also not detonate at full load, reason being there is no
restriction in a diesel engine's intake, so the intake pressure will not change with load as it would with a standard throttle plate controlled
gas engine. if there is no change in intake pressure there should be no change in effective compression ratio.  the only thing that likely would change is the cylinder temperature, wherein it ought to run hotter at higher loads.

now this brings up a possible explanation as to why the engine does not detonate, that being the engine has a fairly wide overlap, wide valve overlap allows for additional cylinder cooling because of the extra airflow across the cylinder. it might be that because of the extra airflow the cylinder is just cool enough to keep from having hot spot sufficient to promote detonation.

if this turns out to be the case, higher loads might generate high enough cylinder temps even with the overlap cooling to cause detonation at some point in loading. also it might be that i would have to time the injection event to start after the overlap event so as not to draw gasoline across the cylinder and out the exhaust.

the more i think about it, the more i am thinking there are a whole lot of things going on that might have bearing on what is happening.

bob g

[Crofter]

Partial load should equate to lower cylinder temperature= lower pressure in charge (BMEP) = less tendency to detonate. Flame propagation rate increases rather linear with increases in these conditions up to the detonation point then the pressures and temperatures multiply 4 or more times almost immediately.

Just a guess, maybe the lean mixture does not ignite immediately at the start of the diesel spray like it would with spark ignition. Perhaps no ignition of the gaseous mixture until there is a sizeable flame for radiant heat to ignite the charge. That could be somewhat behind the spike of the main diesel ignition.It could be late enough that the piston is descending rapidly enough that pressures doesn't rise too quickly to the detonation point.

The last part of normal diesel injection can be poor combustion under part load; perhaps the accompanied burning of the gas mixture cleans this up. Strange! There must be some very different things going on than would be the case if you tried to spark ignite a 12:1 air/gas mixture in a 17:1 compression ratio engine.

[mobile_bob]

if nothing else it might be useful to get a gas injection system in place, to add some flexibility
and might prove useful for other fuel use, such as wood gas?

under computer control, closed loop with O2 sensor the injector pulse could be altered in real time
to keep the mix near optimum even if the wood gas became intermittent.

or if one were to get it worked out, it could have the injector replaced for a spark plug and adapter
and then add a spacer to reduce the compression ratio and make a gas burner.

nothing wrong with having a true multifuel if not dual fuel capability should the crap hit the fan

bob g

[playdiesel]

It is interesting that you noticed no additional detonation noise. Maybe because the small amout you dribbled in was tto lean to ignite on it's own? or maybe because the other racket going on with the diesel was so loud it covered up? My DI 1115 is so loud with diesel knock that a person could not possibly notice a small additional amount of noise. In my opinionated (slightly opinion there is no way that a gasoline mixture rich enough for operation is NOT going to detonate  in an engine with high enough compression to light diesel fuel, if it didnt, the diesel would not ignite either this is an unavoidable fact. Add to that that it is not timed when introduced into the intake stream and can ignite at any point BTDC you are inviting a Kaboom!! Adding some gasoline to clean up emissions sounds interesting but unless the engine was one I could laugh at if it gernaded I would be very leary of introducing gasoline fuel without timing its introduction.

Maybe I am just and old grizzled fuddy duddy but I am very leary of reports like the one you quote that would seem to be a whole new world of internal combustion. I have read about hundreds of such things over my 54 years that turned out to be nothing but dreams or total farses. Those universities are competing to get our tax dollars and ho-hum doesn't make the grade. "50% increase in efficicency" ? Ya right, show me something other than words  on a paper,  I know,, I am the ever doubting,,,,

[mobile_bob]

well i did some research into the background of the guy that wrote the press release turns out he is one of the higher ups in control
of the engineering dept of the university of wisconsin

the gent that is responsible for the research has quite a long history in research of atomization, fuel injection issues and some other stuff
going back over 25 years and is a professor in the engine research dept of the engineering school.

i contacted the former who passed along my email to the latter, who kindly sent me a copy of the SAE paper that was presented and accepted
for publication.

now i have not read the report thoroughly enough to understand the finer points, but have read it enough to begin to understand the processes
involved. 

i understand enough to get out and buy an injection TBI unit off of a 2.5gm car, complete with fuel lines, air cleaner and snorkel. this leaves me to
make up an adapter to fit the Throttle body to the changfa manifold, then i will simply connect to an auxilliary fuel source (gm pump in/tank)
and then put together a pwm controlled injector driver chip to control the amount of fuel injected.

while it is extremely doubtful that i will achieve the same level of thermal efficiency they did with their test engine (interestingly it was a single cylinder diesel) because i cannot optimize cam events without an additional expense i am not likely to accept, my hope is to perhaps see some incremental increase in thermal efficiency of the engine.

my testing and measuring ability is good enough to capture and document even a 1% increase in thermal efficiency in BSFC measured in gr/kw/hr
i would just have to account for the amount of both fuels used in dual fuel mode by total weight.

from what i have read so far, the gasoline modifies the burn characteristics of the diesel injected into the engine, it lowers the peak combustion temperatures, alters the pressure angles, and reduces emissions dramatically, along with some other stuff i have to go back and reread a few dozen more times.  it is not a simple diesel btu + gasoline btu= output sort of thing, but rather a more complex process is happening.

this might be what is missing in my understanding of what i think it should be?

maybe between further study of the SAE paper, and some testing on my engine i can then get a better appreciation of whether there is anything to this or not.

bob g

[jimmason]

bob,

it seems the efficiency increases here are mostly from more complete burning of the diesel fuel.  this is the same effect one gets with propane fumigation of a diesel, or dare i say it, HHO (more accurately called hydrogen and oxygen derived through electrolysis).  diesels generally have such poor combustion completion characteristics that even making your fuel with an inefficient electrical based process can be net positive given the gains using more/all of your primary fuel.  you can get a similar effect in an otto engine by cooling the charge with water injection and then running it lean.  the usual rich mixture to keep things from overheating only leaves fuel for the cat to burn to usually not useful heat.

cooling the charge via a second fuel vaporization event will help your total power potential given the greater charge density, but should not effect efficiency, at least in a diesel.  the top and bottom temps should still have the same delta.  in an otto that is throttling the intake, heating the intake during partial load will actually increase efficiency, as you need less throttle thus less throttle losses pulling vacuum against the butterfly.  you can't get this bump in a diesel, as it doesn't have the original problem.  this lack of throttle losses is why a diesel is so much more efficient than an otto at partial load.

if you want to explore these relationships, it might be easier to do it with propane than gasoline.  you will also get more cooling with propane than with gasoline, assuming you feed the liquid and let it vaporize after injection. 

but whether propane or gasoline, i think you are going to find there are fumigation limits based on the detonation character of the fumigation fuel used.  for instance, in the wood gas world, when dual fueling an engine, the substitition ratio limits vary with the compression ratio of the engine.  if the cr is too high, the wood gas will move into detonation at high substitution. to prevent detonation you need to reduce the amount of wood gas substitution.  the total temps achieved on compression are not likely changed much, but the detonation point does, apparently even within the ignitable flammability limits.  outside the flammability limits, as one will have on a very lean mixture, detonation should be impossible.

wood gas seems to have an absolute cr limit of somewhere around 17:1. +/- 3 or so depending on which one of us is waving our hands about it.  either way, it is well known to be well above gasoline, and well known help clean up poorly atomizing diesels, but it too can still detonate if the substitution gets too high.  this gets worse with a turbo or and idi engine.  the changfa at least avoids these problems.

it will be very interesting to see what fumigation limits you find with gasoline or propane.  i know at least with starting ether, the substitution limit is about 0% . . . ;-)

j