air-assisted injectors, sort of

[Chris Conlon]

Greg wrote:

> >Presumably this means that these injectors have a minimum usable
> >"on" time a good bit lower than the usual 1-2 msec or so? (Or does
> >the engine have a fairly low ratio of power at WOT to power at idle?)
> 
> More like the injector has a rather longer time to fill the pre-chamber
> with a metered amount of fuel, then the compressed air takes the
> (pre-metered amount of) fuel past the air valve quite quickly. The actual

But doesn't the issue of power ratio (or turndown ratio) still apply
here? How is it that these direct-injection devices can supply enough
fuel for WOT in 1 or 2 strokes out of 4 (max 50% duty cycle) and still
be able to open briefly enough to produce a good idle? Or am I
misunderstanding the injection-window for GDI engines?


> >Ignoring the issue of detonation, how ignitable are lean mixtures? Or
>
> 17 or 18 to 1 is the limit of flammability for a homogeneous mixture, no
> matter how hot the spark.

Does this depend on CR? I would expect a very hot, high pressure mixture
to sustain burning better than a cooler, lower pressure mixture at the
same A/F ratio. (I could see it being a lot harder to ignite, though.)


> > I just wonder if any of those efficiencies can be brought to a TPI setup.
> 
> I wasn't thinking of bringing the lean mixture efficiencies to a TPI
> setup--you need the stratified charge effect for that. All I would hope for
> is the improved fuel economy and power due to better fuel atomization and
> reduction of compression work by evaporating a higher percentage of the
> fuel during the compression stroke. I think that both gains would be
> significant--and, although this is unexplored territory, no-one has come
> forward with any wagers against it yet.

I do agree with you that there are gains to be made in these areas. I
have no real idea how to *calculate* what these gains might be though.

My own line of questioning is pointing down a different road; I think
there may be a lean operating point (with water injection) which could
give even better BSFC and still low emissions.


> As to your thoughts on detonation:
>
> My SUBJECTIVE feelings on it are that there is a critical
> P-T-mixture ratio for any given fuel above which detonation will occur,
> but that there is also a (material and conditions specific) time delay
> involved. Once you get the mixture above the critical point, the
> countdown starts, then after a certain time , it detonates. This feel
> is based on a lot of engine tuning. Also a bit of (field) experience
> working with explosives

I'd say this is pretty close, but I'd amend it a little bit. The
pressure and temp will be going up (mostly) during combustion, and
IMO the shape and peak of this P-T graph is mostly what pushes the
engine into detonation. The shape of the combustion chamber can
have a big effect because of the burning rate curve - look at how
the flame front expands through the charge, and see *how much*
mixture is burning at any one time, and see how the *amount* of
flame front area changes over time. This is the gas engine equivalent
of "grain shape" that I didn't realize last night.

Anyway my experience is that pressure and temperature by themselves
can be enough to cause detonation, and that the temp/pressure rise as
combustion progresses is mostly responsible for detonation due to
high CR or advanced timing.

One example is smokeless powder (modern gunpowder) in a gun barrel.
It has to burn; detonation is as bad for a gun as an engine. But if
the projectile should get jammed in the barrel, the rapid pressure
rise can cause the powder to go from burning to detonation, with
often disastrous results. 


> My experience with engines seems to indicate that if you can get the burn
> over with quickly (as a result of short flame travel, mixture motion, and
> mixture turbulence, as opposed to fast flame travel) the chances of getting
> detonation are VASTLY reduced,  the need for ignition advance is also
> reduced, and BSFC is improved.
>
> I have actually worked with an 11.5 : 1  C.R. engine whose chambers were
> conducive to the above mentioned things happening, and I observed the
> following: It would run on 101 octane (unleaded) with 45 degrees of total
> advance with no detonation whatsoever, but it made its best power with only
> 32 degrees of total advance. Otherwise the same, but with different chamber
> shapes, the same engine did not make as much power, and could knock its
> brains out on 104 octane. Power was still increasing with increasing
> advance with more than 40 degrees, but detonation prevented going any
> further whilst looking for more power.

Eeeeh... part of me wants to say that if you look at the pressure
profiles in each chamber, you'll find that the bad chamber had a
higher pressure (static or dynamic) due to the shape, or increased
timing advance, and that pushed the mix over into detonation. But
OTOH you could be exactly right, it may be that if the flame front
(which probably does not burn all the fuel) moves very quickly through
the mix, the residual A/F mix (which is still burning, really) is much
less able to be detonated even at a higher pressure/temp.

It would be very interesting to see pressure vs. time profiles for
each of those 2 head designs, and figure out the burn rate profile
as well. It sounds to me like the good chamber burned so quickly
the peak pressure lined up well with the optimum time to "push" on
the piston, and the bad chamber burned more slowly so more advance
was needed to try and get peak pressre at the optimum time.

> Does your experience working with explosives back up any of what I
> have (kind of lamely) tried to say here?

Yes, or at least it gives us more parts of the puzzle. Unfortunately
explosives makers are more interested in reliable, repeatable, definite
detonation, rather than the grey area where a mixture might burn or
might trip over into detonation.

I can tell you, though, that *shock* is 1000x better at causing
detonation that heat or static pressure. Nitroglycerin, in the open,
will simply burn, as will a great many explosives. But add
confinement, sometimes even the slight confinement of a largish pile
of the substance, and it may trip over into detonation. (This
depends greatly on the explosive; I'm sure you've heard of soldiers
using TNT or C4 to heat their food.)  Anyway this suggests that
if the mixture is burning while it's still being compressed, this
might cause more detonation than igniting the mixture at a
higher (but stable or decreasing) pressure. Aka too much ignition
advance.

Huh... think I'll call my head-work Guru and see what he has to say
about burn rate vs. knock resistance. Very interesting.

   Chris C.