Because there's more to the internet than hits alone, Peter Hill
wrote:

Without liquid injection the compression can't be raised to take
advantage of the octane rating.

OK, I'll bite. Yes it can. You seem to be thinking only in terms of
induction density. What's wrong with changing the dimensions of the
combustion chamber?

A change of compression ratio from 9.3:1 to 12:1 increases efficiency
by 12%

12% of what? Do you mean that an engine will start returning in the
region of 35-40% overall efficiency, or do you mean the gain will be
12% of about 25% or so, giving something like 28%?

(it also gives more power). Each increase in octane by 6 will
allow an increase of 1 in compression ratio. Petrol is 95 octane,
Propane is 111, this should allow the engine to run 12:1. BUT only if
the inlet temp stays the same. If inlet temp goes up you can't
achieve the same gain and an inlet temp rise (by feeding it hot vapor)
completely negates the possibility of any gains.

I'm not convinced that your interpretation of figures accounts for all
considerations.

The effect you describe of cooling the charge, making it denser, has a
similar result to supercharging - both are ways of enabling more
charge mass to be packed into the combustion chamber during induction.

But there is another way, too. Increase the stroke. Although
volumetric efficiency will not be comparable, both can achieve much
the same thermal efficiency, which is, in the main, a function of how
much the charge, once inducted, gets squeezed.

The comparisons you give may be right for the same induction
temperature, but that is not the same as saying that induction
temperature has to be the same. We need to consider the nett result of
a specific mass of mixture compressed to a specific volume. My
understanding is that in a) a cooled, dense charge going into a
compression chamber (liquid injection), or b) an ambient, less dense
charge going into a higher compresion chamber (vapour injection),
thermal efficiency would be the same. But the relative power/torque
curves of mechanical output would not necessarily be so, and the
engines may have different characteristics.

A normal petrol engine has it's inlet temp reduced by about 20 deg C
due to vaporisation of the fuel.

In theory, (or so my back-of-envelope clacs. show) LPG that becomes
vaporous in the inlet manifold could reduce the charge temperature by
more than this. In reality, other factors would probably prevent it
from either achieving or sustaining this.

As well as improving volumetric efficiency, this temperature drop
would translate into some cylinder cooling resulting in lower
pre-combustion temperatures. But probably not as much as you may think
- there is a lot of heat capacity in the elements around the cylinder
compared to the relatively trivial capacity of the inducted charge. A
lower charge temperature could mean slightly higher compression
pressures could be run and thus better thermal efficiency. At the
moment I've run out of envelopes, but I'd be surprised if in practice,
lowering the temperature of the charge by a few degrees on a working
engine enabled much of an increase in TE.

Every 4 deg C lower inlet temp
results in 1% efficiency gain (hot vapour feed results in 5% loss!).

Are you talking about volumetric efficiency or thermal efficiency
here? VE figures are lovely to bandy about at dinner parties, but
don't necessarily equate to good TE. If you do mean TE, then do you
mean 1% of 100%, or 1% of 25%? And is this on a specific engine
configuration, or does it account for variations up to the compression
limit?

[..]
to 30 deg C (2.5% efficiency gain)
[..]
allow an increase of about 15% on
[..]
an engine that today runs at 33%
[..]
should give close to 38% efficiency
[..]
could give over 40%! A gain of just 9% to
[..]
36% would still make it worthwhile.
[..]
looses about 5%, it regains some
[..]
burns cooler than petrol (1925 vis 2030 deg C)

Where does your prolific abundance of figures and statistics come
from? Just out of interest, like.




--

Stewart Hargrave

Finally visible on www.hargrave.me.uk

I run on beans - laser beans


For email, replace 'SpamOnlyToHere' with my name