Water injection (was: Re: fuel cooling intake air)

[Axel Rietschin]

01bfea1c$0286b630$1201a8c0 at fotowire.com> <031701bfea22$35a533b0$900ac9d8 at nonea878d5l7jw>
Subject: Re: Water injection (was: Re: fuel cooling intake air)
Date: Mon, 17 Jul 2000 07:32:34 +0200
MIME-Version: 1.0
Content-Type: text/plain;
	charset="iso-8859-1"
Content-Transfer-Encoding: 7bit
X-Priority: 3
X-MSMail-Priority: Normal
X-Mailer: Microsoft Outlook Express 5.50.4133.2400
X-MimeOLE: Produced By Microsoft MimeOLE V5.50.4133.2400

From: "Bruce Plecan" <nacelp at bright.net>

> Well water does respnd to electrolyse, what is the ignition spark?.  So
> while how ever small, some most likely does seperate (into it's compnents
H

Interesting. But given that the spark is extremely short and extremely
localized, I'd not expect a lot of water to be separated, but you are right,
every bit counts. All I can say is that effect didn't noticeably show up on
the torque display, which had one (not very stable anyway) digit after the
decimal point :)

> > Well, more boost means more fuel, hopefully.
>
> But you excluded it in your first statement.

I implied it. The AFR is supposed to be maintained, some way or another,
with or without water.

> The ECU will see a lower air temp and add more fuel,
>
> So all ecu / ecm do this?, hmm wrong.

OK, maybe some ECUs don't take the air charge temperature, the most
important correction beside the base fuelling parameters, into account. Now,
would you use such an ECU to run a highly tuned race engine, if regulation
allows you to use a more sophisticated unit?

> > oxygen (your turbo is not going to swallow any more air because you
inject
> > some water after it, right?)
>
> No

No? Are you saying that injecting water in the air charge at some point in
the intake system, after the turbo, will cause the turbine wheel to
accelerate? Please explain.

> > However, in both types of engines, there is a lot to be gained if a
normal
> > combustion can be maintained with more boost or compression ratio.
>
> If the timing and fuel are correctly adjusted.  (Timing as used here is
for
> ignition (SI)or fuel injection timing on a diesel)

Thank you. However, under some circumstances, detonation starts _before_ you
have a chance to advance the (SI engine's) spark to the point where that
engine is the most efficient. You just can't reach maximum brake torque.
This is obvious on the dyno when you _feel_ there is more (and you actually
gain something by advancing the spark a bit more) but you start hearing some
random detonation, for example one erratic 'shot' every two-three seconds.
You add one more degree, you watch the torque output raise a few Nm's, and
the engine starts detonate more frequently, say one shot every 0.5 - 0.8 sec
or so (it's very erratic, anyway). You can't get to the maximum torque and
you even back off two or three degrees for safety. A little water helps a
lot here. How much water? Just what's needed to allow you to reach MBT
without suffering detonation at that specific load/speed point.

> As soon
> > as you get end gases detonation or, worse, runaway detonation (also
> referred
> > to as preignition) the game is over.
>
> Runaway?, how about uncontrolled?.  Since a diesel engine uses a form or
> controlled detonation.

Heywood describes it as runaway detonation (see, I can read :) Definitely
uncontrolled indeed.

> Also,
> > you suggested cooling the IC with NOS, a substance definitely not
allowed
> in
> > FIA rallying, water however, is allowed there.
>
> They let them drop water on the street, yet gaseous cooling is illegal?.
> Some set of rules, IMHO.

As you wrote, they make the rules. That said, the water is sprayed toward
the IC using several nozzles, in a fine mist. The spray bar is activated
only a second or two at a time, only when the ECU decides it's too hot
(there is various schemes). Rally cars tend to have some protection under
the engine, front drivetrain, intercooler and radiators thus very little
water, if any, will ever reach the ground. Also, 2/3 of the championship (or
so) are made on gravel or snow.

>> It is reasonable to think that if you inject close to the valve, less
water
>> will evaporate outside of the engine, thus less oxygen will be displaced.
In
>> contrast, injecting before the IC obviously gives much more time for the
>> water to turn to vapor and thus displace more oxygen.

> But, you said **any** Oxygen displacement was bad.

I feel like you want me to say there will always be some evaporation outside
the engine. Sure, there will and it is bad. Thus the attempt to minimize the
'damages' by injecting water as late as practically possible. IMHO, the best
thing would be to use a detonation-resistant fuel, raise boost up to the
point of diminishing returns and get rid of WI altogether.

In the future, I'm thinking to experiment further with water injection using
a dedicated sequential ECU, in phase with the engine cycles, using regular
injectors mounted close to the port instead of a modulated but continuous
flow with nozzles in the intake pipes. This will hopefully allow me to
precisely meter the flow and inject at the right time in the intake cycle,
inlet valve open (unlike the fuel, but this is another topic).

Axel



----------------------------------------------------------------------------
To unsubscribe from diy_efi, send "unsubscribe diy_efi" (without the quotes)
in the body of a message (not the subject) to majordomo at lists.diy-efi.org