How do you tune an ECU without knock control

[Bernd Felsche]

Bruce tapped away at the keyboard with:
> > You always risk blowing up your engine. :-(
> > Acoustic method is better than none.

> Always at risk about being hit by lightning.   Simple precautions
> tend to min it happening.
> 
> Acoustic as implimented in the early systems (ie GM) is great, if
> you don't do any tuning, and want to run too low of octane fuel.
> Acoustic, in a serious engine is marginal protection for the
> inexperienced tuner, IMO.  Thou, the J+S unit maybe the exception.
> 
> Course you just said some detonation is to be tolerated, so I have
> no idea of what your standard is for *tolerable*.

I didn't say some was to be tolerated. 

I specifically stated condition under which light knock was
acceptable; i.e. when it's not severe enough to strip the protective
boundary layer from the surfaces. One also doesn't advance the
timing until light knock is encountered; timing is advanced until
the maximum torque is available at the crank; the advance is limited
by the onset of knock - if that should occur.

Exactly when that happens in a particular engine is still a matter
of trial and error; unless you've got a few million bucks to throw
at in-cylinder optics and high-speed photography. The future may see
indirect pressure measurement being used in production cars; e.g.
piezo washers under head bolts. These allow the magnitude of
the pressures to be determined (ref SAE 1999-01-054).

Alternative technologies such as ion-current sensing are still being
refined (Bosch: SAE 1999-01-0204) to improve knock detection by that
method. Ion-current detection requires a "static" ignition system,
with a coil per spark plug and favours a non-central plug location.

The Bosch paper also describes some criteria for knock sensing and
nominal boundaries for pressure and knock-index.

It should be noted however that pressure is not in itself an
indicator of the likelihood of damage due to knock. It's the gas
velocity that does the damage; stripping the boundary layer and
exposing the surface of the metal to excessive temperatures.
The magnitude of the higher-mode oscillations in-cylinder is
therefore more significant than the primary mode due to knock.

One can exploit the knowledge of the manufacturers in acoustic
detection (even if this is a DIY list :-)) and look at the
_conservative_ knock bounds (in terms of magnitude at particular
speeds and loads) as used in production engines upon which your
beast is based.

The bounds have to be conservative, especially on German cars
because these can be used continuously at maximum power, possibly
for hours on end on public roads. It's expensive to replace engines
under warranty.

-- 
Bernd Felsche - Innovative Reckoning, Perth, Western Australia
----------------------------------------------------------------------------
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