[Diy_efi] Estimation of spark advance

[Perry Harrington]

Carsten,

Spark advance is directly related to cylinder pressure.  There are severa=
l
ways that I can think of to get the optimum advance for your motor.

The simplest way is to adjust the parameters throughout the rpm range,
reduce advance when you ping at a specific rpm, and add advance where
no pinging is.  I 2 degrees retard off ping is what I've seen tuners run.

That's kind of a crude explanation, but the whole point of spark advance
is to achieve the highest stable cylinder pressure at a given RPM.  Usual=
ly
a dyno is used extensively for tuning.  However, if you have a method for
loading the engine while the bike is static, that should work as well.

The closed loop spark systems use knock sensors to retard spark, to keep
it on the hairy edge of the curve.  This yields the optimum cylinder
pressure.

There is also a relation to advance vs fuel requirement.  Many systems
add more spark at a steady cruise, to improve economy.  If you add too
much spark when lean, you get pinging.

You may want to look around for a DIY knock sensor setup.  You could
hook this up and use it for automatic peak spark, or just for the tuning
phase.

--Perry

On Sun, Apr 21, 2002 at 12:14:19AM +0200, Carsten Beth wrote:
> Hi all!
>=20
> Some time ago I wrote to the list that I build a EFI solution to replac=
e
> the mechanically controled system of my motorcycle (Honda GL1000, 1977).
> It's a microcontroler based system, which is triggered from hall sensor=
s
> and drives IGBTs. The main goal was to create a low cost,  more accurat=
e
> replacement without wearout and with the possibility to tune the spark
> advance.
>=20
> Now the EFI is running! I can start the engine and it is running quite
> well above 2000 rpm.
>=20
> The next step is to adjust spark advance. At the moment I use the
> mechanical systems original values, respective what I think, what it
> should be. The information I actually have, is that the advance should=20
> be 5=B0 up to 1000 rpm, 37=B0 above 3000 rpm and there sould be a "soft=
"
> transition between 1000 and 3000 rpm. That's not a lot, but maybe all a
> mechanical system can do.
>=20
> Now I think about how to make a good estimation in a theoretical way to
> have a quite well running engine over the whole range. Is it possible t=
o
> do so? In my opinion the main problem is the unknown time from
> triggering an ignition spark to the point where power is developed to=20
> the crankshaft. I just know the first component, from triggering to
> appearance of spark and I know the sparks duration. So the question is,
> how long does it takes from spark appearance to power development? Or i=
s
> there a better sulution for estimation the spark advance?
>=20
> Best regards
>=20
>     Carsten Beth
>=20
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> http://www.diy-efi.org/mailman/listinfo/diy_efi

--=20
Perry Harrington             Linux rules all OSes.               APSoft  =
    ()
perry at apsoft dot com 			                 Think Blue. /\

Those who would give up essential liberty to purchase a little temporary =
safety
deserve neither liberty or safety. Nor, are they likely to end up with ei=
ther.
                             -- Benjamin Franklin


_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi