Multispark stuff and other thoughts
Clinton L. Corbin : Backgrind/Gold : Pager 0544
CCORBIN at INTEL7.intel.com
Fri Mar 29 03:12:56 GMT 1996
>Several OEMs are doing this or designing engines that do this, namely,
> one coil per spark plug. Where's the beef?
>
>1) No plug wires. Like someone correctly guessed in their post, if you
> have no plug wires, you don't have to contend with RFI and losses
>across the wires. You still have to contend with RFI and losses within
>the coil itself, but that's an existing problem. You also eliminate
>one connection when you eliminate the plug wires. That's one less
>thing to breakdown and leave you stranded. Sure, you just added a
>bunch of 12 volt wires to the each coil, but it's much easier to deal
> with 12V than 12kV-45kV.
I agree that this can only be done with one coil/cylinder
>2) Spark energy. Firing once every two revs means you can charge
>longer and deliver more energy per spark than firing three or four
>times each rev.
What is the saturation time of the coil? If you are charging at a fast
rate, I still see no advantage with one coil/cylinder over one coil/two
cylinders.
>3) Multistrike/multiple spark. You can extend the multistrike rpm
>range all the way to redline instead of have to go single strike at
>3000-4000rpm.
Even with one coil/two cylinders you stil have four times (on a 8 cylinder)
the number of coils than the MSD has. Just how high do you need to take
the motor anyway?
>Now it gets really interesting...
>
>4) Combustion sensing. You can read the resistance across the gap
>after firing the plug and determine if you misfired or if it the flame
>went out. You can then strike the plug again until you are satisfied
>that you've burnt everything! Jacob's Electronics claim they can do
>this with one coil firing many cylinders, but imagine how much better
>you can do it if you have an independant circuit for each cylinder.
>This is literally on demand multistrike capability and can extend your
> lean limits and/or reduce feedgas emissions.
Once more, with the wasted spark system, you could do all of this with out
a problem. The only exception that I have heard of would be a engine with
extremely long intake and exhaust duration. Not a problem on just about any
engine going on the street.
>5) Independant cylinder knock control. With two knock sensors, you can
>determine which cylinder is knocking and alter that cylinder's spark
>advance without screwing with the remaining cylinders' spark advance.
Any again, why does one coil/cylinder do this any better than the wasted
spark system?
>So, the beef is, with coil per plug, you are one step closer to
>controlling several one cylinder engines instead of controlling a V6
>or V8 or I4. There are tradeoffs, but if you really need some of these
>things(especially extending the lean limits and reducing emissions),
>you will pay for them.
Except for the first item, almost all of this was the advantages one coil/
cylinder has over a standard distributor system, not over the wasted spark
system. I am also heavily into car audio, so getting rid of one source of
RFI is tempting though...
Clint
ccorbin at intel7.intel.com
More information about the Diy_efi
mailing list