<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>
Quote, originally posted by smakbiam »</TD></TR><TR><TD CLASS="quote">
How much did you end up softening the domes? Assume the Pistals were 13:1 before the softening, did you happen to measure the CR after wards? Was it a true rounded edge or a chamfer? I would imagine the lack of detonation has more to do with lowering the CR than changing the actual dome shape? Can you elaborate how dome shape may effect detonation outside of it changing the CR?
</TD></TR></TABLE>
as i see it:
if you have sharp edges they can get very hot and act as something like glow plugs, igniting the gas before the spark plug gets a chance, pre-ignition. taking the edges off and making stuff round, as jd says, can make a big difference. little changes can make a big difference with lots of comp - running too much squish distance can make engines more prone to pinging. reduce it, increasing the comp, and it may stop pinging.
the dome is a problem because the flame front has to travel around/over the dome to get to all the gas on the other side. which is further than going across a flat top. if it doesn't get there before the overall pressure and temp in the chamber reaches a point that the remaining unburnt gas will auto ignite, the unburnt gas explodes on its own. that's detonation.
again, more comp brings everything closer to being a problem, as pressures and temps are higher due to the compression alone. my sport 1100i guzzi had a 14mm dome on the std piston with 9.3:1 comp. and a huge hemi chamber, much more open than the ducati hemi 2v head.
if you have another plug on the other side of the dome it's no longer a problem. but that brings us back to timing requirements.
<TABLE WIDTH="90%" CELLSPACING=0 CELLPADDING=0 ALIGN=CENTER><TR><TD>
Quote, originally posted by smakbiam »</TD></TR><TR><TD CLASS="quote">
Brad interesting bit of info regarding the long/short intake runners effect on timing and the importance of short runners on hi-compression engines. Out of curiosity have you tried using flow matched long runners to negate the sub-5K pinging? You mentioned the short runners wont ping with a 4-6 degree retard, in your opinion would chopping an inch more off the shorts runners and having them flow balanced further decrease the need for retardation?
</TD></TR></TABLE>
flow matched long runners? i think you're reading a bit too much into it. all the manifolds are the same part anyway, you get two of the same thing in opposed fitted positions.
the long runners promote low rpm volumetric efficiency, leading to higher cylinder pressures, giving higher torque and therefore power in that rpm range. plus the carb bikes have shorter duration cams which have the same effect. so if you add hi comps and again increase the low rpm cylinder pressure you have an engine very high in low rpm cylinder pressure, and will most likely need to reduce the ignition advance.
this is where the problem starts, as the advance is not adjustable. the boxes have 26 degrees advance which comes in about 2,500 rpm as a single step. the std spec of 6 degrees advance at idle gives 32 degrees advance from 2,500 rpm upwards. when you really want 26 - 28ish. running 0 at idle leads to idle and starting problems. twin plugging to solve the detonation problem has the same reduced advance requirements pretty much, so you're back to the same shit pit. it's a specific application problem because of the ignition system.
the injected engines have short manifolds and longer duration cams. so they make less low rpm cylinder pressure comparitavely on both counts. meaning they would be less likely to ping because of that. secondly they have a digital tapering ignition advance curve. so they are less likely to ping on two points - cylinder pressure and spark advance. a short manifold, long cam carbed bike would exibit the same tendencies i'd expect.
if you ran the ignition of a carb bike using either a mappable ignition system of some sort - silent hetnick, supersport ignition (if he's still going, ask alex maybe) or the 1.5m ecu from an injected bike set up running some single pickup timing gears and a modified waterpump type alternator cover (no air cooled 2 phase single injector alt covers exist, nor dual ones either for that matter) you could map it with the fim software. if you added a manifold vacuum sensor you could use that to give a tps-esque input to use a map of sorts as opposed to a single 'rpm only' based curve if you had no tps type input.
or - not sure if this works or not, i haven't tried it yet - with your std 2v carb ignition, you could take your flywheel and reduce the length of the lump on it that triggers the pick ups from the leading edge. this, i am told, will reduce the added advance, as the boxes change from using the signal from the trailing edge of the lump at idle to the leading edge at full advance. using the 'off' portion of the square wave then the 'on' if that makes sense.
i'm not sure if this is true or not - i pulled the 600 apart to try it and it has a stamped/folded from plate flywheel, not the machined ones the 750 and 900 have. so i would've ended up with a hole in my flywheel that may have been a balance issue. now i've sold that and got a 750 engine for the 600m i'll try it at some point. if this does work, you could then keep the std 6 degrees at idle and reduce the full advance as required. it's still a bit of a bandaid solution, but more effective. if it works.
there's lots of good theory about all this stuff, but reality usually isn't so neat. the systems as chosen have limitations that you work around.