Torque cones on a 2-1 exhaust system....98' M900

[brad black]

one thing that you'll notice about torbjorn's system is that the header pipes have an expansion before they hit the collector.  this will give a definitive end to the header.  not sure if that's important or not, never tried it.

[MonsterHPD]

Well, maybe some explanation could be of general interest.
The design was based on two requirements:1) Equal-length headers, and 2) 2-1 system.

Since I had some problems to join the equal-length headers at the right (...?) header length, I reasoned I terminate the effective header length by providing a pronounced diameter increase, and then join the resulting "chimneys" at some convenient spot. Essentially, a pair of dragpipes joined into a common silencer.

The idea was that the big-diameter part of the headers and the collector should just funnel exhaust and noice thru the silencer, having no basic function other than leading the exhaust away and provide silencing.

I don't know if I came up with the idea myself, or if I read about it; I think the 999 exhaust is made basically the same way: Short rear header directly into the silencer, long front header with a pronounced diameter increase along the way, presumably at about the same length as the rear header length.

A few years ago, I had the bike on the dyno to map the Nemesis ECU with some tuning bits. To check different header lengts I made header extensions to a total of 1000 mm for both headers, and made runs cutting pieces away after each run. This was the result:



As expected, longer headers gave more torque at lower revs, and less power at higher revs.
What I did not expect was the result compared to the graph with the complete system: less torque at lower rpm, but more power at higher rpm (that graph is the final, mapped graph). So, clearly, there's something to be had by development work here; if one could combine the best parts into an envelope the result would be quite nice.

As for how to join the headers into the collector, this graph could be of interest:



The M900 curves are from the 900 with no tuning except air box lid and PCIII and the 2-1 shown previously; the other three are from the rain bike with a "normal" Y joining piece and collector but no silencer (it was broken at the time). Pretty similiar in shape and caracter, and probably a bit of performance to be found with a lot of work, but maybe not worth it. It seems that the Y joint and the "chimney joint" work about the same, and the Y joint certainly looks better (I think).        
     

[118811]

Well...I wish I had the time & resources to play with the system and try to maximize it's potential.
But this a daily - commuter bike for me so its just going to get bolted up and we'll see!!

[MonsterHPD]

Well.... so do I   nothing much has happened beyond this mainly academic exercise for 3 years.
Learned some, got confused some, but not quite sure what I learned.

Let's see what these megaphones might do to my ST2-based track bike, to to be finished any year soon