[Madact]

Yikes, this thread really took off

Originally Posted by freethink

There are tons of cheap D-series tubular headers available on ebay that will fit the VX. You might need a DX cat and midpipe to make it fit though.

Good suggestion, unfortunately it's illegal in many, many places to change the cat position. Luckily (a) I have a manual, Australian model with the cat under the car near the transmission and (b) the law here only prevents moving the position or affecting the operation of the cat, neither of which will be a problem (I'll be leaving the cat right where it is - which is why I was looking at rams horns - and quick warmup should be assured using stainless and/or coated primaries)

Originally Posted by ultimx

A main point to also think about is a good collector design. Burns stainless which I believe Neal posted comes to mind since u like tigging.

Originally Posted by Nigel_S

The angle the pipes join in the Y pieces is rpm dependent with Formula 1 exhausts having an unbelievably high angle, so for low rpm with long exhaust pulses you will want a very shallow angle.

Yep, I've been doing a bit of research there, and I'm thinking around 10 to 15 degrees (centrline angle, each side), and also having the incoming tubes take on a slight 'D' shaped profile on the approach rather than just being cut and welded, producing both a slight venturi effect (the cross sectional area decreases when you flatten one side) and a cleaner interface where the pipes meet.

Originally Posted by ultimx

Also for tuning purposes, you might want to convert to obd1 as there are many tuning solutons out there, one cheap one being chrome. Or hondata for better option. Also, I did see you mensioned using a adjustable fuel pressure regulator on your obd2 engine, you should try and get as close to 50-55psi on it as possible as Honda injectors atomize fuel the best at that pressure. That should be set if going back to obd1 and you can adjust the fuel tables to what you want in open loop as well as closed. Just fyi.

I think I'll be running the stock ECU etc. for regulatory purposes in the end, unfortunately. I'm hoping worst case, it runs a little lean in open-loop mode on account of the displacement - I think it should be fine in closed loop - though it may even have some learning feedback from the closed loop into the open-loop to compensate for various things, I'll just have to see how it goes at that stage.

Originally Posted by Nigel_S

Evolution is more concerned with preventing blood clots than making efficient use of the inertial energy of gasses flowing at mach 0.5...

Originally Posted by Old Tele man

...true, but headers designed for *economy* certainly are NOT operating with gasses at mach 0.5 either.

Originally Posted by adam728

Nor are they flowing thru flexible tubes, an incompressable fluid, etc.

No offence to Old Tele man (it's interesting stuff all the same), but I'm going to have to agree with Nigel and Adam here, on a few points:
1) The design criteria for an artery branch is very, very different from the design criteria for a collector merge.
2) The quoted formula seems to give results in the 50-70 degree range for the branch, which looks really wrong...
3) Flexible walls may make a difference, and while for most aerodynamics problems people on this forum are interested in we can assume incompressible flow (like blood, for instance... whoa, that sounds creepy), that's not the case for exhaust gasses.
4) All I've read so far indicates that approximately mach 0.3 (average) is a good target velocity for scavenging - much higher, Poiseuille’s law breaks down completely and transsonic effects & flow resistance start to choke the engine, much lower and you lose the inertial effects.
5) Unless the Reynolds number is similar, you can't use the same optimisation results even if the objectives are the same. Blood vessels are (from a quick web search) in the region of roughly Re = 2000, exhaust flow is (by some quick calculations) more like Re = 50,000 to 100,000. This puts blood flow near the upper limit for laminar flow (makes sense, as this is more efficient), while exhaust systems are very firmly in the turbulent regime, even if they weren't disturbed by valves, gaskets, bends and merges on their way out. The large difference means that a lot of the 'rules' change, and optimal geometries certainly go out the window when you have this kind of change of regime.