This is the problem with messing around with scavenging on crossplane v8's its not as easy as the flatplanes.
People have spent a **** load of time making headers over the years and using all the science only for someone to come along and dyno a ****load of combinations or drag/roadrace them including odd firing order unequal length arrangements that make no sense (well probably some) and they end up making more power. People stumble onto things that work is usually what happens. Because the actual science involved to work this **** out is pretty hardcore.
The e55 is easy to get big gains from any combination of header as practically anything is going to be better than the stock log manifolds. Its cool to see someone like alex attempting to make something really make some power. If you look at the m3 or audi rs4 guys the stock factory headers are so decent that aftermarket stuff hardly makes any gains at all usually.
It's been busy two months but I'm back at it now. Got the engine out today, bolted to the stand, ready to start mocking up the new headers on Monday.
There will be a couple of changes to the original idea.
1. I'm going with Titanium back boxes, middle silencers and main pipes. Main reason is weight saving. I've not welded Titanium yet so it will be a first. It turns out it's cheaper (and quicker) to buy ready made Vibrant Titanium silencers than to buy material and make my own.I know I've said I'd do everything myself but I'm behind (I want to race in May) and I need to get a lot done.
2. I'll also try to use 8 wideband Lambda sensors instead of EGT sensors. Basically individual cylinder tuning all the way. So if previously one bank was running hotter and we just dumped fuel into all 4 cylinders, I can see which cylinder is running lean and address that. This should improve the power for the engine quite a bit.
3. Lastly, as I'm evening out the exhaust pulses I will not actually need an X-pipe, which is used to balance out the pressure and improve scavenging. With evenly spaced exhaust pulses it is no longer required.
Are you just doing the 3" sections in ti or the whole thing? Where did you source the ti?
Headers and downpipes in 304, exhaust tubing, mid silencers and back boxes - Ti. Someone was selling a box of unused Ti 3" bends and some off cuts locally and cheap (about £50 per bend). So I jumped on that deal as I'd never be able to justify it otherwise!
Time to make this thread alive again. Had to squeeze a couple of other exhaust jobs which pushed the work on my car back a little but I've now cleared the schedule and can dedicate all my time to it. No more long delays!
First things first, I've drawn a diagram of our firing order to help illustrate what options M113 owners have when making firing order specific headers:
The crankshaft rotates clockwise and each 90 degrees of that rotation a numbered cylinder fires (starting with 1).
From the diagram it's easy to see pairs of cylinders, firing 180 degrees apart. These would be pairs 1-4, 5-2, 6-7 and 3-8. 180 degrees headers are the most popular type as it makes easier to pair cylinders on the same bank (1&4 and 6&7), effectively only leaving us with two primaries that would need to cross underneath the engine/gearbox (3 and 2 or 5 and 8). Unfortunately for us, this doesn't stop there. As the primaries grow a bit long and reach the optimum 30" for this engine's 6500 RPM, the secondaries have to be as short as possible (without losing some low and mid range torque). That means that pairs 1-4 and 6-7 have to be as close to each other as possible. Basically throwing out of the window the whole point of pairing them on their respective sides.
Which bring us to the next arrangement: 360 degree headers. As the name suggests, we pair cylinders firing 360 degrees apart: 1-6, 4-7, 5-3 and 2-8. Then we pair secondaries (dotted boxes): 1-6 & 4-7 and 5-3 & 8-2, leaving us with two 3" downpipes ready to connect to the rest of the exhaust. 360 degree headers are much more difficult to make as one have to cross 4 primary runners, not just 2 like the 180 degree type. And as if it's not hard enough, we have to equalise all runners within an inch of each other.
Sounds pretty challenging. 360 degree headers it is. Let's go.
Step No. 1 is to mock up a design. I rely on Icengineworks 1-3/4" modular set for my exhaust work. Since we're going with 1-7/8" primaries, the design is not going to be 1:1 accurate but still good enough to give a quick idea on all do's and don'ts. After a bit of playing around with this 'Lego' system, here's the runners 1 and 6 (being the farthest from the bulkhead and hence setting the max length):
The primaries ended up being 31" long and meet up close enough for a collector. Again, this is only a view so the length may wind up being a little less (or more) when done in metal, which will be reflected in all other runners as well.
Here's one bank in plastic:
The design allows for all 8 lambda sensors to be installed 8-10" from exhaust ports which is another goal of this job.
The next step is to make metal sections according to the bends in plastic and build the actual primaries. This is a tricky part because the CLR (centreline radius) of the plastic kit is not the same as 1.5D bends available for me to buy (the plastic kit has tighter radius) but I should be able to make it close enough.
Very sorry for the lack of updates. Didn't forget about this thread, just been inundated with some side work/cars that took all of my spare time. Again.
Checking in as restarting very soon. Some changes due to a different platform this engine is going in to - 209 now instead of the 211 - but essentially it will all stay the same, i.e. firing order specific headers albeit out of a more exotic material due to turbo heat.
Slideshow: A one-of-one U.S.-spec Mercedes-Benz SLR McLaren Roadster became even rarer after a factory-backed transformation at McLaren's headquarters.
