Alex L

Given that I'm about to start making a new exhaust system for my E55, I thought a few of you might be interested in seeing how this will be done. I will be sharing my ideas and work from start to finish. This is no botch, "git 'er done" type job. It will take a few good weeks but will be done with quality of fit and finish in mind. This is going to be an upgrade to my existing exhaust (also a DIY) which is: 1.75" dia, 22.5" equal length primaries => 4-1 merge collectors => X-pipe => two straight through silencers. 2.75" overall dia tubing (factory size)

The new exhaust will be as follows:1.88" dia (port matched) => stepped to 2" primaries half way down (space dependent) => 4-1 merge collectors OR Tri-Y system (space dependent) => X-pipe => two straight through post X-pipe silencers => two rear silencers. 3" overall tubing size

There's a few reasons why I want to redo the whole system:

- My headers were my first ever weld & fab project. Since then I've done a few complete systems and have better ideas as for the implementation and outcome. I've learned a lot more too which will benefit this job.
- I've reached the limit with my 630cc injectors at only 65% throttle (physical butterfly opening, not pedal travel). In order to move forward I bought a set of 1300cc injectors which will require a whole new map done from scratch (I use standalone ECU). I want to tie up the new and improved exhaust with new injectors which will save me doing this job later and adjusting the map.
- I currently monitor EGT as per each bank. I want to monitor EGT on every cylinder.

Essentially it's making a better and less restrictive system for the power level I'm aiming at with this car. My last dyno was 600rwhp. Whatever another 35% of butterfly travel add, I want to have the system that will be capable of supporting the top end which I need for the Vmax200.

Before we can get on with fabrication, there's a few design features that need straightening up:

- Headers must be of equal length design. Equal length offers the best scavenging across all cylinders. This means minimum to no gases left in the combustion chamber after exhaust stroke which in turns means more oxygen on intake stroke and ultimately more power.
- Only mandrel bent swept tubing being used. Less turbulence, more laminar flow for our exhaust gases.
- Equal diameter catback tubing. Exhaust gas velocity drops when tubing diameter increases. On the other hand, stepped up design of primaries helps with scavenging. The further the step is from the headers though, the lesser the effect. I want to try and step the primaries up, once and to 2" each. It may not be easy due to space issues. We shall see.
- All tubing is SS304, 16g & 18g. All must and will be back purged using argon (headers, collectors, x-pipe) or flux (rest of the exhaust).

I will be making all the parts myself apart from material and small hardware. I've never done collectors or x-pipes. I'll start with the latter as it appears 'easier'. I don't expect the whole project to take over 8 weeks due to other commitments but then again we shall see. This is not my daily job. I'm a weekend warrior when it comes to weld & fab or any car related stuff for that matter.

Lastly, feel free to share your thoughts on the above. I will be adjusting the plan as I move along. It's not possible to just imagine something in your head and make it fit the car. There will be compromises but I will try to keep them to a minimum.

Thanks for reading the intro. I've got one last car to finish before I start on the E55. Yes, there will be plenty of pictures of the process and outcome. Stay tuned.

tlambert

tw2

Thanks for the awesome info. I am looking forward to seeing some tri-Y's on an E55. Great stuff.

JoeJErnst

Awesome! Can't wait to follow along.

1ForcedBenz

Here is one build very similar https://mbworld.org/forums/w211-amg/337272-diy-long-tube-headers.html for insights sake.

I wish I could find a thread from back around 05 when a guy pulled the engine and had flawlessly built headers custom made. In sure some one with enough diligence to search will find it. Those were truly a bench mark example.

austingtir

Heres mine: For RHD guys in NZ, Australia, UK etc etc I will be making another set in the new year so if there is much interest I'll keep making them. Pictured is my prototype set. So got most clearance/design issues worked out. If I sell any they wont be slip fit (unless requested) and will have the pictured vibrant flexi joint installed instead.

Alex L

Engine in mine had to come out a thousand times when I built my set. Won't be able to fit the headers without pulling it out and it made it much easier to mock them up and fabricate with the engine on the stand. Would be great if someone could find that thread. I don't recall seeing it before.

Some beautiful work Austin! Tri-Y and all. I hope you don't mind a few questions. Size of primaries you went with and the secondaries (past the first Y and into second). Also is this a 3" system or 2.75". Thanks!

austingtir

I made this prototype without pulling the engine. Its doable but does take a lot of time either way.
In the states you guys have easy access to finer sizes of tube than us here in NZ. I maybe would try 1.875" on the primaries and 2.25" on the secondaries but I really cant see it making a massive difference on this engine without cyl head porting and so on. I might try 2.25" secondaries on this prototype set sometime when I have time.
Mine are 1.75" primary 2" secondary 3" to midsection then factory 2.8" from there) I guess doing the rear section thats the next thing BUT im actually very happy with how it sounds and runs I modified the rear mufflers.... If I build past resonators I will make everything myself including the new 3" dual in out resonator and mufflers/tips at the rear just need time. Im still working fulltime for a restomod shop till the end of the year so not enough time for that atm.

Heres an idea of what it sounds like. If people PM me I'll show you a good high rpm burnout. Im not posting that on a public forum due to laws in my country. IMO no X pipe sounds better at higher rpm whereas X pipe sounds better at low rpm. Im happy with how it sounds at low rpm so thats why I have not fitted one and the stock resonator has an H section anyway. The vid doesnt do it justice once its revved out its totally different to this vid it sounds bloody good.

Alex L

new55

what are the white tube structure between the mid pipes ? cant figure it out ?
cheers

tw2

Looks like body bracing.

austingtir

No I just focused on firing order and step length more so than actual equal length as you have found if you make these equal length Id struggle to see someone coming up with a tri y design that could be fitted with the engine in 4 into 1 yeah probably be easy but not tri y. If it was a e60 m5 with the v10 I would certainly of being trying alot harder just for the sound.

These prototypes I can fit in like 45minutes pretty easily. Engine in, didnt raise engine off mounts or disconnect steering.

Drivers side steering clearance in the major pita when making these things.

Alex L

Got the car on the ramp now so the work has officially started. First thing I wanted to do was to measure the back pressure in the exhaust. This is purely for reference as the exhaust is being replaced with a bigger system, which I will also measure. Anyway, the needle fluctuated around zero which is to be expected as I don't have cats and my silencers are a straight pipe design not a chamber like the OEM ones.




I've also had a closer look at my current setup and thought what if it was possible to make a true 180 degree, firing order specific headers? I think with no valve body hanging below the gearbox I can just about pull it through. For those who don't know, 180 degree headers refer to sorting out one of the V8 design problems which comes down to this: two cylinders in each bank fire consecutively which is not ideal as pulses are not spaced out evenly. The M113K firing order is 1-5-4-2-6-3-7-8, meaning that cyls 4 & 2 and 7 & 8 fire in sequence. A 180 degree version of such firing order will look like this:

Collector 1: 1-4-6-7
Collector 2: 5-2-3-8

If you check the firing order against the new arrangement above you will see that it will be pretty much perfect, i.e. all exhaust gas pulses enter the collector at evenly spaced intervals.

Obviously this would mean that runners 2 & 3 and 6 & 7 have to cross under the engine to go into the opposite collectors. This is what I want to try and do.

So tomorrow I'm going to start knocking up an engine/gearbox stand which I can wheel around and do most of the header work on. Then it's engine out job.

austingtir

Awesome cant wait to see this. Have you dynoed your current setup and is there a thread somewhere on this car of yours? I only just noticed the 6 speed conversion! Your car sounds amazing!

Have you got pics of the current headers?

trabots

I am always interested in the "equal length" primaries are best but no indication that the optimum length of the primaries, length of collectors has been determined. For real scavenging to occur the overall firing order should be followed, not just for each side. That means crossing over primaries which is impossible for most street cars.

austingtir

Its difficult when you have to live within the constraints that the mercedes engineers have allowed for space particularly above the engine mount area and ground clearance on these cars. If it was not for that engine mount and the trans being in the way it would make it alot more feasible to make stuff like what alex is doing here. Im personally a fan of the way renntech did their slr headers and in a perfect world mine would be equal length but when you add in the fact most people are not going to want to pull an engine to install a set of headers (I certainly wouldnt to gain a handful of hp) you've just got to be realistic.

On the other hand you can be like Alex and go all or nothing. That is the mercedes motto I believe.

Alex L

Thanks. I'm at the limit with my injectors (630cc). Once I've done the exhaust, I'll take the car to the dyno place, put bigger injectors in (1300cc), redo the file and see where we're at. Currently it's only pushing 600rwhp and 600rwtq.
Here are the headers. They took a while to get made mainly as I was learning welding and fabrication along the way.






I agree. Most exhaust companies make headers that fit the engine bay and can be installed in a couple of hours. This obviously leads to many compromises on size and length of the runners. I think if it was easy on this platform, a lot more development would take place. Good thing I'm not a shop driven by profits or meeting targets and time is on my side.

Haha, I can't get any more out of the engine with forced induction so more finesse has to take place now. I don't have any expectations nor do I believe the new headers / exhaust will make a massive difference so with this approach it's kind of try and see what works and how. I have my current dyno output, I will dyno the car with the new exhaust as well. It's a bit more pipe to cut and weld but it's not a mega outlay.

Made some modest progress on the stand last night. More to be done today:

austingtir

Thats some impressive work alright.... im kinda in disbelief you are going to redo those. What you've done looks really nice. Im not a fan of the tight radius bends as I just used some in the last set of headers I made and didnt really like them but you got to do what you got to do.

Are you going to use the burns style collector again and how did you find that one? Iv got to make a decision soon on weather to make my own 4 into 1 collectors or just buy them off burns.... Burns is pricey but seems to be fair for what your getting. Having never held one in my hand they look good enough in the pics thats for sure.

Alex L

Thank you. There's a few things I can do better now. As you said, tight radius bends are not ideal. With collectors in their new place (to each side of the gearbox instead of downpipes) I should be able to use 1.5-2D bends instead of 1D and a lot more of straight pipe as the length of runners will be longer now due to crossing over.

I'm looking at making my own collectors to keep it all within the budget. Burns parts are outstanding, very symmetrical and uniform. I'd be all over them as it would save the time. For the money they are excellent given the amount of work that they put into making them. Vibrant is another manufacturer that a lot of people use and they are cheaper than Burns but I've got no first hand experience with any of their products.

StarvingArtist

Nice work, especially for learning while you go. Wondering if the bends might change the speed of the exhaust gasses to make your length off. I don't know if it would speed it up from the slingshot effect or slow it down from resistance. Maybe you have already calculated for that. Definitely looking forward to seeing your next run at it.

What boost pressure are you running that you can't get any more from?

Alex L

Thanks. The bends don't help. The gases have to stretch and compact as they make their way in the tubing. Even though there was no back pressure, I will try to stay away from tight bends in my new system.

The highest boost I've seen so far was 21 psi. When I say I cannot get any more power out of it, I mean with the pulley that doesn't overspin the supercharger (or rather doesn't exceed the rpm load on the bearings).

Alex L

I did a bit more "*** packet" math on the headers design. Technically, there's not 1 but 3 different version of 180 degree headers possible. And in a Tri-Y combination.

Given that in a V8 for optimum balance every cylinder fires at 90 degrees [720 degrees (=2 revolutions) / 8 cylinders] and our firing order - in case you forgot - is 1-5-4-2-6-3-7-8, cylinders
1+6, 5+3, 4+7 and 8+2 fire 360 degrees apart. Combining these pairs into Y's would ensure perfect scavenging for this engine as every gas pulse would enter and leave the collector without 'clashing' with another one.

So 3 configurations for the headers will look like this:

No.1 (runners 2 & 3 crossing over to the left bank, 6 & 7 to the right bank)
Right bank: 1+6, 4+7
Left bank: 2+8, 5+3

No.2 (runners 5 & 8 crossing over to the right bank, 1 & 4 to the left bank)
Right bank: 2+8, 3+5
Left bank: 1+6, 4+7

Basically, the second config is a mirror image of No.1: we're crossing over the 4 cylinders that remained at their "home" side in No.1.

Lastly, config 3 is a bit of a compromise. We're only crossing 2 cylinders - 4 & 8 - due to their close proximity to the collectors. This will allow for easier equalisation.

Right bank: 2+8, 1+3
Left bank: 4+7, 5+6

Why this is a compromise is because 1&3 do not fire true 360 degrees apart. They fire 450 (1-5-4-2-6-3) and 270 (3-7-8-1) degrees apart. Same happens with 5&6: 270 from 5 to 6 and 450 degrees from 6 to 5 again.
Such config however provides the easiest way of making the headers as only 2 runners will go under the engine.

Hopefully, as I make progress with the actual design it will become clearer which of the 3 configs will be the most suitable for the engine bay / chassis.

In the meantime, I've welded and painted the future engine stand and started making the uprights to support the engine:

cnterline

Fountain35

I see why you had to have the engine out, a lot of work. I am really interested in the dyno numbers after they are done.

Great work Alex!

austingtir

Yeah infact being in welding for 20 years everything from structural steel or jobshop alloy stainless and restomod shop I have never seen anyone pull anything like this off to this degree on a first attempt. Usually there are signs of shortcuts or dog**** lookin welds somewhere. Alex has done exceptionally well on his first attempt.