Anti-Reversion Headers

General engine tech -- Drag Racing to Circle Track

Moderator: Team

NewbVetteGuy
Expert
Expert
Posts: 779
Joined: Sun Aug 07, 2016 4:11 pm
Location:

Re: Anti-Reversion Headers

Post by NewbVetteGuy »

digger wrote: Wed Oct 25, 2017 2:54 am i've seen tests where a bigger primary makes more power than a smaller primary that would be classified as being a "more correct" size, translating to if you have a header that is considered too large then maybe making it the smaller size in itself might not produce much benefit on the dyno.

So does this inherently mean that something about the combination is "off" that you need alot more primary pipe than what you might get away with on a fully developed engine? what have people seen as steps that they needed to also make to get the smaller pipe to work for them? of course there will be no single answer as it depends where you start from....
IMHO, there's at least two problems here:
1. How are you defining "correct" size? -This is a huge part of the issue. Are you sizing the initial pipe diameter based upon the area of the heads' exhaust port? The Valve? The HP level? --The different methods applied to different heads and engine combos will end up with different answers to "correct"; some of these old models are proving to be wrong, and some are proving to be right. I think basing the "correct" size upon the exhaust port area is probably destined to fail, unless you have a head with a "correct" sized exhaust port area to start with. -Should we instead base the correct header primary starting pipe diameter upon the valve or the HP or the view of both?!?? How SHOULD we define whether a head's exhaust port is "correctly sized?!?" (valve area to exhaust port area???) -I'd love a good answer to either of those questions, for sure.

2. Much of the old-school, stereotypical header primary diameter recommendations are based upon the assumption that you have a single pipe diameter (non-stepped design) for the entire length of the primary; many of the pros providing headers to pro racing teams are using stepped designs. Even the magazines have plenty of dyno evidence that to get max power potential out of an engine, you don't just start at the pipe diameter that you'd use if you were using a non-stepped design and then work your way up. As far as I can tell part of the benefit of going with a stepped design is that you can start off with a SMALLER initial pipe diameter at the head to improve velocity/blowdown length / scavenging AND maybe even gain a little bit of anti-reversion benefit for the first 7-12" of the primary pipes and THEN step up 1 or two more times and you get the torque and scavenging benefits of the smaller diameter pipe, and the HP benefits of the large pipe.

If I copied/pasted correctly, a direct quote from Calvin Elston on this subject: "If you hold a tube size for more than 10"-12", ie not stepping, you are giving up top end power". Note: This seems to line-up very well with the guidance from Larry way back at the beginning of this thread in terms of placement of the anti-reversion chamber in the primary pipes ("within 12 inches of the head"), and those devices also perform the step-up to the next diameter. Note2: Over email John Grudynski @ HyTech also stated that the anti-reversion step-up devices are most effective about 9" off of the head.

-Isn't it time more of this tech "trickled down" into affordable street headers? I certainly think so because maintaining low-to-mid range torque, a broad powerband, and decent fuel economy with a performance-oriented cam is so important in the street space. I'm honestly shocked and borderline appalled that this technology and know-how seems to be limited to the $4,000+ exhaust system racing team domain still. (Calvin started doing this stuff in the 90s and it still isn't available in the street perf bolt-on market?!?!?) I really hope if more consumers start asking for it, it'll finally trickle down... Or maybe it will take the modern models being democratized by being loaded into something like PipeMax v4 or Burns Stainless releasing X-design to enthusiasts to really drive awareness to the influential enthusiasts...???

Here's Calvin's guidance on "off-a-port tubing diameters and power levels supported" (the language of "off-a-port", seems to be intentional and hinting at stepped header design, and I'm of course assuming an "ideal" end-to-end designed header) -If 590 HP out of 1 5/8" tubing (outside diamter & off-the-port) is possible, the old-school guidance is just terribly outdated if we're talking about custom or DIY headers or even headers produced with modern understandings of the science, at all in the conversation:
  • 1.5″ OD 18ga tubing off a port can handle 430hp in broad powerband, (BP) and 475 in tight powerband, (TP)
    1.63″ tubing is good to 550 in BP and 590in TP.
    1.75″ tubing is good to 700 in BP and 780 in TP.
    1.88″ tubing is good to 875 in BP and 1000 in TP
    2.00″ tubing is good to 1000 in BP and 1200 in TP


The header primary pipe diameter discussion seems to always go back to a "tyranny of "OR" situation (either you can go with a smaller primary diameter and get more torque, but limit upper HP, OR you can go with a larger primary diameter and sacrifice lower-end torque and gain upper RPM HP), but it can definitely be an "opportunity of "AND"" situation (you can start with a smaller primary and keep low end torque while also supporting higher RPM HP capability with a properly designed stepped header), if we maintain the complexity in the conversation that comes from the real option of stepped header designs.

-From what I'm gathering from the likes of Elston Headers & Burns Stainless, is that roughly the same situation exists with Tri-Y header designs: although people have long been framing the "Tri-Y" vs. "4:1" trade-off as one of low end torque vs. high RPM HP, superior understanding of the science, allows a properly designed Tri-Y (for the application) to balance things like the collector size / choke and pipe lengths to make this an "opportunity of AND" situation, too.



Adam
NewbVetteGuy
Expert
Expert
Posts: 779
Joined: Sun Aug 07, 2016 4:11 pm
Location:

Re: Anti-Reversion Headers

Post by NewbVetteGuy »

<Begin Rant>
My wife's $18k, 2.0 Turbo 4 Cylinder Mazda CX5 comes with a tuned-length, tri-y, stepped header from the factory, but I can't buy the same thing off-the-shelf for my C3 Corvette with a Small Block Chevy engine and a custom header version is unobtanium for under $4k?!?? Something's really wrong with this picture...

Why don't enthusiasts have affordable access to technology that was developed while Three's Company and M.A.S.H. were the most popular shows on TV, especially considering that under $20k Japanese 4 cylinders come with it from the factory?
</End Rant>


Adam
User avatar
modok
Guru
Guru
Posts: 3314
Joined: Sun Jun 06, 2010 1:50 am
Location:

Re: Anti-Reversion Headers

Post by modok »

NewbVetteGuy wrote: Wed Oct 25, 2017 12:50 pm
1. How are you defining "correct" size? -This is a huge part of the issue.

Adam
i would not use the word correct, but more like normal, or classic. I would say the classic size is, average or mean primary ID Is same as valve OD,
in the case of a NA, sporty engine,with one exhaust valve per cylinder.

There could be MANY reasons why an engine would like larger.
If the only variable tested is primary SIZE then that only proves it's the best size for that setup. Doesn't mean the setups is the best tho, in fact it -could- be a clue that the engine doesn't LIKE that particular length or design.

There is a old rule of thumb about cam timing verses header design, that basically says if you have really short cam timing and more flow than you need, then any design of header will work fine as long as it's not restrictive. in some cases you are just fighting pumping losses on the up stroke.
The majority of the power you can gain or lose with a header is about the FIFTH stroke of the four stroke engine....overlap, but if there really isn't any overlap, or not enough to do good for the RPMS you are wanting to improve, then your not going to gain jack, and can lose more in pumping losses then could gain with good scavenging.
NewbVetteGuy
Expert
Expert
Posts: 779
Joined: Sun Aug 07, 2016 4:11 pm
Location:

Re: Anti-Reversion Headers

Post by NewbVetteGuy »

Thanks Glen, for that previous post. I think I'm more agreeable to sizing the primary based upon the exhaust valve size than the head's exhaust port size, for sure- the head's exhaust port size seems to be more of a variable than a constant that you'd want to base something else off of.


On a very related note to this thread; here's a SolidWorks flow simulation of a tri-y header; it's very interesting to see where and for how long the "blue" (stalled / backwards flowing?) air seems to primarily occur. Seems to lend further credence to the idea of anti-reversion steps close-ish to the head (7"-10" or even 12").


Anyone have any more Solidworks flow simulations of different header designs that they've come across? Super, super interesting to see.
https://www.youtube.com/watch?v=yl34m7x0M6w

Adam
Bazman
Pro
Pro
Posts: 332
Joined: Sun Aug 21, 2011 9:07 pm
Location:

Re: Anti-Reversion Headers

Post by Bazman »

Some good stuff in this thread. Has anyone done any research on anti-reversion exhausts for a turbo application?

As long as the AR chambers were shaped and sized to keep any velocity drop to a minimum I'd have thought there could be some useful gains in the lower rpm range just as in NA and that could aid spool times. With a lot of turbo motors having very modest amounts of overlap however perhaps there is nothing to be gained other than improved vacuum off boost??
Bazman
Pro
Pro
Posts: 332
Joined: Sun Aug 21, 2011 9:07 pm
Location:

Re: Anti-Reversion Headers

Post by Bazman »

Thinking further on this, for turbo engines there could be more downsides to upsides of running AR chambers along the header primaries, but maybe this simple idea may work without significantly reducing velocity????

Build shorty turbo headers with primary pipes based around the exhaust valve sizing (go up 1 size if in between available tubing sizing) then run 4 into 1 AR collectors which over a distance (say 15-30cm/6"-12") conically reduce to match the area of the turbo inlet (for twins). By running a square collector front plate and extending the primary pipes inside the collector by approx the distance equal to the primary pipe diameter you'd have in effect an AR chamber via all the space behind the pipe ends. Velocity loss could be controlled by ensuring the space was not too large.

I haven't got a decent drawing program but the collector would look something like this before being fabricated properly and blended to keep velocity up while providing adequate gaps for an effective AR chamber.

Image

Grumpy used the following design on some applications as a collector extension, and I wonder if it could work on turbos if the outlet end tapered down to the same smaller size (as the cone inside) and the gap above the "cone" was kept modest (like 1/4") to retain as much velocity as possible? Any thoughts?

Image
Jim2527
New Member
New Member
Posts: 29
Joined: Sun Apr 02, 2017 4:03 pm
Location:

Re: Anti-Reversion Headers

Post by Jim2527 »

NewbVetteGuy wrote: Wed Oct 25, 2017 1:59 pm <Begin Rant>
My wife's $18k, 2.0 Turbo 4 Cylinder Mazda CX5 comes with a tuned-length, tri-y, stepped header from the factory, but I can't buy the same thing off-the-shelf for my C3 Corvette with a Small Block Chevy engine and a custom header version is unobtanium for under $4k?!?? Something's really wrong with this picture...

Why don't enthusiasts have affordable access to technology that was developed while Three's Company and M.A.S.H. were the most popular shows on TV, especially considering that under $20k Japanese 4 cylinders come with it from the factory?
</End Rant>


Adam
Economies of scale. Supply/demand
Newold1
Guru
Guru
Posts: 1963
Joined: Fri Apr 08, 2016 9:50 am
Location:

Re: Anti-Reversion Headers

Post by Newold1 »

Just happened on this thread today and I can say that the anti-reversion exhaust header, manifold has a third direct benefit and use in marine (wet exhaust) racing and high performance engines that I have been involved in since about 1983.
In the marine racing engines we were building and running from those earlier days to today the need for increased power was accompanied with longer duration camshafts with increased overlap and the exhaust reversion was a tuning and performance robbing characteristic in our normally aspirated engines

Add water to that exhaust which these water cooled headers and manifolds required and the exhaust reversion became a major engine killer! Especially when we ran in saltwater offshore.

In all of the exhaust headers and manifolds we ran then and many still use today the we always used some form of a exhaust reversion device either at the header to head flange or upstream of the cooling water exit into the exhaust at the exhaust elbows or collectors. These engines run a various rpm ranges and quick throttle backs and changes in throttle position and engine rpms changes are a constant during almost all usage. This in turn amplified a lot of reversion and made a reversion blocking of the wet exhaust flow a mandatory feature. You should see what hot saltwater does back inside a marine engine reversion when the reversion is allowed to continue even for one race! We are talking valves, seats, guides, pistons, rings bearings and camshaft and lifter surfaces and components- They are Toast at that point!
The Older I Get, The Dumber I Get :wink:
Bazman
Pro
Pro
Posts: 332
Joined: Sun Aug 21, 2011 9:07 pm
Location:

Re: Anti-Reversion Headers

Post by Bazman »

Thanks for your post, that makes a lot of sense. I think any application that requires a motor to run up and down through the rpm range a lot with many throttle positions and some cam overlap (eg. hot street or road racing) then AR has a place, though the detail may differ on how it's done.

I'm keen to try it on my twin turbo set up but am trying to learn all I can so I can run a decent mid sized cam (e.g. increase from 2 to 14-20 degrees overlap @ 050) reduce reversion and improve vacuum/throttle response at one extreme, but do not hurt spool at the other. Most hot (efi) turbo cams are only 2-6 degrees overlap @ 050 for street apps through cam timing changes, even though they run decent duration. The turbo efi motors would likely run better with a bit more overlap if lower rpm reversion was reduced making them act like a smaller cam off the power band. Losing bottom end on a turbo efi street based motor is usually not a good plan, hence a lot of 8 second turbo cars on stockish cams but big boost. Would be good to run the motors up a bit yet retain bottom end and retain some fuel efficiency at cruise - this is where I think AR can help
Post Reply