Anti-Reversion Headers

General engine tech -- Drag Racing to Circle Track

Moderator: Team

Post Reply
David Vizard
Guru
Guru
Posts: 1161
Joined: Thu Jan 15, 2015 12:19 pm

Re: Anti-Reversion Headers

Post by David Vizard » Thu Oct 19, 2017 4:15 pm

The first move toward determining if an AR ex. system will do anything is to get on the flow bench and establish what sort of velocity gradient is occurring across the port at the manifold face.

I remember when Fueling was all excited about testing on a Cosworth DFV F1 engine. I told him before hand it would fail to show a positive difference anywhere in the rev range because the velocity gradient across a DFV exhaust was very low. Tests showed no before and after differences.

DV
David Vizard Small Group Performance Seminars - held about every 4 weeks. My shop or your shop. Contact seminar organizer at marvingvx@gmail.com for details.

User avatar
modok
Guru
Guru
Posts: 1307
Joined: Sun Jun 06, 2010 1:50 am

Re: Anti-Reversion Headers

Post by modok » Thu Oct 19, 2017 8:51 pm

first, to be clear the topic of discussion is the "AR chamber", which IS a tube which exits inside a larger tube, as illustrated earlier. There is some variation on the size difference, or there might be a slight taper to the inner tube so the exit is a hair smaller, but that's the idea, general outline.
The use of a sharp edge tube which exists into a large volume, OR small details of the port/valve which promote flow in one direction more than another....are separate topics IMO.

The "device" is a concept that is fairly easy to understand, but how it actually functions is not as clear. The actual difference in flow forward/reverse in steady state testing is actually very small. The difference in reflecting pressure waves forward/reverse is also very small.
To "put in" an AR chamber....you MUST also increase the volume at that point. The added volume is what is making a difference just as much or MORE-so than any special flow characteristics of that shape.

If you are thinking about AR chambers you are thinking about stepped headers, and tapered pipes. Header pipes with steps or a header pipe that is tapered actually function in a very similar way. It's the overall volume/length/size that makes a difference. It is true that steps will give a return wave SLIGHTLY different than a gradual taper but it is NOT a large difference. If the step is a few inches one way or another....still operates largely the same.
As compared to a straight pipe, Tapered/stepped pipes tend to be far more PEAKY. You can even hear it. trumpet VS flute.
A "reverse stepped" or reverse tapered header is weaker but a lot less peaky than a straight one.
You may have never considered the possibility of a reverse stepped header....but they can exist, and they do exist. I bet a lot of folks who DO report large gains with AR chambers would have similar results with a "reverse stepped header".....but never tried one. People will not build or try a reverse stepped header because it's theoretically bass-acquards, and I'm not denying it, I agree! :lol: BUT, if you try a stepped header, and it's WORSE, then the logical thing to try next is the opposite....right? sure. OR try different lengths, valve timing or change the port or valve, really. it's all a system. But what we all want is a free lunch. CAN the AR chamber be part of a balanced breakfast? Maybe.

We know the PORT ends up being smaller than the valve, say, 86% give or take, and the far end of the header pipe ends up being as big as the valve or sometimes larger, so WHERE and HOW exactly does the size step up from port size to header size? In some cases it works great to stay small (port size) 8-12" inches out of the head, because... you can. IF you do that....it's kind of like a stepped header, it's more peaky. If that's great, then great. If that's a bad thing, put an AR chamber in there where it steps up...then you add back the volume you took out, and it should resonate more like a straight pipe did.....and there might be a net gain. That's where there MAY be the free lunch.

So TO ME, the AR chamber does not work as imagined, BUT, it is still the best way to add volume to the engine side of the header pipe with the least downsides.. Think about it. If you can think of a better way.... please share. :D
Glen Urban

User avatar
MadBill
Guru
Guru
Posts: 12624
Joined: Tue Nov 15, 2005 10:41 am
Location: The Great White North

Re: Anti-Reversion Headers

Post by MadBill » Thu Oct 19, 2017 8:52 pm

I just tracked down that A/R test I referenced on page 6. It's in David's 1981 book How to Build and tune Cars for performance and Economy and I got all the info right.
Felix, qui potuit rerum cognscere causas.

Happy is he who can discover the cause of things.

77cruiser
Expert
Expert
Posts: 949
Joined: Fri Dec 29, 2006 10:32 pm
Location: I Falls MN
Contact:

Re: Anti-Reversion Headers

Post by 77cruiser » Thu Oct 19, 2017 10:15 pm

tenxal wrote:
Tue Oct 17, 2017 9:16 pm
NewbVetteGuy wrote:
Tue Oct 17, 2017 7:41 pm
Has anyone test the benefits on the dyno enough to provide some practical rough guidelines with #of foot-pound estimates for improvement in a few real-life examples? Adam
Tested back-to-back: 1 3/4" dyno headers with standard 4 into 1 collectors, then the same headers with excellent merge collector, then a set of very good 1 5/8"-1 3/4" steps with standard 4 into 1 collector, then the same headers with the previous excellent merge collector and finally Elston Tri Y's with the collector furnished by Elston.

The results show that the theory is fact. :wink:
So what were the numbers?
Jim

joe 90
Guru
Guru
Posts: 1991
Joined: Tue May 27, 2014 4:02 am
Location: The land of the long white cloud

Re: Anti-Reversion Headers

Post by joe 90 » Thu Oct 19, 2017 10:16 pm

modok wrote:
Thu Oct 19, 2017 8:51 pm



So TO ME, the AR chamber does not work as imagined, BUT, it is still the best way to add volume to the engine side of the header pipe with the least downsides.. Think about it. If you can think of a better way.... please share. :D

To me.......it works in the same way as LCR and rectifier circuits used in electronics.
The chamber acts like a capacitor.
The step acts like a leaky diode.
The pipe is an inductor / resistor.

User avatar
modok
Guru
Guru
Posts: 1307
Joined: Sun Jun 06, 2010 1:50 am

Re: Anti-Reversion Headers

Post by modok » Thu Oct 19, 2017 11:17 pm

I agree.

As a diode I think it would work better -inside out-, meaning the outer pipe would have the main outflow and the inner pipe would be the dead end a ways upstream....is that a good idea?.... probably not, but....feel free. I sure don't know. Only way to prove it doesn't work is try it.
Glen Urban

joe 90
Guru
Guru
Posts: 1991
Joined: Tue May 27, 2014 4:02 am
Location: The land of the long white cloud

Re: Anti-Reversion Headers

Post by joe 90 » Fri Oct 20, 2017 3:12 am

A bit like tesla's vacuum pump?

https://teslaresearch.jimdo.com/death-r ... ator-pump/


That link also has a digram of a "Tesla fluid diode".

User avatar
modok
Guru
Guru
Posts: 1307
Joined: Sun Jun 06, 2010 1:50 am

Re: Anti-Reversion Headers

Post by modok » Fri Oct 20, 2017 4:50 am

Yes, like that, proportions just a little different, I'm sure it will flow terrible backward!! but not as good forward either #-o

you know, tesla's vac nozzle would probably work better inside out too, although both ways are used today, so it probably works ok both ways.

And....hmmm, perhaps next we try out a co-axial collector?

https://www.blowervacuumbestpractices.c ... fficiently
Glen Urban

n2xlr8n
Pro
Pro
Posts: 369
Joined: Tue Jun 28, 2005 9:01 pm
Location: Bama

Re: Anti-Reversion Headers

Post by n2xlr8n » Fri Oct 20, 2017 9:02 am

joe 90 wrote:
Thu Oct 19, 2017 10:16 pm
modok wrote:
Thu Oct 19, 2017 8:51 pm



So TO ME, the AR chamber does not work as imagined, BUT, it is still the best way to add volume to the engine side of the header pipe with the least downsides.. Think about it. If you can think of a better way.... please share. :D

To me.......it works in the same way as LCR and rectifier circuits used in electronics.
The chamber acts like a capacitor.
The step acts like a leaky diode.
The pipe is an inductor / resistor.

...and to me, it's a tuned cavity circuit / waveguide, as in a particle accelerator. Except it's not- the "signal" from an engine is not continuous wave, it's variable frequency pulsed, which reeks havoc on this line of reasoning.

In max-effort race engines, the builder/engineer shoots for a specific rpm range, which is very much like a tuned circuit/cavity. They design it for max Q throughout the desired range, but it's still a trade off.
He who is in me is greater than he who is in the world.

NewbVetteGuy
Member
Member
Posts: 196
Joined: Sun Aug 07, 2016 4:11 pm

Re: Anti-Reversion Headers

Post by NewbVetteGuy » Mon Oct 23, 2017 4:00 pm

I feel like part of the thread is dredging up some sort of "holy war" / two opposing schools of thought on the subject, and I know Glen has stated that the subject of the thread is in anti-reversion devices being built into headers, but I'd REALLY love to have some conversation on other decision points in header design that can minimize reversion and it seems highly relevant to the thread.

If there's a controversy, it seems like getting it out into the open will be the best way to help people make an informed decision on the best way to minimize reversion for a given application and budget.


I've been reading a lot of Calvin's posts and his blog on this subject this weekend so I'm going to ask a few pointed questions:
  • Are there other header design decisions that can effectively limit reversion OTHER than anti-reversion chambers? If there are multiple options that can help reduce reversion, please place them in a prioritized order from most useful to less useful.


I'm going to TRY to attempt to summarize what I think the "Elston school-of-thought" is saying, if for no other reason than to screw it up royally to goad him to respond in his own words. :D :
  • High velocity in the first 7-12" of the header primaries (obtained by using an initial primary diameter on the small side and then stepping up the diameter) can not only help with reversion by increasing the "blowdown length" but also results in better cylinder scavenging (and is cheaper than adding mini, individual cylinder anti reversion dams at the step)
    -If that's a fair characterization, is there still a benefit to implementing Feuling / Widmer-style anti-reversion dams at the step? (I'm assuming some, reduced benefit, otherwise why would Larry be spending the extra $$ to add them?)
    • Merge collectors can also act as anti-reversion devices and, unlike the after-the-collector anti-reversion devices (ala Fueling/HyTech, Hendren Racing/ DynaTech, or Shoenfeld) can help improve both mid-range AND high RPM power

    I am quite certain I've gotten some of this wrong, because based upon my understanding, I wouldn't expect to see any traditional anti-reversion chambers on an Elston exhaust and other's in this thread have already pointed examples of end-to-end Elston exhaust systems that DO have anti-reversion chambers on them, so I know I'm confused.

    Adam

    hoffman900
    Guru
    Guru
    Posts: 1093
    Joined: Sat Feb 23, 2013 5:42 pm

    Re: Anti-Reversion Headers

    Post by hoffman900 » Mon Oct 23, 2017 4:25 pm

    I am quite certain I've gotten some of this wrong, because based upon my understanding, I wouldn't expect to see any traditional anti-reversion chambers on an Elston exhaust and other's in this thread have already pointed examples of end-to-end Elston exhaust systems that DO have anti-reversion chambers on them, so I know I'm confused.
    Calvin has used them in the primaries of 4-1 systems and at the merge of both banks into a single tailpipe on a road racing set up.

    Not sure how much he wants to share.

    Here is a post of an application I'm very familiar with:
    Ok, Erland, as I would deal with all customers, do you know the area of your exhaust port at the flange? If this 4 banger makes about 220hp? then primary size should be no bigger than 1.5" OD or 38mm OD with a .050" wall. That is all you need max. So right off the bat I would tell you the header is too big and if the port is that big then I would recomend starting there, ... why is it so big, (probably from the manufacturer, :( ,sigh), do the rules allow making port smaller and would you consider that, and how big is the valve and can it be made smaller? and would that allow a larger intake?
    1.75 primary can handle 400 hp so I don't need to go much further to tell you probably the only thing you are ever going to get your header to change anything is the overall length. Almost any motor will respond to that.
    I just recently did a Triumph TR4 2.1, (or maybe 2.2?) litre tractor motor, made 194+ wheel hp and the header was 1.375" off the head, which was actually 2.5% smaller in area than the port. Going from his old header of 1.625 off head. Engine went from 180 to 195 hp. :shock: The choke was 2.0" with a taper out to 2.5 but after 8" the tailpipe went back down to 2.25 for the rest of the car. It is a 4into1 header with AR valves in the primaries about 10 inches out. If we had done a 421 the choke would have been even smaller. 1.75" Secondary would be getting too big but there is a much larger area difference in smaller tubing going from one gauge to the next than in larger tubing. I would probably be inclined to step the secondary from 1.63 to 1.75 and 2.0" max choke, or stay at 1.63 with a 1.88 choke.

    It can be hard to imagine this but that is an example of an engine that was dealing with contaminated reversion the entire power band. If I remember it topped out at 8k but usually only run to 7500. The big header could not keep the burnt gases away from the cylinders. If you hear what I am saying, we did not make this engine pull close to 10% more air for the entire power curve with the header, the header was able to keep burnt gases away from the cylinders and we had more available oxygen to make more power the entire power curve. It is amazing and he is very happy.

    Can we agree that if you have a beautiful depression across overlap from a "tuned" header, that does not mean that the chamber is pulling from only the intake. 8)
    Note, the peak is at 6500 and runs out at about 7000rpm. Everything else is dead on.

    If you think of a Tri-y header, where each collector is an A/R node, plus the better velocity gradient through the header, that should help you understand how they work really well at what they do.
    -Bob

    digger
    Guru
    Guru
    Posts: 1004
    Joined: Sun Jan 27, 2008 6:39 am

    Re: Anti-Reversion Headers

    Post by digger » Tue Oct 24, 2017 5:39 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....

    hoffman900
    Guru
    Guru
    Posts: 1093
    Joined: Sat Feb 23, 2013 5:42 pm

    Re: Anti-Reversion Headers

    Post by hoffman900 » Tue Oct 24, 2017 7:18 pm

    What tests would these be? Engine type / combination / how developed was it actually?

    I think larger than typical headers only make a difference when:

    [*]The header is poorly built - tight bend radii, lack of attention to the flange / port transition
    [*] Something in the combination is so wrong, that the engine rather just pump it out. Likely a camshaft issue.

    You really can't compare something like Calvin, Mark Lelchook, and a handful of top headers build with what you can order out of a catalog (unless that catalog is theirs).
    -Bob

    digger
    Guru
    Guru
    Posts: 1004
    Joined: Sun Jan 27, 2008 6:39 am

    Re: Anti-Reversion Headers

    Post by digger » Wed Oct 25, 2017 2:54 am

    hoffman900 wrote:
    Tue Oct 24, 2017 7:18 pm
    What tests would these be? Engine type / combination / how developed was it actually?

    I think larger than typical headers only make a difference when:

    [*]The header is poorly built - tight bend radii, lack of attention to the flange / port transition
    [*] Something in the combination is so wrong, that the engine rather just pump it out. Likely a camshaft issue.

    You really can't compare something like Calvin, Mark Lelchook, and a handful of top headers build with what you can order out of a catalog (unless that catalog is theirs).
    one example purely from an observation perspective was reading the vizard book mentioned before. it was a pretty low spec and low developed engine 300hp 327ci might be a stocker not much else is stated. pipemax would recommend 1-3/8" to 1.5" at most but the bigger 1-3/4 made more power from 3500 up than the 1-5/8" and 1-7/8" was down alot the engine was happier at a certain size. you cant draw too much conclusion from that as not enough info . perhaps small cam cant use the proper tuning benefits of a header so it is just the free flowing manifold coming out on top or they didnt test small enough....

    there are few other tests you see floating around i dont have any links saved though, it was more of a general what have others observed.

    the gradual bends and better port transition makes sense, ive also seen calvin recommend slightly shorter length pipes. what about cam timing and or lobe shape any trends or specifics people have seen?

    User avatar
    modok
    Guru
    Guru
    Posts: 1307
    Joined: Sun Jun 06, 2010 1:50 am

    Re: Anti-Reversion Headers

    Post by modok » Wed Oct 25, 2017 4:43 am

    NewbVetteGuy wrote:
    Mon Oct 23, 2017 4:00 pm
    I feel like part of the thread is dredging up some sort of "holy war" / two opposing schools of thought on the subject, and I know Glen has stated that the subject of the thread is in anti-reversion devices being built into headers, but I'd REALLY love to have some conversation on other decision points in header design that can minimize reversion and it seems highly relevant to the thread.
    Perhaps....you are right. Lets talk about the forward and backward properties of all the pieces?
    I don't see opposing schools of thought, much less a specific number of them. Mainly not enough perspective to see the WHOLe thing, I think, although I don't know a lot about particle accelerators or rocket nozzles so....I'll plead the fifth on those. I'll think about how to discuss the forward/reverse flow and oscillations all the way to the tailpipe
    Glen Urban

    Post Reply