I tried to design this header based on what math and expierence and advise from more senior engineers I have to draw on (which is comparatively little). I did not look at the basic design dimensions I have for other existing headers, and when i put my header down next to the OE manifold for the first time...... they looked pretty similiar
And what do you know, it performed in a similar (but a litle better) way. I did a head to head test, just like the last serries, here are the most average test runs (with no vafc, as that introduced to many variables last time, and i dont have an wideband O2 to test with, yet). Please remember that these numbers are only relative to themselfs, and that this is power at the wheels, not compensation for aero drag (which at 64mph is about 13whp on a prelude) or compensated for environment (which i did calculate to be at about 0.992, thanks mc_honda).
heres the env variable data for the test times:
061104 8:53 PM
temp: 59.0 °F / 15.0 °C
dewPoint: 33.1 °F / 0.6 °C
hum:38%
pres: 30.01 in / 1016.1 hPa
wind: E 6.9 mph / 11.1 km/h
061104 11:53 PM
temp: 59.0 °F / 15.0 °C
dewPoint: 33.1 °F / 0.6 °C
hum:38%
pres: 30.07 in / 1018.1 hPa
wind: ENE 6.9 mph / 11.1 km/h
All in all, im not happy enough with this design to make it the final version. Im going to go +1 on the pipe sizes and play with length (been doing valve timing/gas and pressure timing "math" all weekend, I might even go shorter on the primaries). I also, saw some F1 motor pics (not honda's of course) and Renault uses a steped header design, I still have never seen a "steped vs non steped" test though.
Lepto, you deserve mad props for this. I wish I could be there with you playing with header designs.
You are one of a very few set of people with the skills and resources to play with things like this. If you want, I can go dig up some more exhaust header design resources for you. I'm sure you have many already, but maybe I can find a few more interesting ones.
__________________
I am not an active moderator on this site anymore. Please contact Phryxis or liquidxnitez for help with your issues.
Nice looking header! Yeah, of course it looks similar to the OEM one, but your primaries and secondaries are longer than OEM. That's got to acount for something. And obviously the exit is larger.
Is is the persective of the pic, or is the pipe before and after the flex-pipe different size? It appears that the pipe before the flex pipe is smaller.
As for the gains, yeah they are not alot, but it appears you are on the right track. IMHO the stock header must be very well designed b/c even "expereinced" header manuacturers have trouble making increased power every time via a header on an H22.
Props to you and keep up the good work.
- Paul
__________________ Paul
2009 Mazdaspeed3 | 1995 Miata | 1999 Crystal Blue Base - bought 6/13/02, sold 3/11/09
I swapped in one of Marcucci MS's vafc harnesses, and now that my obdii work around isnt pressent im getting an occasional CEL code 1... so im trying to tease that one out, maybe thats affected my previous testing.
The more thinking i do about my next pipe lengths the more i like the idea of short primaries and corrispindingly tuned seccondaries. Does anyone (pauluce ) know what gas temp is in a near stock H22? I remember paulace's pyrometer was at arouns 1200f pre-boost, but i dont know how an NA H22 does... ? anyone?
I wonder if the OE O2 sensor freaks out being in a 2.5" tube... Maybe its not far enough past the boundry layer to get good readings. Or maybe i dammaged in durring one of the 15 or so header swaps i've done this year Any senior mech/eng have any thoughts?
I don't know about the temps but the boundry layer in gas flow is usually small and it is not really discussed (at least not in my text books). Boundry layer discussion seems to center around water and other liquids. This makes sense as most gasses have a very low viscosity (2 orders of magnitude or more) than most liquids.
Another thing to note is that in my Gas Dynamics text it talks about the effects of laminar and turbulent flow on the boundry layer in gasses. A turbulent flow will actually give a smaller boundry layer and more uniform velocity profile than laminar flow. I would have to assume, in that area of the header, that the flow is at least somewhat turbulent, if not fully turbulent. One thing you can try to do is calculate the Reynolds number of the flow. To do that you would need the approximate velocity (not easy), the density (also somewhat of a guess as you need temp), the viscosity of the exhaust gas mixture (maybe you can find this online or in a text), and of course your pipe diameter. I'll have to look in my ICE book when I get home to see if it gives any hints.
__________________ Paul
2009 Mazdaspeed3 | 1995 Miata | 1999 Crystal Blue Base - bought 6/13/02, sold 3/11/09
The pyrometer in my exhaust shows about 1300F at WOT. That is measured in the collector, so there is some heat loss from the port to the collector. From experience I would estimate 1350F to 1400F in the primary pipe. If you've leaned the car out some, then your EGT may be a bit hotter.
My experience with headers is to not be too worried about laminar or turbulent flow..... it's all turbulent. Focus more on manipulating the pressure pulses and using them with the natural frequency of the pipe section you are tuning to reach performance objectives. Use pipe lengths to place the system gains where you want to achieve gains, then use pipe diameters to control the magnatude of the gains and losses. Basically, this is a very complex and difficult to model system you are trying to characterize, so don't get carried away. For my experience, you will learn much quicker what the engine wants by experimentation and emperical model building than by keeping in the books. The best references out there are SAE technical papers because they are very practical. Look for papers on tuned pipes and branch resonator systems for the most relavent information.
EGT are going to vary a lot on thermocouple location... My car has always been boosted, so I don't know what EGT's were pre-boost. but now 1500F is common at WOT & MAX RPM. Thermocouple probe is placed 3" from the head which is around the hottest location.
thanks for the temp info! i'll assume 1300f at WOT (but how often are we at WOT at 6K rpm + on the street? )
I've been using Smith's math for basic pipe lenght estimates so far (seems thats what most people use, in some form or another). But i just got a copy of GT Suite maybe i'll play with that for fun
so i've been playing around, thinking about pipe diameters (while drinking wine after my reccent 9hr+ work days) i've been using
1 5/8" OD pipe and:
H22 exhaust port area = aprox 1196.5mm^2
1 5/8" OD area = 1368mm^2
1 5/8" ID area = 1210mm^2
maybe not so good...
1 3/4" ID area = 1414mm^2
1 3/4" OD = 1551mm^2
... probably the best
1 7/8" ID =1634mm
... bi-polar or ok?
back to the sae paper search. i havent been able to find anything in SAE's tech papers about anti reversion pipe design, but alot of people use them (even some F1, but not Honda, as of a few years ago)....
i have been enjoying the v4 header though, even thought the power gain wasnt huge the throtle response is nice, and the 15lbs weight loss might help too (as much as a cfrp hood)....
I've to add to the GTech, i've got an LM1 on the way (on loan, THANKS!!!), weather data logger, and water and EGT temp gauges (its my birthday on the 20th hehe). so i'll be a one-man-data-loggin-band (imagine a little song here).
Also, the math i was using at first (from Smith), see my first post from last year , changes alittle with the better temp data MC and paulace gave me (thanks fellas):
L = ideal length in inches
V = velocity of sound in exhaust gas, can range from 1100 - 1900 fps
T = 120' exh port open time in crankshaft degrees (is this correct for h22a4?)
RPM = disired torque peak RPM
H22 EGT = 1200-1500f
speed of sound at 1300f = 2050fps
speed of sound at 1400f = 2115fps
speed of sound at 1500f = 2171fps
*in standard atsmophere
the above length estimates line up with my v4 accelerometer testing... but the narrow pipe size should have givven better torque down low... hmm
This and the better OE manifold measurements, accelerometer testing and 4 prototypes worht of cut-n-try are helping me to re-think my design. As MC suggests, im trying not to model the system in to much detail, but just enought to get me going again (friggn obdii gremlins kinda shook my confidence in how well i understood the design problems...) anyway...
... so im helping our formula sae tema out this year and one of the fellas noted that the best way to calc the speed of sound in exhasut gas is by it's "gamma" which i assume to be a variable that includes pressure, density and temp... atsmospere get a "1.4" and EG tends to be at "1.2" he recalled... i guess that the speed of sound of 2050fps for 1300f EG is a touch fast, probable closer to 1900fps or so.
oh yeah.... GO BARF1 i cant believe that thiers actualy thier web site. honda came in 3&4 in F1 this weekend... but more importantly Sato didnt blow anythign up this time
oh yeah, water temp gauge and EGT pyro are on the way (gotta love birthdays, 26 now) and I GOT A (ceramic) TANUKI !!!!! im going to do all further gtech testing at same water temp (curses to OBDII).
I changed the v4 last night, adjusted pipe lenght and added a 3rd, plugged O2 bung for the LM1 (borrowed, TY!!) work i'll be doing. I wanted to push the power gains down in the RPM range by 1000rpm, hopefully it will be an improvement. Im working on the desgin changes for the next (I hope-hope-hope it will be the final version) prototype, and im expecting material for a whole exhaust soon .
Based on the results of the last test i made a design modification to my v4 prototype. v4.5 went on the car last night and feals good.
I changed pipe lengths to move the gains i was getting down by ~1,000rpm. The engine feals more powerfull in the 3,500 and up rpm range (which is were a street driven, USDM geared H22 can use power).
Im going to do a head to head test with the OE manifold ASAP. Im also doing as much reading and basic analytical work as i can (even thought this design is pretty good already). Im waiting on flanges to build the final prototype.
Once you get a full out production model would you consider selling one of these prototypes to me. I want somethng better then DC/Neuspeed and that's about all I require.
Leptolude, the work youa re doing shows the kind of dedication that is needed with Honda modification, especially the Prelude. Is it possible to do any of the theoretical testing on a computer? You know, the various bends and tube lenghts. Whatever happens, this is great work. Another aspect of exhaust tuning that should be looked at is the location diameter and length of the muffler and the relocation of the resontor at the end of the system rather than the reverse as in OEM. I have been looking at the design of various high performance european exhausts and they have this configuration. In addition, you are then able to get that special european exhaust sound which is built into the resonator.
Im still waiting to get the time to test v4.5... i've had family visiting and works been busy. But I did to some basic resonance modeling (helmholtz from SAE papers). Its not to much better on the exhaust side than the methods Smith had in his book, but it was a good learning expierence.
I've also been sourcing materials for the next (2) manifold(s) and an exhaust prototype. As for the exhaust design, at those pipe lengths there isnt too much pressure tuning to do so i've been working on a muffler design (finding perforated tubes on low volume is frig'n tough).
I've had the v4.5 on the car for the past few weeks and its been nice. Im looking forward to the testing!! If it yields the results i want then im going to move quick to get to production before Fall (when i start grad school )
The AutoGuide.com network consists of the largest network of enthusiast-owned enthusiast-operated automotive communities.
AutoGuide.com provides the latest car reviews, auto show coverage, new car prices, and automotive news. The AutoGuide network operates more than 100 automotive forums where our users consult peers for shopping information and advice, and share opinions as a community.
) know what gas temp is in a near stock H22? I remember paulace's pyrometer was at arouns 1200f pre-boost, but i dont know how an NA H22 does... ? anyone?
)
Linear Mode
