U20 Headers

[gtrbg]

The latter exhaust is the one on my car. Had front half of a BRE exhaust when I bought the car and had to get the remainder made. It runs through a 3 inch straight through muffler and exits in front of the drivers side wheel. One rocker support had been previously modified but the exhaust tucks in nicely between the rocker and frame so no bottoming issues yet on the track and hill climbs. Fairly loud even with a helmet on:). Just got it Cerakoted to hopefully have it last longer. Seems to make decent torque since it spun/broke the 5th gear:(.

[casadeclovis]

I'm, Bob Swan, who was crew chief for Balboa Racing 1972-1977. Bill Swan is my brother. The exhaust system on all our roadster engines were 4-2-1 interference type. and the lengths were the same for all including our '95 Solo National Champion Honda 1200. This system gives superior bottom end performance. For the R1600 and Honda 1200 we used 1 1/2 " tubing for the primary. The length out of the head is 12". At the "Y", end with about 1" of straight tuning. A 3/4in long sleeve (1 5/8" dia) is slid up about 1/8" above the end and welded on. The 2nd "Y" ends slide up onto this and is 1 3/4" dia. Keep both 1 1/2" 1st pipes parallel to facilitate installation and removal of the 1 3/4" second "Y" pipes. The second "Y" tubes are 24" long and end with a 1 3/4" spigot. Add a 3/4in long 1 7/8 sleeve and install the single 2" exhaust pipe from here. The U20 is similar but starts with a 1 3/4in primary, 1 7/8 Spacer, 2" 2nd "Y", 2 1/8 spacer and 2 1/4 final exhaust pipe. The R1600, which had competition Datsun intake and exhaust valves ( 1.69 Int and 1.38 exh.) and a .5" lift Crower cam put out 135HP at 6500, 141 at 7000 and 7500 and 135 at 8000. 4-2-1's give wide range. In 1977 the engine blew off Col Hauser's 1600 and had over the length of the front straight lead when a miss fire set in and we couldn't rev over 7K. The same Car ran a U20 engine: had 205HP but strangely at 6500rpm. We changed the rear end ratio to allow us to top out at 133 at 6500 in 4th gear. 1 to 1 straight thru so no HP gear loss. We were fastest car down the straight in the 1st practice. Road Atlanta. The U20 ran the 44mm Mikuni Carbs with 40mm inserts. We also ran the Hitachi carbs on the dyno and put out 202HP but the idle and midrange were clearly inferior. The Honda we never ran on a dyno. We could run it easily to 9000rpm if needed but tried to keep it to 8500. Also and VERY IMPORTANT. The exhausts on these cars were all wrapped with asbestos (all we had then) and had cold air boxed. This gives a significant horse power gain. Again I say significant.

[rwmann]

Excellent! A photo or drawing would be magic!

[drieseck]

Fantastic write-up Bob cascadeclovis, thanks for sharing the actual specs and experience in detail.

Here's another header configuration and a sweet Jenvey FI set up. FWIW, I think they sell the header on their website.

https://www.facebook.com/photo/?fbid=33 ... cale=sw_KE

[casadeclovis]

Just for understanding. An exhaust wave leaves the combustion chamber with positive pressure. It is reflected at the end of the exhaust as a negative pressure wave. This in turn will be reflected as a positive wave (1). by a closed exhaust valve or (2) it will go thru the combustion chamber at overlap and will be reflected as a positive pressure wave at the carb bell. This can lead to "stand off" where a small cloud of fuel will form just in front of the bell. The bell reflected positive wave can cram extra air/fuel into the combustion chamber if the exhaust valve closing is timed correctly which is good. If exhaust valve closes late some air/fuel will be pushed out of the chamber which is bad. Too early and the chamber will still have exhaust in it, also bad.
The 4-2-1 interference exhaust has the opposite effect of a tuned exhaust. A tuned exhaust will give a boost in HP at the RPM it is tuned for. This tuning is effective over a small RPM range but it can actually reduces power above and below Max HP. A 4-2-1 exhaust reduces the strength of the wave by splitting it up thus decreasing max HP but increasing HP above and below. Thus you get a wider HP range at the cost of peak HP. Many street cars use this system. The interference system maximizes this buy further reducing the strength of the wave accomplished by reflecting part of it at each intersection both by a pipe diameter change and by making the change sharp edge guaranteeing that the part of the wave going up the the wall of the pipe will be reflected. I got this design from a book, "The scientific design of exhaust" (or close to that) copyright about 1926.It has served me well. I don't believe the lengths are extremely important, certainly not like they are in a tuned exhaust, and all our exhausts were built as near to the lengths stated as was convenient.
I liked to have a tab welded on that I could use to drive the slip fit connection apart with and and tabs on the upper and lower pipes for a spring to hold them together. I believe our friend Jerry Jones also used this type exhaust on his 2liter roadster driven by Bro Bill Swan 1980 ish.

Tri Y.jpg

[drieseck]

Exceptional Bob, thank for all the follow-up clarification and diagram, very nice education for some of us. Much appreciated. Dave
Hopefully Steve can put these in the Tech Wiki.

[Gregs672000]

Agreed! I come away with a much better understanding... thank you. I note that the original stock manifold was a 4-2-1 design, but obviously nothing as well designed as what's been described. Would be fun to run a stock vs 4-1 header back to back on the dyno. I THREW my old cast manifold down the alley after giving up on welding the cracks again, but have since heard it was missing an important connector to protect it from that. It was heavy beast though and I have no idea how well it flowed etc... just curious. Did anyone race them, or was it automatic to replace with a header?
Thanks again for these very informative posts.

[Daryl Smith]

Smooth bends are very important for max performance on a header, and too many bends used to get an 'equal length' header can also hurt performance. It is important to get a good, experienced welder/fabricator for your header....with references!
Equal length is beneficial for a 4 cyl, but, not at the expense of additional bends to get it.
Diameter of the header pipes is important for horsepower output, and length is important for the rpm range. Length is probably the less important of the two, within reason, iirc, length will have more of an effect on your torque peak rpm and curve....possibly restricting power slightly at higher rpm if too long.
While the specs are important for MAX output, say for a 'tuned' header, for the most part 'close' would be good enough for most. The exception might be collector/exhaust total length, where it can make a noticeable difference.
A couple of benefits of the 4-2-1 header is that it should be slightly lighter than a 4-1, and will also have more clearance from the starter because there are only 2 pipes running below it rather than 4, so your starter will be cooler. Ceramic coat it and every thing runs cooler still....

[casadeclovis]

Smooth bends are very, very important. On our Solo Nat Champ Honda our 1st set of headers {yes tri Y interference) were 1 3/4 diameter primary. Poor fabrication (me) and only adequate welding (Bro Bill). Later I built a set or 1 1/2" primaries very well fabricated and welded. These flowed much better than the 1 3/4 set. Thus we had a win, win. Better low end from higher low end velocity and better top end from better flow. Bob Swan