The exhaust is almost always the first thing to go. Before the seat, before the bars, before anyone has looked up what a fork brace does, the stock pipes come off and something louder goes on. It is the modification that most obviously turns a commuter into a cafe racer, it takes an afternoon, and it is the one every builder is most confident about.
Two threads on the SOHC/4 Owners Club forums suggest that confidence is misplaced in two completely different directions, and we haven't covered either of them in this series.
The first is a 249-reply, fifty-three-thousand-view thread from 2013 that begins with a CB550 owner who has just bolted on a stainless four-into-one and wants to know what jets to buy. It takes ten pages, two respected veterans disagreeing flatly with each other, one trip back to square one, and four months before anyone finds the actual problem — and the actual problem is not a jet.
The second is a 997-reply thread started two months later that has been running for more than a decade and has nothing to do with performance. It is about what the pipes you just took off are worth now, and it is the closest thing the forum has to a cautionary tale.
Read together they cover the whole exhaust decision, which turns out to be two decisions and not one.
The question everybody asks
The thread opens in August 2013. A member has a '76 CB550K, a MotoGPWerks four-into-one, and the stock airbox with a foam filter in it. His question is the one this forum answers about forty times a year: what jets?
The first answer is the obvious one — ask the pipe maker. The second is more useful: buy an assortment. Several members pointed him at the small jet kits you can order in a chosen size range, which is genuinely the right first move. If you're going to change jets on a set of Keihins, changing them once is a fantasy. Budget for doing it three or four times and buy accordingly.
Then the reports start coming in, and they don't agree with each other. One member with the same pipe on an otherwise stock '75 550K was running #105 mains, #40 idles, the needle in the middle position and the air screws at 1.25 turns, at four to eight hundred feet of elevation, with plugs that everyone agreed looked fine. Another member ended up at #110 mains, #38 idles, needle third notch, 1.5 turns — but he'd also swapped in a 650 cam. A third, quoted from a video, was at #110 / #38 / third notch / one turn out and declared his tuning "spot on."
Four bikes, four sets of numbers, all with the same pipe. That spread is the whole problem in miniature, and the thread spends the next several pages working out why.
The disagreement: does a freer pipe want more fuel?
The intuition every one of us brings to this is simple. A four-into-one flows better than four restrictive stock mufflers. Better flow means more air. More air means more fuel. Go up on the mains.
One of the board's most experienced members went after that assumption directly, and his version is worth understanding even if you end up disagreeing with it: bolt on an aftermarket exhaust while leaving the intake alone, and no jetting change is required at all. No extra air is coming in. The carburettor doesn't know what happened downstream of the exhaust valve. At most you might want a quarter turn on the mixture screws. Change the intake — pods, stacks, a hacked airbox — and you have changed the entire game. Change only the pipe and you mostly haven't.
The rebuttal came from another long-time member and it is the single best-argued post in the thread, because it doesn't rely on opinion at all. It relies on Honda.
Compare two bikes with the same engine, the same cam and the same induction system. The CB550K of 1974 to 1976 came with the four-into-four exhaust, a #100 main jet and the needle clipped in the fourth position from the top. The CB550F of 1975 to 1977 came with a four-into-one, a quieter muffler running at higher pressure — and a #98 main jet with the needle in the second position from the top. Same motor. Leaner jetting on the four-into-one.
The mechanism he offered is that higher exhaust pressure reduces cylinder scavenging slightly. Some of the previous burn's hydrocarbons stay in the chamber, so the next cycle wants a little more oxygen and a little less fuel — which makes the engine more efficient through the low and mid range where street riding actually happens, almost certainly at the cost of some peak power near the redline.
His supporting evidence is the 1977–78 K models. Still four-into-four, but with a reduced outlet diameter on each muffler, PD carburettors, #90 mains and a dedicated idle circuit for fine mixture control. Honda raised the exhaust pressure and cut the fuel. That is a lean-burn engine, built to an emissions target, and it points the same direction.
The upshot is that both camps are half right, and the half they share is the important bit. Changing exhaust pressure can change what the engine wants. But the direction is not automatic, and it is not necessarily richer. Whether the pipe on your bench flows more or less than what Honda jetted around is a question the pipe's maker probably cannot answer, because that number is not what most exhaust manufacturers are optimising for. That engineering detail, as he put it, is left to whoever bought the exhaust.
Which is a polite way of saying: you're going to have to read your plugs.
Back to square one: which carburettors do you actually have?
Three pages in, after everyone has been trading jet numbers in good faith, someone finally asks the question that should have come first, and the thread grinds to a halt.
The needle position everyone had been calling "stock" wasn't stock. The idle jet everyone assumed was a #38 might be a #40. And the reason is that on a CB550, the carburettor set number matters more than the model year. The chart that came out of that reset is one of the more useful things buried in this forum, and it's worth having on the bench:
- 022A — 1974–75 CB550K: 22mm float, #100 main, #38 idle, needle 4th notch
- 087A — 1976 CB550K: 22mm float, #100 main, #38 idle, needle 4th notch
- 657B and 086A — 1976 CB550K: 26mm float, #105 main, #40 idle, needle 4th notch
- 069A — 1975–77 CB550F: 22mm float, #98 main, #38 idle, needle 2nd notch
Look at the two 1976 K entries. Same model, same year, two different carb sets, and a four-millimetre difference in float height between them along with a different main and a different idle jet. If you get that wrong you are tuning a fuel level, not a mixture, and no amount of jet swapping will save you.
The original poster, incidentally, had 022A carbs on a '76 — which per the chart belong to the '74–'75 K. Whether that was a factory anomaly or forty years of previous owners is beside the point. The point is that "I have a 1976 CB550" is not enough information for anyone to help you, and three pages of well-meant advice had been offered on that basis.
The other thing that came out of the reset is a warning about trusting history. When the poster said the needle was in the stock position because a Honda shop had been the last to touch it two and a half years earlier, one of the mechanics in the thread was blunt: he didn't care who had worked on it, and he wasn't convinced anyone had actually pulled the carbs and looked. Checking and changing needles on these carbs is well under an hour once you know the job. Do it, and then you know.
Stop staring at the main jet
The most quietly valuable correction in the thread is about which circuit you're actually riding on.
The main jet governs roughly three-quarter throttle to wide open. That is a vanishingly small share of your riding, and it isn't the freeway either — you are not at wide-open throttle at 65mph. The needle, which rides up and down through the main jet, controls the quarter-to-three-quarter band, which is where nearly everything happens. Below that, the pilot circuit and the mixture screws own idle and just off idle.
Moving the clip down raises the needle in the emulsion tube and richens the mid-range. Moving it up leans it. That is the adjustment most people should be making, and it costs nothing but time.
You can watch this play out in the thread. The poster went from #100 to #105 mains and immediately reported worse hesitation coming off a dead stop — which puzzled him, reasonably, because the mains shouldn't be doing much down there. What actually fixed the hesitation across the mid-range wasn't the main at all; it was moving the needle clip. And when he tried #40 idle jets with the #105 mains, the plugs came back sooty and the bike ran, in his words, like poo. The #38s that looked slightly lean ran better than the #40s that looked right.
That's the practical lesson: mains are the last thing to change, not the first, and a plug that looks faintly lean on a bike that runs cleanly beats a plug that looks perfect on a bike that stumbles.
The answer nobody guessed
By November the bike was close. Genuinely close — strong pull, no stumble, plugs that a couple of veterans looked at and told him to stop fiddling and go ride. But there was one fault left: a small blubber or flutter at about three-quarter throttle, between six and seven thousand rpm. He chased it with every combination of jet, clip and screw he had.
Two members had already told him where to look. One said simply: check your float level. The other, running eight-hundred-dollar CR carburettors on the same pipe, said the same thing with more force — he'd bolted brand-new race carbs straight onto his 550, chased a tuning gremlin for a week, finally measured the floats and found them wildly out of specification. Out of the box. On new carburettors.
In December the poster posted his update. What cured the flutter was resetting and checking float height to 22mm, the stock figure for his carbs. Nothing else. The final configuration he settled on was a #105 main, a #38 slow jet, mixture screws at 1.25 turns, the needle clip third from the top, and a foam filter in the stock airbox — plus a small trick worth stealing, a spacer nut and stud between the airbox lid and body to let a little extra air past. After that the bike would carry two people to 80mph on the freeway without complaint.
Ten pages, four months, and a jet kit's worth of brass, and the answer was fuel level, not fuel metering. It's the same lesson this forum keeps arriving at from every direction: the jets meter fuel, but they meter it out of a bowl, and if the bowl is at the wrong height every number downstream of it is a guess. Set float height first. Then sync. Then think about jets.
Getting the thing bolted on
The fitment notes scattered through the thread are worth collecting, because a fresh exhaust leak will produce symptoms that look exactly like a jetting problem and will send you back to the parts box for no reason.
Copper crush gaskets are not optional and mostly not reusable. Before the new headers go anywhere near the head, check that the old gaskets came out. They're sooty and they blend into the port, and the reliable way to find them is to run a flat screwdriver around the flange face — they generally pop straight out. Skip that and you're stacking two gaskets and chasing a leak forever. On reuse, opinion splits: one camp says you'll usually get away with it once, the other says never. The stronger argument is the second one, and the reasoning is specific — thin-walled aftermarket headers cut a narrow groove into the gasket, and you will not get the pipe back in exactly the same position, so the groove no longer lines up. The hollow cylindrical copper gaskets were the preferred type in the thread.
Fit the header as an assembly. Several people tried bolting the collars to the head first and then feeding the header on, and it's a fight, because on a four the outer cylinders sit at a slight outward angle. Assemble the sleeves, collars and header together and offer the whole thing up. Expect the outer pipes to need a shove; they settle with heat.
Understand what the springs are for. On the sectional pipes, the spring-and-collar arrangement exists so you can pull the pipe for an oil change without disturbing the head seals. Remove the springs, slide the header off the collars, and leave the collars and their gaskets undisturbed on the bike. That's the feature you paid extra for; using it is the point.
Seal the insert joints. More than one member developed a leak at a header-to-flange joint after a thousand miles and cured it with a thin bead of copper RTV on the inserts. Worth doing on assembly rather than after.
You will probably lose the centre stand. The popular four-into-one designs all run where the stand lives. On the MotoGPWerks pipe the secondary comes off with one 10mm bolt and two springs, which gets the stand back for servicing — but as a daily proposition, it's gone. Our earlier piece on stripping weight off a 550 covers the two workarounds worth knowing.
And on the question everybody actually wants answered: does it make power? The most honest assessment in the thread came from someone running the pipe daily. There's probably very little in it, the racers would laugh at any claim otherwise, and the sound biases everyone into believing the bike is faster. What he did notice was that the engine revved more freely — enough that he was overshooting by five hundred rpm on rev-matched downshifts for the first week — plus better ground clearance and a genuinely meaningful weight saving. Those are all real. The horsepower is mostly a story you tell yourself over the noise.
The other half of the decision
Which brings us to the second thread, and to the part of this that nobody thinks about on the afternoon the pipes come off.
It started in October 2013 with a member who had acquired a rare early bike and wanted to put it back to stock, asking whether anyone thought Honda would ever produce the four-into-four exhausts again. Nearly a thousand replies and more than a decade later, the answer is still no, and the thread has become a running ledger of what that has cost people.
The numbers in it are grim reading if you're holding a set of stock pipes and a hacksaw. Back in 2013, one member reported a single near-new muffler — one, not a set — going for $685 on eBay. Reproduction sets from the best-regarded Japanese supplier ran to around two thousand dollars shipped, when they were in stock, which frequently they weren't. Members who had bought genuine Honda sets from a UK dealer during the recession for well under a thousand a set spent the next decade telling each other they should have bought more.
And running underneath the price talk is a thread of memory from the people who worked in shops in the seventies. New bikes coming in and having their factory exhausts pulled off and binned so a four-into-two or a four-into-one could go on, mostly for looks. Racks of takeoff pipes covering two walls of a dealership, hauled to the scrap yard by the pickup load. One member reckoned he'd personally done half a dozen such swaps. In a cold-winter climate the stock pipes rusted out in a couple of years anyway, so nobody thought twice.
There's a useful maintenance note tucked in there too. The header sections rarely rust from the inside; it's the mufflers that die, and the cause is short rides that never get the system hot enough to drive out condensation. One member pulled his original headers after 75,000km and sold them as good pipes. If you're keeping a stock system on a bike you ride occasionally, that's the mechanism to defend against.
The most recent chapter is the one to take seriously if you're planning on buying reproduction stock pipes for anything in the family. An Italian manufacturer took on reproducing the four-into-four systems, ran group buys through this forum, and then took years to deliver. Some members waited eighteen months and more. Some paid and never received. Members who did get theirs reported wrinkled head pipes at the top bend. There were reported quality problems with the CB500 pipes specifically. By early 2025 the business was in administration and up for sale, and the thread's veterans — several of whom had led the group buys personally — had settled on a one-line verdict: buy the Japanese ones, they're worth the money.
None of this is an argument against putting a four-into-one on your CB550. It's a very good modification and it's most of what makes these bikes look the way we all want them to look. It is an argument against the bin.
The through-line
Put the two threads side by side and the exhaust decision separates into two decisions that have nothing to do with each other.
The first is what you bolt on, and the forum's answer is that the pipe is the easy part. The hard part is that a CB550's carburetion is a system — float height, pilot circuit, needle position, mains, sync — and a new exhaust nudges that system without telling you which way. Set the float height to the figure for your carburettor set, not your model year. Sync it. Then work the needle and the mixture screws, which own the throttle openings you actually ride at. Change the mains last, and only if the plugs and the top end tell you to. And don't accept a stumble as the price of a loud pipe — four months of that thread is one man refusing to, and being right.
The second decision is what happens to the pipes that came off, and there the forum speaks with one voice built out of a decade of regret. Box them. Label them. Put them in the loft. The four-into-four you think is ugly and heavy is the single most valuable thing on that motorcycle to the person who eventually wants it back to standard, and the supply is not coming back. Everyone in that thread who is now paying two thousand dollars for replicas is paying it to somebody in 1976 who felt exactly the way you feel this afternoon.
Credit where it's due
None of this is ours. It belongs to the members of the SOHC/4 Owners Club community, who spent four months and a drawer of brass working out that the problem was a float, and a decade keeping a running tally of what the rest of us threw away. Both threads are worth reading in full — the four-into-one jetting thread and the stock exhaust thread — and the forum is worth supporting. There is no substitute for an archive that deep, and no search engine that will reconstruct it for you.
When the pipe is on, the old one is safely boxed, and you're working through everything else the project turned up, that's where we come in. The Thirteen Motorcycles CB550 catalog is built for exactly this stage — parts made for this bike and designed to work together, so the afternoon you spend under it is spent riding at the end of it.
0 comments