Yamaha R3 Forums banner

1 - 20 of 25 Posts

·
Registered
Joined
·
208 Posts
Discussion Starter · #1 · (Edited)
This belongs in its own thread (used to be a reply); this makes it easier for me to locate & update with new information.

This is a compilation of what I've read as well as the theories & analogies I've developed through attempting to understand.
I will be available to debate my claims and I will correct them if I find reason to (for example: you reply debating a section, providing your reasoning & sources of information).

This was an attempt to understand the following things:
  • Order of operation for 4-stroke engines
  • Exhaust gas flow
  • The effects of scavenging
  • Exhaust gas flow & restriction
  • Tuning potential for street & racing applications
  • Effect of exhaust gas velocity at collector
    • Help to explain the potential drawback of Delkelvic YZF-R3 exhaust system O2 sensor placement
Anatomy of an Exhaust System
  • Header (the "primary tube", source)
    • Features flanges which bolts to cylinder head
  • Collector (which joins the header and mid-pipe)
    • Combines exhaust gases from ports 1 & 2
    • Collectors designed to induce venturi effect on exhaust gases are called "Merge Collectors" (source)
  • Mid-Pipe (comes after the collector and before the slip-on)
    • This often bolts to the underbelly, somewhere near the swingarm
    • Common location for O2 sensor bungs (M18x1.5-thread for wideband O2)
  • Adapter (usually a sized adapter between the mid-pipe and slip-on)
    • These adapters are designed to bolt onto stock mid-pipes; so often are found in slip-on kits (not full exhaust systems)
  • Slip-on (or silencer, muffler, resonator)
    • These are often bought via aftermarket for aesthetics
    • These produce negligible performance gains
    • May enhance full exhaust system performance (based on length and modes of exhaust flow restrictions (db killer/baffle))
  • Expansion chamber
    • Some full 4-stroke exhaust systems may feature expansion chambers to modify the effects of scavenging and noise (not pictured)
Anatomy of an aftermarket exhaust (image)

Slip-On kit for stock exhaust (image)

The slip-on adapter for stock mid-pipe will not come with a full exhaust system (a system whose mid-pipe will match slip-on diameter)


The basics of engine vacuum & exhaust flow

Searching for more information (whether it's fact is up for debate), I found intimid8er's thread on MotorcycleForums.net, view it here: EXHAUST--Buyers Research Guide. I edited the original post for clarity and simplification.

First, watch this video to learn how a 4-stroke engine operates.

[Engines operate as vacuum pumps. Outside the engine (at the air filter and at the exhaust muffler outlet pipe) is unequivocally lower pressure than what is inside the engine while it is running. After you start the engine via crank and spark, the engine operates without any further input.]
...
When the intake is opened during its stroke, a pressure differential is created (negative) and air flows in to fill [the cylinder].

[After the compression, combustion, and at the end of the exhaust stroke,] the exhaust valve opens, and a [positive pressure is created]. This rise in pressure is followed by a [negative] pressure in the [exhaust] pipe.

[T]here are two waves that exit the exhaust. One [energy wave travels faster than the other wave]. [This means there are two exhaust pulses for every exhaust stroke.]

When the energy wave[s?] hit a wider part of the exhaust system or the end of the pipe where the atmospheric pressure is greater, it rebounds off the greater pressure and heads back up the pipe with no loss of speed. On the way back up the pipe, it encounters turbulence from the exiting combustion and energy waves, but does not experience any significant decay. If it hits the exhaust valve at the exact time it is closed it simply heads back down the pipe.

[Meaning the two exhaust waves are racing through the header, seeking low pressure. The leading exhaust wave bounces off an expansion of pipe (wider tubing, or baffle) and reverses flow up the header toward the (still open) exhaust valve (this guide does not mention what the slower exhaust wave does). By the faster exhaust wave passing the slower wave, it causes turbulence but does not deteriorate its form.]

Timing the exhaust event to happen after the reversion bounces back again is critical to creating a low pressure event that will help your engine scavenge more efficiently, more or less “suck charging” the engine. This is where diameter of the pipe, length, thermal efficiency, and a ton of other variables come into play.
...
[T]he smaller the diameter of the pipe. The faster the gas can flow.

[By] putting a larger diameter pipe on an engine...you are slowing down the speed of the energy pulses, which again, is slowing down the timing of negative pressure in the exhaust pipe near the valve area, which retards the efficiency [of] scavenging.

As I have shown, the wave arriving at the right time at the exhaust port affects the efficiency of the engine.
...
RPM or wave tuning depending on the length of the pipe affects the timing of the reversion event, and this in turn affects power output.
...
There is no Holy Grail for horsepower and torque, but it is generally agreed a 2:1 system will preserve both and be the most efficient. Also, a “crossover”,“equalizer” or “balance” chamber will equalize and flatten the torque peak or widen the power band by using the volume of both chambers like a 2:1 system.
...
When an energy wave from one primary hits the collector tube in a 2:1 system, the negative pressure behind the energy wave helps pull the waves from the other primary along faster. This reflects the term I used before of “suck charging”.
...
Backpressure is not necessary for an engine to run...Backpressure helps with torque by interrupting the energy waves. Backpressure can be manipulated by mufflers, baffles, and other devices...A properly designed pipe with the correct diameter and length will still be more efficient that an improper diameter and length pipe fitted with baffles or devices to interrupt the reversion waves.
...
Conclusions intimid8er has drawn from his research:
  • Long pipes will increase power earlier in the powerband.
  • Short pipes will increase power later in the powerband.
  • Large diameter pipes will cause you to lose low end torque and horsepower.
  • Small diameter pipes will preserve low end torque but lose upper end horsepower.
  • Crossovers will fatten and flatten (no peaks or dips on a dyno chart) up the torque curve and give a more usable powerband.
  • 2:1 systems will do the same.
  • (Due to using the combined volume of two pipes to scavenge both cylinders versus one pipe: one cylinder.
intimid8er's EXHAUST--Buyers Research Guide

The skinny on exhaust thermodynamics
  • Back pressure may (or may not) be the correct term when discussing exhaust flow restrictions
    • I believe "back pressure (or backpressure)" often represents "negative pressure" (will update when I read otherwise)
  • Exhaust flow is not constant, and sound waves travel multiple directions
    • I haven't found information on whether exhaust gases travel multiple directions
  • Reducing exhaust gas velocity while enhancing torque throughout the rev range by using different diameter pipes, baffles/db killers, ignition timing, etc., is a matter of research, development, and tuning.
Altering exhaust gas velocity
  • Altering the exhaust flow rate may affect the ability scavenging has on providing vacuum to exhaust the combustion stroke.
Negative Positive exhaust pressure
  • Scavenging is responsible for greater torque production in high-performance engines.
  • Scavenging is produced with camshaft overlap, an engineered moment when the intake valves opens just after the exhaust valve, inducing exhaust gas vacuum.
The exhaust gases that exist in the combustion chamber and exhaust during the exhaust stroke are met by camshaft overlap, producing the scavenging effect whereby vacuum from the exhaust flow (high pressure) suck air through the air intake (low pressure), then the cylinder, and out the exhaust port (also low pressure).

Now that I've quote myself and reflected on the explanation above, I feel like I'm not accounting for the butterfly valves in the throttle bodies; does it open via vacuum even when the TPS is zeroed? I'm assuming so.

Without exhaust pressure
  • When exhaust flow is unrestricted,
    • but the intake of air is not increased (airbox mods, velocity stacks, ported throttle body)
    • and fuel supply is not increased (fuel pump)
    • and the ignition timing is not adjusted (advance/retard)
    • (and some other variables I'm not aware of)
      • synergistically eliminates all forms of torque throughout the rev range
This means that unrestricted, high-flow exhaust systems deplete the high pressure in the exhaust system which also produces the exhaust gas velocity needed to maintain the engine's vacuum.

Without scavenging
If the exhaust stroke does not escape the complete mixture of exhaust (traditional 4-stroke without cam lobe overlap), the next stroke has a mixture of hot exhaust gas, thereby reducing the ability of the engine to produce an optimal compression stroke, thereby reducing the efficiency of the combustion and exhaust strokes.

The effect of reducing exhaust gas flow
Reducing the ability an exhaust can produce vacuum (baffles, resonators, expansion chambers, etc.), means less air & fuel is required to operate the motor. By slowing down the exhaust gas velocity, the stock air delivery permits optimal scavenging, thereby producing torque in the low and mid RPM range.

On the contrary, by increase air & fuel on a high-flow exhaust system, this gains torque and horsepower in the top-end of the rev range. (Or maybe I'm too tired for editing right now).

Scavenging for torque
  • AFR (air fuel ratio) is mixed based on the complete air-fuel exhaust mixture at the O2 sensor in the exhaust mid-pipe.
    • To maintain proper AFR mixture, the engine requires exhaust scavenging to complete the intake of fresh air into the combustion chamber while simultaneously efficiently exhausting the exhaust gas from the previous combustion stroke.
  • Without scavenging, torque is lost because the engine has to recirculate its previous combustion strokes (traditional 4-stroke operation).
  • The AFR is mixed based on the complete air-fuel exhaust mixture at the O2 sensor. A more complete AFR combustion produces a piston movement, energizing the crankshaft with torque.
This video shows everything a tradtional 4-stroke engine does


While this video shows what scavenging does (from 2:26)


By scavenging, exhaust gases are always a step ahead of the intake valve. The vacuum caused by exhaust flowing out of the exhaust port and into the header is what pulls the air into the cylinder and out of the exhaust port, further pushing the exhaust gases, priming the motor for the next cycle.

What I learned 21 Jan 21: There are two exhaust waves, each one produced per exhaust stroke. One is faster than the other.
  • The faster of the two waves "bounce" off an exhaust expansion (larger diameter pipe), or a baffle
    • This faster exhaust wave/pulse then fires back up the header pipe,
    • Passing the slower exhaust pulse without deteriorating its volume,
    • And intercepts the exhaust valve
    • This is the sound wave
  • The slower of the two waves is a volume of exhaust gas
    • Exhaust gas travels much slower than sound waves
intimid8er from MotorcycleForums.net writes about how timing this event to the precise moment the exhaust valves close is how intricate and precise design, research, and development of exhaust systems must be to maintain and enhance torque throughout the rev range. See the gist of the thread in the spoiler above (or read the original post by clicking here).

Street applications
  • Plated (licensed) motorcycles for the street must adhere to strict specifications produced by manufacturer R&D (Research & Development).
  • Stock exhaust systems aim to restrict flow enough to produce low-end and mid-range torque for street applications.
  • Air & fuel consumption must meet the flow rate of a highlyrestrictive exhaust to meet emissions laws.
    • Mileage decreases by using aftermarket exhaust systems & tuning
  • Baffles cause negative pressure by reducing exhaust flow rate at specific points in the exhaust system.
    • Reducing exhaust gas velocity on a high-flow exhaust will more appropriately match the capacities of our restrictive ECU tunes
Racing applications
  • High-flow exhaust systems are designed for racing applications.
    • Requires more air and more fuel to properly maintain exhaust and subsequent intake air flow rates via scavenging
As I excerpted in My Notes on Jesse's "dyno test and superbike build", Jesse states:
[O]n the race track, we don’t care so much about the area from idle to 9,000 rpm, and are much more concerned with the power above 9,000 rpm.
Exhaust modification criteria
Source: intimid8er's EXHAUST--Buyers Research Guide (drop him a like on the orignal post)
  • Small diameter pipes will preserve low end torque but lose upper end horsepower.
  • Large diameter pipes will cause you to lose low end torque and horsepower.
  • Short pipes will increase power later in the powerband.
  • Long pipes will increase power earlier in the powerband
Norton-Motorsports.com on R3 exhaust performance
...I, and other tuners have seen over and over again, which is the fact that the R3 performs better with the noise restriction silencers installed in the exhaust, and always seems to be down on power with exhausts that are completely unrestricted in the end.
Yamaha YZF-R3 / MT03 exhaust dyno test and superbike build. How much power does the R3 make?
Balloon scavenging analogy
It's like you poke a hole in a balloon and try to keep it inflated, but the hole is exhausting more air than you can inflate with your breath, this causes the balloon to lose its volume. Losing volume decreases its ability to push air out. Cyclically (and chaotically) starving itself of air, while tiring you out, the balloon deflates and you're too tired to blow air into the balloon.

This is an analogy as to how a high flow exhaust will not permit the build-up of pressure in the balloon (cylinder), because the exhaust is letting more air out than what you're able to put in.

I am missing something here, why is the bung hole location not optimal with the Delkevic? Can the bung hole register O more accurately further downstream of the collector? Is there data to support that?
Collector intersection analogy
I imagine a collector as a busy intersection just before entering a tunnel, where two lanes are merging into one, and there are no stop signs or stop lights.

From one of two terminals, we fire off toward the intersection. For split-seconds at the intersection, we take a look at the other incoming lane and then find ourselves jumbled together, because we like to rubberneck (looking where we don't need to be looking).

Realizing we're both hot as ****, we want to seek that low pressure, cool air; using our combined velocity, we shoot for the tunnel exit. At some point after the intersection, we're completely merged as one pulse.

It's at this point we take a photo together, relishing on the peace of our togetherness - accepting the past was turbulent and uncertain. This is where we should capture an O2 reading, a clear, decisive AFR. The point at which velocity is up and turbulence is down. (somewhere after the collector, closer toward the slip-on).

Collector jumble
  • Exhaust gas flow may be turbulent up to and including the collector.
  • Collector merges the hot gases into a cohesive flow thereafter (in theory).
  • Placing an O2 sensor at this junction may cause sporadic O2 sensor readings

Delkelvic collector
The Delkevic places the O2 bung just atop the exhaust collector. From what I've read, this could cause sporadic O2 readings. This could easily be capped and a new one be installed on the mid-pipe.
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #4 · (Edited)
Notes excerpted from OnAllCylinders.com
Author: Wayne Scraba
Title (Part 1): Header Theory (Part 1): Looking at the Science Behind Exhaust Header Tuning
Title (Part 2): Header Theory (Part 2): Looking at the Science Behind Exhaust Header Tuning

Part 1: Header Design
Header design can affect two types of scavenging, known as "intertial" and "wave" scavenging.

Wave Scavenging specs:
  • Wave as sound
  • Being mass-less, waves starts before intertial gases
  • Based on many variables, waves travel at 13,397in/sec (inches per second)
  • A negative shock wave is generated once wave reflects off collector; wave travels toward exhaust port (place of origin)
Inertial Scavenging specs:
  • Inertia as gas
  • Being volumetric, intertial starts after sound waves
  • The best power is produced at 300fps (feet per second)
  • Flowing through exhaust, inertial gases cool & slow
  • Inertial flow produces low pressure in its wake
What the header builder has to do is time the arrival of this negative pressure wave to occur just before the exhaust valve closes and while the intake valve is opening. If the negative wave arrives too soon or too late, the power potential of the next combustion chamber cycle is diluted.
...
[If low pressure from intertial gas coming down exhaust port 1 travels up to exhaust port 2, this high pressure may help scavenge the next exhaust cycle from exhaust port 2.]
...
By manipulating the diameter and length of a specific tube, a header builder can influence the size of that low-pressure front.
...
Reher-Morrison Racing Engines tells us there is a direct relationship between the diameter of the primary header tube and the exhaust velocity:

“The key when selecting a tube diameter is to find a happy medium between the free-flowing characteristics of large tubes and the superior scavenging of small, high-velocity tubes..."
...
Hooker notes that by varying the length of the primary tube, you physically change the time it takes for the vacuum pulse (low-pressure area) to reach the header collector.
...
...the proper header tube size will allow you to obtain a superior scavenging effect with an increase in exhaust speed.
...
To a point, a header manufacturer can control the speed of the exhaust gas by changing the diameter of the primary tube. The smaller the diameter of the tube, the faster the exhaust will flow (keeping in mind the exhaust flow slows as it cools). A clever header builder will recognize that by varying both the length and the diameter of a given primary tube in a header, those tubes can be tuned to provide the largest amount of inertial scavenging.
...
“At lower engine rpm, long tubes help maintain good exhaust scavenging and increase torque output...It is also very important that all header tubes are as close as possible to the same length...
Part 2: Collector Design
According to Hooker: “The major objective of the collector is to take advantage of the ‘secondary tuning impulses’ from the other pipes in the system. The influence of the collector is generally restricted to the rpm level below the torque peak of the engine in question. Collector sizing is more critical in engines that spend the majority of their time in the lower rpm ranges.”
...
Try-Y (4-2-1) headers collect the exhaust by pairs of cylinders that are as far apart as possible in the firing order. The idea is the gases racing past the silent tube will scavenge a vacuum in it, reducing exhaust reversion. We have found no advantage in this design for the majority of racing applications.”

David Reher added a caveat to the above.

"...it may be worth testing a properly designed Try-Y, particularly in applications with lower engine speeds,”
...
The headers need to compliment the engine you build.
...
the diameter of the collector affects the exhaust restriction and velocity.
...
For example, Hooker tells us that a merge collector minimizes the reflected wave, so it has less effect upon the inertia pulses. The actual hourglass shape of this collector design serves to regulate the high and low pressures. It also tends to speed up the velocity of the gas flow. Additionally, the shape of the collector (particularly in the case of a merge collector) helps equalize the transition of the individual exhaust tube gases.

The overall length of the merge collector is important, as is the angularity of the tubes that physically “collect” in the collector. You’ll also find a “spear” or “pyramid” inside a merge collector. The purpose of this pyramid is to control the exhaust gases so that they do not re-enter the primary tube and travel backwards into the engine by way of the low-pressure area (exhaust gas dilution).

The neck of the hourglass is important too. By reducing the size of the neck, the exhaust gas velocity will increase as it is carried into and out of the secondary tube (collector). This form of collector can prove efficient; some folks have experienced an actual increase in the EGT (Exhaust Gas Temperature) in certain applications. The reason is the header assembly is extracting more of the air/fuel mixture out of the cylinder and likely burning it in the collector...
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #5 ·
Notes excerpted from HotRod.com
Author: Matthew King
Title: What's a merge collector?

..a merge collector, also called a venturi collector, works on the opposite theory: necking down to a very small diameter before opening back up to the larger size of the exhaust system.
...
The concept of a merge collector is similar to the theory behind a carburetor venturi, and in both cases, they work on the principle of building velocity through the venturi effect. Air passing through an hourglass-shaped venturi accelerates as it passes through the narrowest part of the venturi then slows down and expands as it emerges on the other side. This expansion creates a pressure drop, in effect siphoning air into the low-pressure side of the venturi and drawing more volume through the venturi.
...
...the venturi effect improves exhaust velocity and scavenging, which has effects on cylinder filling, the intake tract, and camshaft timing.
...
Merge collectors are designed to minimize th[e] change in area at the transition from the primaries to the collector, which allows [inertial exhaust] gasses to maintain the greatest possible velocity.
By increasing exhaust gas velocity, other systems (such as air intake, valve timing, and exhaust header tuning) must be accounted for.

Dynatech Merge Collector


Dynatech cautions that it's possible to "overscavenge" the exhaust when switching to merge collectors. This is why the company suggests additional camshaft testing may be necessary to get the best results from a merge-type collector.
...
..it's not unheard of for a merge collector to initially lose horsepower or torque if it upsets the balance between the camshaft timing and engine airflow.
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #6 · (Edited)
Bullet-point note-taking makes details easy to remember, but to get a full understanding, follow the links to original articles.

This post will expand on design exhaust flow theory, baffles, and (chambered) resonator design theory.

On flow simulation
Article from Motordyne.com
Transient Analysis of Exhaust Headers Using Computational Fluid Dynamics In Solidworks Flow Simulation

Introduction (keywords)
  • flow resistance
  • synchronizing acoustic resonance
  • charge air dilution
    • "In some engines, for emissions reasons, exhaust gasses are intentionally left in the cylinder for charge air dilution to reduce combustion temperatures and smog forming NOx." (Source: Motordyne)
    • Yamaha R3's "charge air dilution" system is the AIS or Air Injection System
  • intake charge management
    • ...introducing relatively high temperature exhaust gas to intake can impact the temperature and composition of...air supplied to the combustion chamber. [For] proper functioning of an engine with [AIS (Yamaha R3)], components such as [reed valves and charcoal canister] have to be introduced to control the flow, temperature and distribution of [emissions following air injection].

      As can be imagined, introducing relatively high temperature exhaust gas into the intake air can have significant impacts on the temperature and composition of the combustion air supplied to the combustion chamber. In order to ensure proper functioning of an engine with EGR, various hardware components, such as valves and coolers have to be introduced to control the flow, temperature and distribution of EGR supply and the resulting mixture with intake air.
      Source: dieselnet.com

      Watch this video to understand how AIS works:
  • mass flow rates of exhaust
Exhaust Pulses and Scavenging (edited for concepts)
  • Low-pressure scavenging is brief
    • Measure duration of sound wave from (exhaust) port 1 to port 2
      • Utilizing collector or crossover after primary tubes
  • Long tube headers are a "quarter wave acoustic resonator"
  • Resonant frequency determined by primary tube’s length and diameter, including exhaust gas temperature, molecular weight, and ratio of specific heats (gamma)
  • Long tube headers generally start with small diameter primaries close to the engine with gradually increasing diameters as distance from the engine increases
    • Maximize exhaust velocity without creating excessive pressure drop
Solidworks Flow Simulation
  • Solidworks Flow Simulationis a computational fluid dynamics (CFD) program with the ability to simulate liquid and gas flows
  • In order to model the time dependent nature of exhaust pulses in exhaust tubes, a transient simulation must be used.
  • nested iterations must be used to properly capture transient compressibility effects
    • transient shock waves
    • acoustic waves
On baffle design
How Exhaust Baffles Work - ItStillRuns.com

On resonator design
teriks shared links on the FMF Power Bomb patent designs (click link to see discussion) regarding expansion chamber designs for 4-stroke engines. I linked keywords for further reading:
A question on R&D of designing products
What it makes me wonder is, do these guys have some way of directly measuring exhaust pulses. If they did, they could theorize and experiment and measure results.

[Answer is yes: Solidworks Flow Simulation]

If the dyno is their only measuring device, that would make the development process pretty difficult and time consuming. Maybe they have FEA software that can handle those weird configurations? After all, it is just holes and chambers with air as the flexible medium. If you got an FEA model to "work", you could experiment with stuff like chamber configuration and hole configuration at various pulse frequencies.
NemadjiMan
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #7 · (Edited)
I'm going to be reading about Solidworks Flow Simulation next.

I will update this post with the notes of my research and list any alternatives to SolidWorks that I come across, because let's face it - SolidWorks wasn't developed for mere hobbyists.


Maybe someone else can chime in on Solidworks Flow Simulation - I'm not the guy.
 

·
Registered
Joined
·
300 Posts
File this under "caution: this is harder than it looks".

There's a chap in New Zealand who has posted, on youtube, countless dyno runs concerning various mods to a Toyota 4A-GE engine (1600cc 4 cylinder DOHC-4v, fuel injected, normally aspirated). This is the engine that was in the high-performance version of several 1990s-2000s era Toyota cars (Toyota MR2, etc). NO that is not a Yamaha R3, but that's not the point. It's a 4-valve-per-cylinder DOHC fuel injected 4 stroke engine without forced induction. Our engines have smaller (and fewer) cylinders and spin faster but the fundamental concept of operation is the same.

On his youtube channel, you'll find all manner of experiments with cam timing, air-fuel ratio, intake runner length and diameter, exhaust system configuration and all manner of other things. But this one - just posted today - is to drive home the point of "choose your parts wisely, because mismatches don't work".

Among today's experiments included ... exhaust header pipes, and exhaust tailpipe diameters, that were too big for the engine ...

Bear in mind that the total power that an engine makes is very close to being in proportion to the mass flow-rates of air and fuel through it (and the mass flow-rate of exhaust = the mass flow-rate of the air plus the mass flow-rate of the fuel, so that's in proportion, too). And the speed of sound is a function of the temperature ... but under analogous operating conditions, the exhaust temperature from his 400cc-per-cylinder engine won't be much different from our 160cc-per-cylinder engines, so certain things (intake runner cross-sectional areas, exhaust header-pipe cross-sectional areas, etc) will tend to vary in proportion to horsepower per cylinder, and other things (intake and exhaust header-pipe LENGTHS) will tend to vary inversely-proportion to the target operating RPM range.

For what it's worth, the engine that he is working with has a higher BMEP (brake mean effective pressure - an indication of how effectively it is tuned) than any R3 engine that I have encountered in any state of tune ... so don't come complaining "it's a car engine" ...

If you are interested in this stuff, there are hours of videos on that guy's channel ...
 

·
Registered
Joined
·
300 Posts
Another small thing. In the R3 world, Jesse Norton's article and website is a wealth of information (and I have a bunch of his parts on my race bike): Dyno Testing and Tuning

I will also grant that his eventual peak power number is a little higher than what I've seen on mine. Different dyno, though. I have a rulebook that I have to follow (for the roadracing organisation) which limits some of the things that I can do.

He ran into a problem in which there was some sort of mismatch leading to a big hole in the powerband around 10000 rpm, which is absolutely not good on a race bike. His solution: Bigger throttle body diameter (changes the way the intake runner tunes because it affects the way the pressure waves bounce around inside), bigger fuel pump, the aRacer ECU, $$$. Aside from $$$, I cannot do that, have to run stock throttle bodies and fuel pump and ECU hardware because rulebook. My solution: reconfigure something on both the intake and exhaust sides that dealt with (i.e. got rid of) the flow reversion that was causing that. "For otherwise similar state of tune with stock throttle bodies and stock ECU hardware if not programming and stock fuel pump" ... I've got a little more peak power, a little more peak torque, and a flatter torque curve than he got (Disclaimer, different dyno)... And at that point, I shut up.
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #10 · (Edited)
You always bring interesting points to discussions, @GoFaster. I hope there are more lurkers out there who will contribute their perspective they way you have - it's admirable.

And the speed of sound is a function of the temperature ... but under analogous operating conditions, the exhaust temperature from his 400cc-per-cylinder engine won't be much different from our 160cc-per-cylinder engines...
Based on my limited research, I understood sound was a function of vacuum, and temperature was a function of inertial gas (combustion). But perhaps I'm misunderstanding the science or connections between sound and temperature (I likely am, and I will include the concept on my search for understanding).

the exhaust temperature from his 400cc-per-cylinder engine won't be much different from our 160cc-per-cylinder engines
I feel this is correct, but my understanding is it's based on many variables. But unless we're talking about the differences of diesel versus gasoline engines - yes, I think you have a point here.

In the R3 [racing] world, Jesse Norton's article and website is a wealth of information
Jesse Norton's website is geared more toward racing applications, as his articles clearly discuss performance modifications for top-end horsepower...I actually read the superbike build thread and took notes, then created a post using the notes: My Notes on Jesse's "dyno test and superbike build".

Granted, my notes don't include the science or racing-only applicable information, but I feel like I did a good job of summarizing it for street applications, noting the major differences between what makes them different.

I will never forget this quote on racing:
[O]n the race track, we don’t care so much about the area from idle to 9,000 rpm, and are much more concerned with the power above 9,000 rpm.
My Notes on Jesse's "dyno test and superbike build"
It supports the fact that you mentioned he stressed with the issue at 10,000 rpm. Clearly, this is not confidence-inspiring when you're racing and never drop below 9,000 rpm.

You may be referring to loss of mid-range using the Yamaha GYTR racing kit, as he mentioned in the article.
*Racing kit linked is not the one Jesse reviewed, but linked for clarity - it is not a Norton-Motorsports product.

My solution: reconfigure something on both the intake and exhaust sides that dealt with (i.e. got rid of) the flow reversion that was causing that. "For otherwise similar state of tune with stock throttle bodies and stock ECU hardware if not programming and stock fuel pump" ... I've got a little more peak power, a little more peak torque, and a flatter torque curve than he got (Disclaimer, different dyno)
You have developed trade secrets, then (I would say), with your ability to make as much power without Jesse's modifications. One thing I do not understand is what you mean by "different dyno".

If this has anything to do with Correction Factor STD, then this is also something Jesse brought up in the article to say it doesn't read-out as accurate a figure as CF SAE.
 

·
Registered
Joined
·
300 Posts
Jesse discusses this in the first part of his article. His dyno is HIS, located in San Diego CA (USA) near sea level, in a temperate-to-warm climate. The one that I used is (a) not the same one although a similar model, (b) located north of Newmarket ON (Canada) a few hundred metres above sea level, (c) tested in February although on a day that was warm (ish) for February (IIRC about 10 C). You can't honestly make direct comparisons unless you do the testing back-to-back on the SAME dyno in the SAME place under the SAME environmental conditions (temperature, barometric pressure, humidity) at the SAME coolant and oil temperatures and so forth. The correction factors for these conditions can only do so much - the accounting is almost guaranteed to be imperfect. It's also pretty likely that our test procedures were different, and certainly the bikes are different, probably different oil in the crankcase, probably different rear tire, probably different tire pressure, probably different chain, and on and on.

I'm still mildly frustrated at not understanding why the BMEP is so low. Mine's better than stock, but it's not in the range of what a well-tuned engine of this type should be, and I've yet to see any R3 engine that is. The key cylinder dimensions are all within a mm or two of those of an R6, but it doesn't make half the torque of an R6 (which is what one would expect with half the number of cylinders that are pretty close to the same size). Something's holding it back, and I've only found part of it, not all of it.
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #12 · (Edited)
On tuning environment variables
I wrote about this very thing in my ECU tuning write-up (which I've started the process of rewriting).

With regard to developing a better machine, the Z-AFM module helps your motorcycle become the best it can be within the limitations of your riding conditions:
  • Elevation
  • Season (time of year)
  • Atmosphere (temperature, air density, humidity, precipitation, etc.)
  • Riding style (spirited, casual, commuting, racing, etc.)
  • Performance modifications (air box, air filter, exhaust, cams, etc.)
...a bike at sea level with 70 degree air and 0 percent humidity will always require more fuel and make more power than a bike at 6000 feet elevation with 90 degree air and 90% humidity.
Yamaha R3 / MT03 Exhaust Dyno Test and Superbike Build
On the BMEP
Reading from sciencedirect.com,
[BMEP] is artificial and superfluous as it is derived from measurements taken by a dynamometer (or brake), which are then used in the calculation of mechanical efficiency.
So this BMEP figure is highly dependent on which dynamometer being used to take measurements.

If I recall correctly there's a bit in his superbike article about a thing he wouldn't share (because it's like a trade secret), that pertained to his performance increases, because revealing it would give his competition the upper-hand.

Their website just broke the old link to the superbike article (Error 404): https://www.norton-motorsports.com/...p-on-and-stock-exhaust-tuned-with-bazzaz-zfi/

Here's the new link to the Norton-Motorsports superbike build article: Yamaha YZF-R3 / MT03 exhaust dyno test and superbike build. How much power does the R3 make?
 

·
Registered
Joined
·
300 Posts
I thought of another difference between Jesse's situation and mine. He would have been using California-spec oxygenated 91 octane petrol. I'm using Canadian-spec Shell V-power 91 octane which does not contain ethanol. My handy bike-mounted real-time air/fuel ratio display shows that there is a measurable difference between Shell 91 and PetroCanada 94 (which is loaded with ethanol - that's how they got the 94 octane rating). N.B. these octane numbers are the North American (RON+MON)/2 convention and differ from those listed elsewhere in the world. Bottom line, you have to tune on the fuel that you plan to use, and stick with it unless not possible. The "pump gasoline" specified in Canadian Superbike (CSBK) which all competitors have to use in that series, is known to be Shell 91.

Your reference in which you found a definition of BMEP glosses over its importance. Yes, it's derived from the dyno measurement but describing it as "superfluous" is absurd. It is an extremely important engineering parameter that represents a combination of how well an engine "breathes" plus how efficiently it turns what it breathes into power.

For a naturally-aspirated 4-stroke-cycle spark ignition piston engine running on petrol (no exotic fuel chemistry) ... The most exceptionally well-tuned examples all have a BMEP very close to 15 bar. A Formula 1 car (before the forced-induction era!) may seem very different from a NASCAR's pushrod V8 - very different RPM range, very different valve actuation, but they are both exceptionally highly developed - both have maximum BMEP around 15 bar. The premium production superbike engines (ZX10R, R1, GSXR1000, etc) are around 13.5 bar, which is quite exceptional considering that these have OEM catalysts and quiet mufflers. The bone-stock production Chrysler 3.6 litre Pentastar V6 in my van, going by the factory ratings, works out to 12.8 bar BMEP (this is also a DOHC 4-valve engine, which also has variable valve timing, which extends the RPM range at which this torque can be achieved).

And yet ... the R3 struggles to reach high 11 bar BMEP. Stock, they're 10.something.

Of course, the BMEP is not the only story; at first glance it may appear that the humble engine in my van is 85% as good as a Formula 1 car engine ... but of course, the F1 car is doing that at 17,000 rpm to get the car around the racetrack, and the one in my van is doing that at 2000 rpm to cruise on the motorway and 3000 rpm to get up a hill. Very different applications ... and yet the BMEP is not hugely different ... because still, they are both naturally-aspirated 4-stroke spark ignition piston engines running on petrol; the thermodynamic and fluid-mechanic principles underlying their operation are exactly the same, they are both playing by the same thermodynamic rules ...
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #14 · (Edited)
Yes, [BMEP is] derived from the dyno measurement but describing it as "superfluous" is absurd. It is an extremely important engineering parameter that represents a combination of how well an engine "breathes" plus how efficiently it turns what it breathes into power.
I apologize - I have done it before, where I take the first explanation to something and refer to it as fact when I'm in a hurry. It's problematic when I don't have the time to sit down and really absorb information and cross-examine definitions from multiple sources. That carelessness, is the biggest downfall of internet research.

I found a video on youtube describing BMEP by Engineering Explained.

If I understand correctly, torque is for publicity, while BMEP is what engine-builders & enthusiasts want to know.
Basically, the R3 isn't producing efficient torque to compete with the BMEP ratings of your examples.

Using ConvertUnits.com, 321cc is 19.58ci.
Then I get claimed torque at 9,000RPM of 21.8 ft-lbs (wiki).
And plug them into this BMEP calculator and I get 11.58 Bar for the R3.

From what I can tell - you're correct about BMEP; it will vary depending on many variables, but that also, the R3 seems "low". Is it possible Yamaha intentionally designed it this way for beginner riders?

Low BMEP means low torque, right?
So whichever way more torque could be produced at 9,000RPM, where (correct me if I'm wrong) the R3's peak torque is (at 9,000RPM) in stock form.

So a finely-tuned, highly-specialized exhaust may positively affect torque, right?

I'm not an engine builder; perhaps you know more about the engineering aspect of what could be causing the R3 to be inefficient.

I went back and read your response. The third time I read through it - it clicked.

He ran into a problem in which there was some sort of mismatch leading to a big hole in the powerband around 10000 rpm...My solution: reconfigure something on both the intake and exhaust sides that dealt with (i.e. got rid of) the flow reversion that was causing that.
...
"For otherwise similar state of tune with stock throttle bodies and stock ECU hardware if not programming and stock fuel pump" ... I've got a little more peak power, a little more peak torque, and a flatter torque curve than he got (Disclaimer, different dyno)... And at that point, I shut up.
Where you mentioned your solution was eliminating flow reversion, I recalled the section I wrote on this post regarding resonators.

One thing I don't see on most exhausts for the R3 is an anti-reversion chamber. However, the conical flow seems to be engineered into the stock exhaust (from the head exhaust port of ~28mm to the slip-on inlet of ~31mm).

I'm interested in merge collector design - I think that could be another point of where the R3 could pick up some torque, and I'm sure that's another thing that the aftermarket is racing to develop.

Regardless of the reversion the occurs, the BMEP will still read low, wouldn't it? Or are you saying that once you designed a custom exhaust and eliminated the flow reversion, you were able to gain a greater BMEP?
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #15 · (Edited)
Response conclusion
  • Jesse's 10,000RPM dip was claimed to be at the fault of the throttle bodies not inducting more air, but you understood the issue and resolved it by modifying the intake & exhaust (good on you)
  • Reversion may not be related to merge collectors
  • Merge collectors won't positively affect BMEP
  • A finely-tuned, highly-specialized exhaust may positively affect scavenging, while enabling us to maintain the stock fuel pump and throttle bodies
Quoted some interesting points @GoFaster made about fuel mixture & BMEP:
...you have to tune on the fuel that you plan to use, and stick with it unless not possible. The "pump gasoline" specified in Canadian Superbike (CSBK) which all competitors have to use in that series, is known to be Shell 91.
Agreed.

On BMEP and the R3's inefficient bar standard; 10-something:
The most exceptionally well-tuned examples all have a BMEP very close to 15 bar.
And yet ... the R3 struggles to reach high 11 bar BMEP. Stock, they're 10.something.
Intriguing & eye opening to beginning tuning enthusiasts. I know for fact this information will be helpful to those who are on the brink of understanding air & fuel delivery.
 

·
Registered
Joined
·
208 Posts
Discussion Starter · #16 · (Edited)
For street applications...

After all the things I've read - an ECU might be a good choice for street applications, minus all the other recommended performance mods.

I think the most benefit we can get from tuning the R3 is by:
  • Use aRacer RCMIni5, AF1, and a smartphone to monitor AFR whilst test your tune changes
  • Leave everything else stock (including air filter, airbox, velocity stacks, engine, slip-on, etc.)
I would do this - as I've learned even altering the stock slip-on is detrimental to performance, thanks to @kiko who bit the bullet and proved it. I would even entertain packing the hollow stock slip-on with quality packing material, or purchase a nice big slip-on with a restrictive baffle.

There was no visible packing anywhere that I could see.
Products I'm referring to:

SpeedTek aRacer products page:
SpeedTek aRacer RCMini5 links:
SpeedTek aRacer AF1 Autotune links:
 

·
Registered
Joined
·
33 Posts
I think for those of you that race the R3 all the above info is a very helpful learning resource for those of you that take the time and make the effort using it and learning how to apply it on the R3 will only benefit from what you learn as you move up to increased displacement classes.

For most begining riders and experienced riders that just want to enjoy a lightweight small displacement sportbike the R3 is fine as is, I only added a slip on for a little more sound and took another forum members advice and bought a cheap one (musari) rather than a name brand like Akrapovic . And I installed a fender eliminator for looks. So I think I am done with mods.

Experienced riders generally agree that 75hp is adequate, 100 hp is entertaining and anything above 125hp up gets into thrilling. I enjoy the R3 for what it is. It is a excellent lightweight small displacement sportbike and can be huge fun on tight twisty roads even at 37hp. It is a great tool to learn the fundamentals of riding for new riders and lots of fun for experienced riders.
 

·
Registered
Joined
·
67 Posts
I'm still mildly frustrated at not understanding why the BMEP is so low. Mine's better than stock, but it's not in the range of what a well-tuned engine of this type should be, and I've yet to see any R3 engine that is. The key cylinder dimensions are all within a mm or two of those of an R6, but it doesn't make half the torque of an R6 (which is what one would expect with half the number of cylinders that are pretty close to the same size). Something's holding it back, and I've only found part of it, not all of it.
Aloha, I too, have spent some time with my calculator comparing output of R6 - R1's, Suzuki's and different hp increasing mods and claimed hp results and wondering why an R3's with the Norton, TST, and other racing talent why only 5-6 and up to 49 hp is achieved? Whereas the double size motor of the R6 gets (116.8 bhp) @ 14,500 rpm (claimed). (the R3 should be at 58 hp in stock form) even maybe explaining the higher RPM, higher compression, Cams of the R6, that does not make up the discrepancy. What do you think?



 

·
Registered
Joined
·
208 Posts
Discussion Starter · #19 ·
Aloha, I too, have spent some time with my calculator comparing output of R6 - R1's, Suzuki's and different hp increasing mods and claimed hp results and wondering why an R3's with the Norton, TST, and other racing talent why only 5-6 and up to 49 hp is achieved? Whereas the double size motor of the R6 gets (116.8 bhp) @ 14,500 rpm (claimed). (the R3 should be at 58 hp in stock form) even maybe explaining the higher RPM, higher compression, Cams of the R6, that does not make up the discrepancy. What do you think?


I think it's merely the science behind volumetric throughput. Perhaps the nature of airflow succumbs to the highest rate of flow that a design permits (cylinder displacement).

That is to say a large displacement motor more naturally permits a rate of flow which the physics of naturally-aspirated motors operating in regular atmospheric pressure environments allow. I'm not a scientist and this could be bunk.

Small motors are already maxed out, but usually tuned for a progressive powerband for novice, street riders.

Whereas middle and large displacement motors are actually highly governed to remain street legal and rideable.
 

·
Registered
Joined
·
67 Posts
OK, I tend to disagree ^ -- if we look at a SMALLER MOTOR
KAWASAKI NINJA ZX25R
which is more a of a setup compared to an R6. it runs at 50hp stock. If we increased the ZX25R up to 320cc (28% larger) then making the HP 28% more (50hp x 28%) 64hp for a theoretical Ninja ZX320R. OK, lets back up to 61hp ZX320R. and compare it to a R3 setup. 61hp vs 36hp R3. Where is that extra HP coming from? Differences are 4 cyl vs 2 cyl and rev 17,000 (cam) vs 12,000. rpm. Do those two differences make the 25 hp difference? I know the contributing secret sauce is that Kawasaki made the ZX25R with power in mind rather than just power AND streetability, but there has got to be more to the equasion.
 
1 - 20 of 25 Posts
Top