Ninja 400 Riders Forum banner
1 - 20 of 55 Posts

·
Registered
Joined
·
521 Posts
Discussion Starter · #1 ·
What this posting is about

The purpose of this posting is to explore the potential for using an Oxygen Sensor Eliminator as a moderate Air-Fuel ratio (AFR) improvement device on our emissions-equipped 400s.

Warning - I am writing here about a topic I am NOT experienced on. I am basically examining the potential for using what is commonly called an Oxygen sensor eliminator as a tuning aid to moderately improve the air fuel ratio (AFR) on a Ninja 400 / Z400 that has an AFR that is somewhat lean as a result of modifications that might include a less restrictive exhaust and a higher flow intake.

If, like me, you have done any research on the specifics of O2 Sensor Systems on motorcycles, you already know that ACCURATE and DETAILED information on them simply does not exist in the public space - especially when it comes to the specifics of any one motorcycle model.

So, I freely admit here that my knowledge base for this posting is necessarily shallow. if anyone here on the forum can see significant errors or omissions in this posting, please, DO post any better information that you might somehow have gotten from a reliable and truly informed source - ideally someone who works in the industry on fueling systems and emissions.


This posting provides the following information:

  • What is an Oxygen (“O2”) Sensor System
  • How an O2 Sensor System Apparently Typically Works
  • Why a Modified Ninja 400 / Z400 Might Need a Tuning Device
  • Why an O2 Sensor Eliminator MIGHT Be Useful for tuning AFR on a modded Ninja/Z 400
  • What is an O2 Sensor Eliminator and how does it work?
  • Why an O2 Sensor Eliminator must only be used on a bike that has had its catalytic converter (“Cat”) removed
  • One thing to watch out for on any brand of O2 Eliminator
  • Actual testing of this theory and the results (The results are good)
  • An invitation for Feedback

I apologize for the length of this posting, but in my defence, it is really not possible to explain this all properly otherwise. If you have a short attention span, don’t need to understand how and why something is happening, and just want to see “the results”, go to that section directly.


What is an Oxygen (“O2”) Sensor System

An O2 Sensor System is a relatively primitive emissions system, installed in the vehicle exhaust, that helps a motorcycle (or car) stay in compliance with emissions laws.

An O2 Sensor system exists on many motorcycles and virtually all cars for primarily 2 reasons:

The first reason is to maintain the engine’s AFR within a narrow range near 14.7 AFR for emissions regulations reasons, at any combination of rpm and load that falls within a range defined by emissions laws. Over-simplifying, in practical terms, this means (a) whenever the engine is operating at an rpm that is both steady and moderate, and (b) whenever the amount of throttle being applied is modest. Whenever the vehicle is operating in that manner, the O2 Sensor System is active and AFR is as a result very closely controlled and held around 14.7. This is called “closed loop” operation.

A process that is “closed loop” is one in which a control command is not just sent out, but rather is sent out and then been verified as to its actual effect.

The reason that the O2 Sensor System is called “closed loop” is that the ECU is no longer simply sending a signal to the fuel injection system based on the fuel map. It still sends that signal, but then “listens” to the signal coming from the O2 Sensor in the exhaust, and adjusts the fueling (“closes the loop”) if the O2 signal is saying that the AFR is too rich or too lean compared to the ideal 14.7 AFR that minimizes emissions.

Naturally, if there are any errors or tolerances in the system (and there are!), the OEM emphasis is on leaning the mixture versus richening it, because a leaner mixtures is cleaner from an emissions perspective and is therefor less likely to fail an emissions test.

For operation outside of the conditions stated above, the ECU ignores the O2 Sensor signal, as operation in conditions outside those stated is exempt from emissions regulations. The exemption is because engines under high rom or high load conditions would be damaged by a lean mixture, and because The Government agrees with vehicle manufacturers that operation outside of the closed loop conditions accounts for only a small percentage of a vehicle’s usage. i.e. We street (not racing) users normally spend much more time at low rpm or cruising under moderate load, than we do at high rpm or accelerating hard.

The second reason for the O2 Sensor System is to protect the catalytic converter (“Cat”). Cats cannot survive very long in an exhaust environment of unburned residual fuel - the kind of environment produced by the richer AFR that makes the most power in an engine. By ensuring that the AFR stays at 14.7 or moderately leaner, the O2 Sensor System protects the (expensive) Cat from overheating and gradually melting. Since a working Cat is required to meet emissions regulations, and since an overheated and plugged up Cat will severely impede engine exhaust flow and therefor engine operation, protecting the Cat is very important.


How an O2 Sensor System Apparently Typically Works

O2 Sensors as used in cars and motorcycles, come in several versions. The version used on our 400s is a “4-wire” version. These 4 wires provide a reliable ground, power a heater circuit, and output a voltage signal from the O2 sensor .

O2 sensors only work once they are heated up to a certain temperature. The heater makes the O2 sensor get to that temperature much sooner than waiting for the exhaust airflow to heat it up after each cold engine start. This is important because of the way emissions testing is done in most jurisdictions that do testing. A delayed O2 sensor response will cause an emission test failure.

The dedicated ground wire is required because using the exhaust itself as a ground would potentially degrade the quality of the signal.

The sensor’s output signal is routed to the vehicle’s ECU. It tells the ECU, in a rather approximate way, whether the exhaust flowing past the sensor is the result of a “rich” or “lean” mixture. I said “rather approximate way” because really good O2 sensors are very costly, while the ones used in cars and motorcycles are pretty low quality and cheap (despite the THREE DIGIT price your friendly Kawasaki dealer will charge you for a replacement!).

The good ones are called “wideband” O2 sensors, and they can actually produce an output voltage that is predictably proportional to the richness or leanness of the current AFR in an engine.

The cheap ones used in cars and motorcycles, especially on low cost cars and motorcycles, have to be cheap to keep the price of the vehicle sensible. They are called “narrow band” sensors. They work more like a switch than a rheostat: They will basically only tell the ECU that the mixture of components in the exhaust at this moment (which is way later than the moment that mixture was created in the combustion chamber), is either “rich” or lean”.

So the typical cheap motorcycle O2 sensor is physically incapable of being used as a tuning device that would ideally provide ongoing “accurate correction” of the AFR after the ECU has already used the fuel map to dictate the amount of fuel being sent to the combustion chamber(s).

Instead, the O2 sensor forces the ECU, whenever the vehicle is in “closed loop” conditions (again simplified, that means at steady speed and/or low rpm), to iteratively keep adding or subtracting fuel from the mixture being sent to the combustion chambers until the O2 Sensor signal back to the ECU is “acceptable”. Given the cheap O2 sensor’s measurement capabilities, the mixture never attains stable acceptability. It keeps jumping around the 1.47 AFR target.

Also, notice that the emissions regs require the target to be 14.7 AFR. But, most performance motorcycle and car engines run much better with AFRs in the 12.8 to 13.5 region. That range provides good throttle response, more peak power, and cooler operation.

So, because of this unsophisticated O2 sensor iterative lean-rich cycling, and the rather lean “target” 14.7 AFR that the ECU faithfully tries to produce, closed loop operation of many motorcycles, especially at low rpm, can be rather unpleasant.

Note also that not all motorcycles have been equipped with O2 sensor systems. In fact, as recently as just a few years ago, many of the Kawasaki sportbike models had no O2 sensor. Their ECUs had to be tuned even leaner than a sensor-equipped motorcycle needed to be tuned, because there was no way of correctively leaning the AFR because there was no onboard post-combustion measurement of AFR. Subsequent tightening of emissions rules has driven most (all?) motorcycle manufacturers to incorporate O2 sensors.


Why a Modified Ninja 400 / Z400 Might Need a Tuning Device

An unmodified Ninja 400 / Z400 produces pretty impressive power across its 12,000 rpm operating range.

The combination of OEM tune, OEM intake, and OEM exhaust on a Ninja 400 / Z400 produces a mostly safe AFR, but not an ideal AFR, in open loop operation, and a very lean (but still mostly safe) AFR in closed loop operation.

But the combination of OEM exhaust header, Cat, and muffler seems to be a particularly restrictive one, and was probably the result of adapting the design to the newest European Union emissions and noise regulations which are very tough. In fact, those newest regulations caused Suzuki to actually suspend Hayabusa production temporarily, while its engineers looked for a way to make the bike compliant.

The restricted exhaust airflow in the Ninja 400 / Z400 apparently results in a lot of extra heat beyond what either Kawasaki engineers, or We, would have wanted, as evidenced by at least two very prominent symptoms: the engine coolant temperature gauge runs high compared to what we would expect, and the OEM exhaust tends to acquire a sort of burnt colour in the area where the OEM headers join at the collector (I know mine sure did with the OEM exhaust. You can highlight it by waxing the exhaust).

Those who have replaced the entire OEM exhaust with a lower restriction header and muffler system have found that this REALLY wakes up this engine. If you examine dyno charts from reputable tuners, you will see that almost any quality full exhaust system will immediately add 5% to 7% more power, and that is often all along the rpm range, not just at “peak”. The difference in on-the-road performance is VERY notable.

When a simple exhaust change adds THIS much power, without either an ECU reflash or an actual custom dyno tune, you KNOW two things right away:
  • The OEM exhaust system is VERY restrictive, and
  • The OEM ECU tune is remarkably successful in handling the extra airflow of a freer-flowing exhaust without any corrective retuning.
However, while this is great, it is not perfect.

This dyno chart from Jesse Norton’s website shows what happens to the 400’s AFR when a quality full exhaust (in this case the Akrapovic full exhaust) replaces the OEM exhaust. It shows the less than perfect AFR management at high rpm that can occur:

19894


This is not a “complete” graph, as it fails to show the OEM AFR for comparison, but it does show how changing from the slip-on exhaust (which left the Cat in place) to the full exhaust (which removed the Cat) leaned the mixture notably.

In my opinion, Jesse overstates the situation by calling the full exhaust AFR “bad”, since a peak AFR of 13.8 is not really “bad”, but I do agree it is not optimal for best power and cooling.

It would be desirable to richen the AFR throughout the rpm range, as that would:

  • Provide better part throttle response
  • Provide slightly more power
  • Provide some extra cooling, which the 400 would really benefit from

continued below . . .
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #2 ·
And, IF the 400 depicted in this chart had also had Jesse’s subsequently developed intake kit, which increases airflow even more via specially tuned snorkels and a modified airbox, it would very likely have needed even MORE enrichment of the fueling.

So, you can see that a Modified Ninja 400 / Z400 could really benefit from a Tuning Device that can richen the AFR a bit.

Again, I will repeat that either an ECU reflash or a custom dyno tune could certainly address this need, but both are costly, and not always logistically convenient, or practical, for some 400 owners (like Me).


Why an O2 Sensor Eliminator MIGHT Be Useful for tuning AFR on a modified Ninja 400 / Z400

There are a number of potential issues when trying to improve the performance of your 400 via some modifications:

Some quality aftermarket performance exhausts do have a “bung” (a fitting) that will accept the OEM O2 sensor. If your aftermarket exhaust has the bung, you can re-install the O2 sensor into it, and normally the exhaust manufacturer has located that bung so that the O2 sensor will work properly. In this case, your exhaust system has likely leaned the AFR, but as shown in the Norton chart, probably not to an unsafe level. And, the O2 sensor signals will at least direct the ECU to richen the AFR when in closed loop operation, although the ECU’s internal programming may limit the extent to which it can richen the AFR.

But some aftermarket exhausts do not have a bung, and most exhausts designed purely for off-road track riding or racing do not have a bung. But if an O2 sensor is not present, the 400’s ECU programming lights up the Check Engine Light (“CEL”), “semi-permanently”. The CEL can only be cleared by reconnecting an O2 sensor and going through a 3-times repetitive code clearing process. But if you have no bung, you can’t do that. Yes, you could drill a hole in the exhaust and add a bung and install the O2 sensor, but the routing of some aftermarket exhausts precludes mounting an O2 sensor in a location where it would actually work correctly (It has to be as close to the engine exhaust ports as possible while still being after the 2 headers have already been joined at the collector, and there has to be physical space between the exhaust and the engine crankcase to accept the O2 sensor).

If you cannot find a workable location, you have no choice but to get either an ECU reflash, or a custom dyno tune, that includes disabling the O2 sensor programming in the ECU. That might or might not be in your budget, and the ECU reflash will require removal and then replacement of your ECU. That means you will either have to do the ECU uninstall and reinstall yourself, or pay a Kawasaki mechanic to do so.

And, even if you do find a location to mount the O2 sensor in a less than optimal location, it won’t work properly (and the integrated electrical cable may not reach), and will potentially deliver improper signals to the ECU which could result in the ECU improperly altering the AFR.

Here’s where an O2 Sensor Eliminator might be useful.

But before getting into how, you need to know what an O2 Sensor Eliminator is and how it works.


What is an O2 Sensor Eliminator and how does it work?

An O2 Sensor Eliminator is a device that replaces the O2 Sensor physically. It continuously sends a signal to the ECU that causes the ECU to believe that a working O2 Sensor is still in place, even though it really is not.

An O2 Sensor Eliminator is inexpensive - typically under $30 CDN = $24 US.

Why does it exist?

Because:

- If an O2 Sensor fails and needs to be replaced, a 400 owner might be shocked to find that the price at your friendly Kawasaki dealer is measured in 3 figures. This is despite the fact that brand new generic O2 sensors on Amazon and eBay go for as little as $40CDN = $32US. (But their cables and connectors are not necessarily a match for the OEM Kawasaki cabling and connectors!)

- Production tolerances vary, and some motorcycles, including probably some Ninja 400s and Z400s, run rather poorly with the O2 sensor installed, so the owners buy an O2 Sensor Eliminator to stop the pain

- If an owner buys an exhaust that does not include a bung, and does not have the cash, or the practical ability, to get an ECU reflash or custom dyno tune, this becomes his/her best short-term option

On the 400, the O2 Sensor Eliminator is remarkably easy to install. On my Z400 for example, it requires:

- Removing the RHS upper body panel just below the fuel tank. which involves removing only 3 bolts, 3 pushpins, and then pulling on a number of places on the panel to free it from a few hidden mounting pins

- Locating the O2 Sensor connector that plugs into the OEM wiring harness (center of this photo):



19895



- Disconnecting the O2 sensor connector from the OEM harness

- Plugging in the O2 Sensor Eliminator instead

The above photo shows the O2 sensor connector disconnected and capped with electrical tape to waterproof it, and the O2 Sensor Eliminator plugged into the OEM wiring harness to replace and mimic the O2 Sensor.

You don’t even have to physically remove the O2 Sensor and its wiring cable from the exhaust and the motorcycle frame. (Leaving the O2 Sensor in place enables you to reconnect the O2 Sensor quickly and easily if you need Kawasaki dealer work, so that the Kawasaki technician does not try to diagnose why the simulated “O2 Sensor” signal is so constant.

In my testing, the entire process from beginning to end took me under 15 minutes the very first time I did it (10 minutes second and subsequent times), and requires only a cursory knowledge of how motorcycle electrical connectors connect and disconnect. Doing this on a Ninja 400 versus a Z400 will take a bit longer simply because the Ninja potentially has more bodywork to remove (I don’t have a Ninja model to try it on).

Ok, so it is cheap and easy to install. So where’s the potential use as a tuning device?

To answer that question, you need to understand HOW the O2 Sensor Eliminator does its job.

Recall that I said it sends a constant signal to the ECU that the ECU recognizes as a “normal and valid” voltage value for a working O2 Sensor to be sending.

The normal output signal for an O2 Sensor ranges from “0” volts (which means quite lean) to 1.0 Volt (which means “quite rich”). The breakpoint from “lean” to “rich” is roughly a value of 0.5 Volt.

An O2 Sensor Eliminator designer would not want to send a signal that is in the 0.5 Volt to 1.0 Volt range, as that would be telling the ECU “Hey the AFR is rich, so lean it down”! So, the designer creates a circuit, using the existing heater wiring, that produces a constant voltage that is somewhere between 0 volt and 0.5 volt.

I asked the designer and seller of an actual O2 Sensor Eliminator product, “Kawasaki Brad”, what his “simulated” O2 Sensor signal actually tells the ECU to do. He says it causes the ECU to believe that the AFR is lean, and to enrichen the AFR by between 2% and 3%.

So here’s the important thing to recognize:

The O2 Sensor Eliminator is actually causing the ECU to enrich the mixture by 2 to 3%. That makes it a potential useful tuning device. The owner of a 400 that is running slightly lean due to installation of an aftermarket exhaust, or simply an owner who dislikes the way his 400 behaves in closed loop operation, can use the Eliminator as a means of enrichening the AFR by 2 to 3%.

Now 2 to 3% does not sound like much. But if your current AFR is, for example, 13.5, a 2 or 3% change moves it to 13.2 to 13.1. You would feel that difference in the bike’s performance, and maybe see it also in the bike’s engine temperature coming down a bit.

For just $24US and 15 minutes of easy, relatively unskilled work. Done by yourself.

The ease, speed, and relative lack of skills required also means that if a rider’s bike is still under warranty, the O2 Sensor can be reactivated within 15 minutes so that a trip to the dealer does not prompt dealer attempts to blame any unrelated engine issues on the removal of the O2 Sensor.

Now you might ask why the ECU does not continue to iteratively apply the 2 to 3% enrichening repeatedly, since it is still receiving an “AFR is too lean” signal from the O2 Sensor Eliminator. The following paragraph will explain why.

I had actually asked Brad a related obvious question: Why limit the output signal to produce “only” a 2 to 3% enrichening. Brad explained that the ECU will tolerate the 2 to 3% signal without questioning it. But with larger voltage signals, it expects to see a threshold change in the signal voltage once it has made the enrichening adjustment. i.e. With a larger adjustment, it expects to see a notable reactive change in O2 output voltage. Since the Eliminator is a steady state device, and not an actual O2 sensor, it cannot produce that changing result. So, the ECU then assumes there is a problem with the fuel system, and lights up the CEL. It could even force the engine into a safe mode. This is why Brad limits the enrichening to the ECU’s 2 to 3% apparent tolerance limit before it reacts.

The 2 to 3% enrichening is obviously not a “perfect” solution. BUT, it is inexpensive, fast, and easy, and should produce favourable changes in the bike’s performance, and in many cases, will be “sufficient”.

Also, note that while this solution disables the O2 Sensor, both an ECU reflash and a custom dyno tune do too. This is because if the O2 Sensor is left operative during or after a reflash or custom dyno tune, it will cause the ECU to “fight” the reflash or custom tune, trying to restore that 14.7 AFR, and in that unsuccessful attempt, it would also ultimately trigger a CEL. So, there is currently no aftermARKET performance tuning method that preserves the functionality of the OEM O2 Sensor.


Why an O2 Sensor Eliminator must only be used on a bike that has had its catalytic converter (“Cat”) removed

It is important to remember that using an O2 Sensor Eliminator on a motorcycle still equipped with a Cat is a bad idea.

This is because the enrichening of the AFR will cause the Cat to run even hotter than it normally already does, and will thus greatly accelerate the longterm overheating and ultimately the plugging up of the Cat. That would be a very expensive outcome, as a replacement Cat is very costly. (On the Ninja400 / Z400, the Cat can only be bought as an entire exhaust assembly less only the muffler).


One thing to watch out for on any brand of O2 Eliminator

One thing to watch out for on any brand of O2 Eliminator is the quality of its fit into the OEM wiring harness connector.

The OEM connectors used are quality pieces that are made to be both water-resistant and able to handle vibration, but they are not indestructible, and there are production tolerances in the manufacture of any OEM or aftermarket fitting. If you buy an O2 Eliminator, and find that it does not “plug in” relatively smoothly, do NOT apply too much force trying to get it connected. You can dislodge the pins in the connection, and then have to figure out how to get them back into aligned and proper-length position. And if you bend a pin, it might be tough getting it straightened given the shrouded location of the pins. If an O2 Eliminator does not fit in and lock smoothly, simply ask the supplier for a replacement. They evidently know that some pieces just aren’t perfect in their tolerances.

continued below . . .
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #3 ·
Actual testing of this theory and the results

Ok, so we have a working theory. But does it actually translate into real life?

If it does, we should see:

  • Better feel when cruising, and better throttle response to moderate throttle inputs
  • Reduced fuel mileage (since we are supposedly running richer)

We MIGHT also see:

- Slightly cooler engine temperature (might or might not be enough lower to register on the coarse temperature bar graph used on the 400)

- Slightly better power in harder acceleration (although the difference might be hard to feel without a dyno)

Because it is so fast and easy to install / uninstall the O2 Eliminator, it is actually possible and practical to test the O2 Eliminator versus O2 Sensor as fairly as possible, by switching back and forth between them, between different test segments and days.

What I therefor did was run an identical 86km = 53 mile route multiple times, alternating the usage of the O2 Eliminator versus the O2 Sensor. This route includes residential neighbourhood streets, commercial business streets, suburban highway, and long distance highway segments, and the entire route is very hilly with lots of elevation changes, turns, and curves. It’s a pretty ideal route because it embodies just about every type of riding except high speed freeway. (We don’t HAVE high speed freeways on Vancouver Island).

The “cruising” speeds range from 50KPH = 30MPH, through 60-70KPH = 35-45MPH, and 80-100KPH = 50 to 62MPH. I ran the route using a mix of acceleration rates, including two 0 to 130KPH = 0 to 80MPH full throttle accelerations on each run.

I was also careful to make each comparison run under similar conditions (temperature, wind, and time of day). And, for each run, the fuel tank was filled, and then refilled right after the run. I was very careful with the fuel fillups, as I knew I was looking for a very small fuel mileage difference if indeed it existed. And for each run, I recorded both the fuel mileage average as stated by the dash instrument, as well as my own calculated average based on trip meter kilometers and actual fuel dispensed.

The results were interesting.

First; the fuel mileage:

(Note that here in Canada, we state fuel mileage as No. of liters used / 100 km traveled, but I show the fuel mileage results in both the Canadian and U.S. units below)

Average Dash L/100km with O2 Eliminator = 4.2
Average Calculated L/100km with O2 Eliminator = 4.1
Average Dash L/100km with OEM = 4.0
Average Calculated L/100km with OEM = 4.1
So Dash/calc blended average with O2 Eliminator = 4.15 L/100km = 56.7 Miles/US gal
Dash/calc blended average with OEM = 4.05 L/100km = 58.1 Miles/US gal

So, the O2 Eliminator used 2.5% more fuel on average

This is interesting because the 2.5% more fuel used with the O2 Eliminator is EXACTLY in the middle of the 2 to 3% fuel enrichment you would expect if the O2 Eliminator indeed enriches the AFR by 2 to 3%.

The fact that the AVERAGE fuel mileage during the testing was 2.5% worse means that the O2 Eliminator affects fuel mileage under ALL conditions - closed loop AND open loop (since my testing included both).

So, apparently, the O2 Eliminator does affect open loop operation as much as it effects closed loop operation. I’m told that this is because once the ECU is in an enriched closed loop map (due to the enrichening caused by the O2 eliminator) it's selection or scale of open loop fuelling is now relative. Even though it is not listening to the O2 signal in real time (during open loop) it is still basing the fuel table off the last time it read and reacted to the O2 signal.

Ok, so we now know that the O2 Eliminator does indeed richen the AFR, apparently under all conditions, just as I hoped it would. That helps to keep the AFR at least closer to what we would like, which improves power a bit and also cools the engine a bit, and might improve throttle response. But, can we actually detect the power, cooling, and throttle response benefits in everyday riding?

The answer is yes, but it’s not dramatic. Let’s take the remaining results one at a time.

First, yes, the cooling did improve modestly (not dramatically). Recall that my Z400 has consistently been running cooler than a stock Z400 / Ninja 400 ever since I replaced the OEM exhaust system with the Delkevic full system. It has generally run with 3 bars showing on the temperature gage, but moved to 4 bars at times in city traffic, or right after sitting for a few minutes at a gas stop (recovering to 3 bars after a fraction of a kilometer), or when running a bit harder than normal on the very hilly and winding roads here on The Island. It would sometimes move to 5 bars if the ambient temperature was abnormally high (like during our recent “heat bubble”). The 5 bars has never happened outside of very hot weather.

With the O2 Eliminator, it took longer to move from 3 bars to 4, and from 4 bars to 5 bars.

But the bars are a very coarse measuring method. I am working on a project that I hope will enable getting live temperature readings from the Kawasaki Diagnostic System (KDS) port via a Bluetooth OBD2 scanner that will feed my iPhone. If that project succeeds, I will be able to follow up this report with some specific temperatures.

I also did experience better low and moderate rpm operation with the O2 Eliminator. Again, the difference was not dramatic. It manifested in an interesting way. There is a suburban section of the route mentioned above that runs for maybe 10 kilometers that is “awkward” because it is mostly up and down hills and very curvy because of the terrain and a very limited road building budget. Consequently, it has a 60 kph = 36 mph speed limit. I normally run it in a combination of 4th and 5th gears, mostly 4th, running at what the 400 engine regards as “low” rpm.

With the O2 Eliminator installed, I kept getting the persistent feeling that the engine would prefer an upshift: instead of a combination of 4th and 5th, a combination of 5th and 6th felt “right”.

On the higher speed highway section, the manifestation was that I am used to looking down at the speedometer every now and then to check my speed. Usually, with the O2 sensor in place, I found that I needed to speed up a bit. With the O2 Eliminator in place, I kept finding myself going a bit too fast.

Throttle response at 4000 rpm in both 4th and 5th gear city traffic seemed noticeably more enthused. I know this is non-quantitative, but as I alternated runs between O2 Sensor and O2 Eliminator, it was consistently there.

As for harder acceleration, I cannot say with any certainty that it felt notably stronger (I am not a dyno). However, full throttle acceleration from a standing start produced modestly higher wheelies with more wiggle in the bars than I had experienced before the O2 Eliminator.

Based on all the above, I think I can fairly say:

1. The AFR did richen, as evidenced by the fuel mileage testing, which does give me that extra margin of AFR safety

2. The engine is running a bit cooler (but not dramatically so), as evidenced by the coolant temperature gage, again giving me that extra margin of safety. And Remember, my Z400 was running cooler than OEM BEFORE the O2 Eliminator, so the further decrease in coolant temperature with the O2 Eliminator is both helpful and impressive, even if slight.

3. The throttle response is noticeably better at low rpm and low load, manifesting best in city traffic.

4. Midrange torque (and therefor also power) may have increased just a bit as there is now an unintentional gain in highway speed in 6th gear compared to before when not paying attention, and the engine now favours running in a higher gear in that hilly section of the test route. But again, it is not dramatic.


Overall, I conclude that the O2 Eliminator did what I hoped it would. And for the price, this is a great cost versus benefit ratio.


An Invitation for Feedback

It would be nice to get feedback here on both my analysis and any experience any of you might have had with an O2 Eliminator. I’ve seen multiple discussions on other forums over the past few weeks, but very few authoritative technical answers and no carefully controlled tests.

Jim G
 

·
Director of Moderation
Joined
·
2,725 Posts
Did you make 3 posts because the forum has limits on word count or characters and you hit that limit twice? Just curious if it does or not because I don't think anyone else has hit that limit lol

Can you or someone that's willing to read all of it post the TL;DR of this? lol
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #5 ·
I don't know what the limit for a posting length is on this specific forum software, but I have found in the past on other websites that 4 pages in MS Word is about the limit.

And yes, the summary is that an O2 Eliminator (at least the one I tested) DOES richen both closed and open loop engine AFR, with a number of AFR safety, engine cooling, and performance improvements. It's a very inexpensive partial solution.

Jim G
 

·
Super Moderator
Joined
·
1,547 Posts
Great write up Jim. All of this info is scattered around this forum, most of which is provided by professionals like Norton and Woolich and verified by many of the racers here. I fully understand why you don't get a dynotune. But why are you so against sending off your ECU to get reflashed by someone like Norton or the many other places that do this. The results and the low cost and simplicity of a reflash make that option a no brainer. Then you will be off enjoying the amazing curvy roads of Vancouver Island with the amazing performance of your Z. To the extent that you went to get the lightest tires made, I would have expected you to not settle for a less than perfect solution to the AFR. But this write up seems very good to me.
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #7 ·
AtomicMonkey: I know that for a rider who lives in the U.S., the Norton ECU reflash is reasonably convenient except for having to get, remove, and then later replace he ECU (becquse of its burial by Kawasaki in the absolutely hardest part of the bike to get to!). But for me, he are the issues that stack up to create a lot of time and hassle, even though I do have a spare ECU that I bought long ago, thinking that I might someday do the reflash:

- Because I live in Canada, I have to mail the spare ECU "internationally". That is costly from Canada (around $40 Canadian), and with COVID-related delays would take a LONG time to arrive (my purchases from the U.S. typically take up to 4 or 5 weeks)

- The $300 US cpst becomes $392 Canadian after 21% currency exchange and 3% credit card exchange surcharge

- When Norton sends it back, they need to me charge me "International" shipping too, which will likely be about $40 Canadian

- So, my real cost is at least $472 Canadian - so far

- I also don't know how replacing my OEM ECU with a different one might affect things on the bike. For example, VIN ID, odometer reading

- I would need to make sure that Norton can and does apply some special instructions. For example, I don't want to lose the strong engine braking on closed throttle deceleration (I LOVE the sound, the feel, the extra load "off" the brakes, etc)

- The timeframe for the whole process so far is likely 8 to 10 weeks after I send it out, plus the turnaorund time at Norton whatever that might be

- Once the ECU gets back to the Canadian border, I will at a minimum need to pay our combined Canadian Federal and Provincial 12% sales tax (applies to imports), which adds $46 Canadian, for a total so far of $518 Canadian

- The above is only if Canadian Customs accept my statement that it is ONLY the reflash that I bought. If they insist on taxing the ECU itself (thinking that I am actually also trying to import a new ECU, versus just buying the reflash, becasue many people apparently try to do that), I get stuck in a bureaucratic process of rounding up sufficient proof that Iowned the ECU before sending it in for just the reflash, and who knows how long or how successful that would be

- Then, there's the duty on the import. For an ordinary consumer like me, it is not possible to determine in advance what the duty might be. Custome itself takes a while to figure that out once they have the item in hand. So, for now, it's "undetermined"

- Once I receive the reflashed ECU, I have to disassemble basically the entire top and middle section of the bike to get to the OEM ECU to make the swap. That means dealing with the fragility of the rigid plastic fuel line section and the tightly packed multiple electrical connectors that prevent direct access to the OEM ECU. That is not something I look forward to doing.

- If ANYTHING goes wrong with the reflash, I have to repeat the entire 8 to 10 week ordeal.

So, the "low cost" solution costs me $518 Canadian plus any duties, takes 2 to 2-1/2 months, and involves a disassembly and reassembly of a significant portion of the bike which I have not yet ever done (haven't needed to!). The tune might get me a BIT more power, but, per Norton's own graphs, not a LOT more power.

In comparison, the Delkevic FULL, stainless steel and Carbon Fiber exhaust cost me way less, took me maybe a couple of hours to install - without having to disassemble anything beyond the exhaust itself, and delivered a BUNCH more power.

This is why I did the exhaust first in preference to the reflash.

So yes, I'd love to get the reflash, but with the high and really undeterminable TOTAL cost for ME (as opposed to a more typical U.S. based buyer), and the hassles involved, I'm just having trouble justifying doing it. If the cost were lower, and also determinable in advance, I'd be tempted, just because I normally favour the BEST solution. But so far, there just seem to be too many obstacles to deal with to make it attractive to me. :(

Jim G



This is also why
 

·
Registered
Joined
·
1,005 Posts
Jim, I think you have a pretty good handle on the basics of how the fuel system work. You’re only missing on a few important details.

In as few of words as possible the Kawasaki ECU has two static fuel maps. These maps are flashed into the ECU and never change unless you have it reflashed by a tuner. Actually the ECU holds two sets of those two maps but forget that for now.

Those two maps are called IAP for intake absolute pressure, and TPS for throttle position setting. The IAP map is manifold pressure vs rpm and the TPS map is throttle position vs rpm. At low speeds and small throttle openings the bike runs on the IAP map and at some predetermined time switches over to the TPS map. The injectors timing signals coming from these maps can be and are modified in real time by sensors for altitude and air temperature. Also by water temperature but only during cold startup.

The O2 system does not change the base maps. When it warms up and goes closed loop it starts building a table of correction values. This table is not real time and does not make corrections to the injector timing signal in real time. This table is built from averaged values it’s collected over the past running 50-75 miles of operation.

Now think about this. When you remove the battery for some time this O2 correction table gets zeroed. You are now running on the base maps only for about 50-75 miles until the correction table is built. And by that same process anything that changes the O2 reading, like your O2 cheater, has to work it’s way through all that averaging of data before you’ll see it’s true effect.

And this O2 correction table only applies to the small throttle openings and low revs that it’s been designed to cover. Usually throttle position below 20% or so and below 5000 revs. I’m really not sure what the range is for the Kawasaki but higher revs and throttle openings are not effected at all.

In the past I have used O2 cheaters on my KTMs and Ducati’s. They work but not terribly well. I found that sometimes the bikes ran well and sometimes not so well. I’m not sure why. But dumping them and getting a true performance flash was the best bang for the bucks you’re likely to find.

19902
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #9 · (Edited)
Duckman: Thanks for adding more info. But, I do have to say:

1. The increaased fuel consumption on my Z400 with the O2 Eliminator installed continues consistently with every single fuel fillup (I've left it installed now for a longer term test) and that proves that the fueling of the bike HAS been altered.

Now, that COULD be because the open loop operation on the 400 begins at larger throttle openings and higher rpm ranges than I spend most of my time riding in. If so, I COULD be very rarely triggering open loop. In my riding environment here on The Island, there are not many opportunities for large throttle operation, because of the winding and hilly nature of the roads. But, I think we'd need a Kawasaki fueling system designer to tell us exactly under what conditions open loop is allowed.

2. I've already really shown that the reflash is not the best bang for the buck. The Delkevic full exhaust was a significantly better bang for the buck. I picked up a LOT more power with the exhaust than I could with the Norton reflash. And judging from Jesse Norton's published dyno charts, the Delkevic is not unique in that - the Akrapovic and other full systems seem to work that well or better as well. The reflash is very desirable, but the exhaust beats it for best bang for the buck.

3. Your comment about an O2 Eliminator sometimes working well and sometimes not does not surprise me. I suspect a lot depends upon how good or bad the closed loop system designers on any specific motorcycle brand and model did their OEM work. You can see lots of evidence of the varied effectiveness by brand and model by reading many of the online posts on multiple brand/model forums.

IF my Z400 is typical versus somehow unique, it would seem that the Z400 is a good candidate for an O2 Eliminator. One other forum member I know of is running one and has, like me, seen favourable results apparently. But, his 400, like mine, is not stock - I cannot recall what mods he has done. I think we would need a much larger sample to get a statistically valid reading on effectiveness. Of course, the low cost of an O2 Eliminator, and the easy install and reversibility, suggest that we might soon see more reports specific to the Ninja/Z 400. For now, I only feel comfortable saying that it seems to have worked for mY Z400 whose only significant power-related mod is the Delkevic full exhaust.

Jim G
 

·
Director of Moderation
Joined
·
2,725 Posts
Jim, I think you have a pretty good handle on the basics of how the fuel system work. You’re only missing on a few important details.

In as few of words as possible the Kawasaki ECU has two static fuel maps. These maps are flashed into the ECU and never change unless you have it reflashed by a tuner. Actually the ECU holds two sets of those two maps but forget that for now.

Those two maps are called IAP for intake absolute pressure, and TPS for throttle position setting. The IAP map is manifold pressure vs rpm and the TPS map is throttle position vs rpm. At low speeds and small throttle openings the bike runs on the IAP map and at some predetermined time switches over to the TPS map. The injectors timing signals coming from these maps can be and are modified in real time by sensors for altitude and air temperature. Also by water temperature but only during cold startup.

The O2 system does not change the base maps. When it warms up and goes closed loop it starts building a table of correction values. This table is not real time and does not make corrections to the injector timing signal in real time. This table is built from averaged values it’s collected over the past running 50-75 miles of operation.

Now think about this. When you remove the battery for some time this O2 correction table gets zeroed. You are now running on the base maps only for about 50-75 miles until the correction table is built. And by that same process anything that changes the O2 reading, like your O2 cheater, has to work it’s way through all that averaging of data before you’ll see it’s true effect.

And this O2 correction table only applies to the small throttle openings and low revs that it’s been designed to cover. Usually throttle position below 20% or so and below 5000 revs. I’m really not sure what the range is for the Kawasaki but higher revs and throttle openings are not effected at all.

In the past I have used O2 cheaters on my KTMs and Ducati’s. They work but not terribly well. I found that sometimes the bikes ran well and sometimes not so well. I’m not sure why. But dumping them and getting a true performance flash was the best bang for the bucks you’re likely to find.
^This! All of that. And to add further...the only good accurate way to measure how rich or lean a bike runs is to measure AFR on the dyno, or with a data logging system that has an O2 sensor.
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #11 ·
^This! All of that. And to add further...the only good accurate way to measure how rich or lean a bike runs is to measure AFR on the dyno, or with a data logging system that has an O2 sensor.
I'd love to do an actual AFR test, BUT

1. There is no motorcycle dyno anywhere near enough to be a day trip for me. In fact, I think I might have to cross the border into The U.S. to get to the nearest one

2. Using a datalogging system would only work if the OEM O2 sensor is replaced with a wideband sensor, as the output of the OEM sensor is far, far too crdue to measure actual AFR. But, I don't have a datalogging system either, and a datalogging system would surely cost more than a reflash (and much more than the $24US O2 Eliminator that ANYONE, ANYWHERE can do). I maintain that a consistent and reproducible change in fuel mileage IS a reliable indicator that fueling HAS changed.

Jim G
 

·
Director of Moderation
Joined
·
2,725 Posts
I'd love to do an actual AFR test, BUT

1. There is no motorcycle dyno anywhere near enough to be a day trip for me. In fact, I think I might have to cross the border into The U.S. to get to the nearest one

2. Using a datalogging system would only work if the OEM O2 sensor is replaced with a wideband sensor, as the output of the OEM sensor is far, far too crdue to measure actual AFR. But, I don't have a datalogging system either, and a datalogging system would surely cost more than a reflash (and much more than the $24US O2 Eliminator that ANYONE, ANYWHERE can do). I maintain that a consistent and reproducible change in fuel mileage IS a reliable indicator that fueling HAS changed.

Jim G
I disagree about that being a reliable way. Too many variables involved. Environmental conditions, where you ride, how you ride, etc. You'd have to replicate everything exactly the same way to make it reliable which is pretty much impossible to do for the duration of an entire tank of gas (so around 150 miles). I have no doubt that this little gizmo changes your fueling, but hard to really know by how much and how it changes throughout the rev range and TPS. AFR is not a single value. It changes based on RPM and TPS (also IAP like Duckman said, but that's for low throttle applications). I just flashed my ZX6R with Woolich yesterday after running Autotune and logging the data from a session at the track. I realized it needs a lot more tuning based on what I saw in the table. AFR was good at 100% throttle, but besides that it runs lean....so back to the dyno it goes. I didn't have enough time and gas last time so I only tuned it for 100% TPS. I need to finish it up now.
 

·
Registered
Joined
·
1,005 Posts
I disagree about that being a reliable way. Too many variables involved. Environmental conditions, where you ride, how you ride, etc. You'd have to replicate everything exactly the same way to make it reliable which is pretty much impossible to do for the duration of an entire tank of gas (so around 150 miles). I have no doubt that this little gizmo changes your fueling, but hard to really know by how much and how it changes throughout the rev range and TPS. AFR is not a single value. It changes based on RPM and TPS (also IAP like Duckman said, but that's for low throttle applications). I just flashed my ZX6R with Woolich yesterday after running Autotune and logging the data from a session at the track. I realized it needs a lot more tuning based on what I saw in the table. AFR was good at 100% throttle, but besides that it runs lean....so back to the dyno it goes. I didn't have enough time and gas last time so I only tuned it for 100% TPS. I need to finish it up now.
This.
There is no way you fill up a few times and calculate your mileage and tell if your afr has gone from 14.7 to 14.3 (3%).

If you install a wide band O2 monitoring system, I’ve used Woolich and Bazzaz, you’ll see that in actual riding your afr is all over the place. It’s never a static number b cause everything is always changing.

THIS IS WHY the stock O2 system works on long range average reading.
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #14 ·
SBK1198: If you re-read the way I did the fuel mileage testing, and the 3x repetition, I think you will see that the results WERE very consistent, across all 6 included test rides, and they have CONTINUED to be consistent in my post-testing riding.

I knwo my bike's fuel mileage VERY well, as I refill it as the last step of EVERY ride, record the bike-computed versus my own hand-computed results, AND also check the longterm averages frequently to look for any changes over time.

I am VERY confident that my results are both reliable and consistent.

Jim G
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #15 ·
This.
There is no way you fill up a few times and calculate your mileage and tell if your afr has gone from 14.7 to 14.3 (3%).

If you install a wide band O2 monitoring system, I’ve used Woolich and Bazzaz, you’ll see that in actual riding your afr is all over the place. It’s never a static number b cause everything is always changing.

THIS IS WHY the stock O2 system works on long range average reading.
How do you explain the fact that the fuel consumption has CONSISTENTLY been higher, and the same consistent amount higher, whenever I have the O2 Eliminator installed and consistently lower when I have the O2 Sensor installed?

It's absolutely and reliably predictable. I was actually surprised at HOW consistent and predictable it has been, and how it is precisely 2.5% different - right in the middle of the 2 to 3% range that the O2 Eliminator's designer told me it richens the mixture.

Jim G
 

·
Registered
Joined
·
1,005 Posts
How do you explain the fact that the fuel consumption has CONSISTENTLY been higher, and the same consistent amount higher, whenever I have the O2 Eliminator installed and consistently lower when I have the O2 Sensor installed?

It's absolutely and reliably predictable. I was actually surprised at HOW consistent and predictable it has been, and how it is precisely 2.5% different - right in the middle of the 2 to 3% range that the O2 Eliminator's designer told me it richens the mixture.

Jim G
I can’t. But I can tell you what that I have taken advantage of that second set of fuel maps in the ECU that I mentioned earlier. I have one set mapped at 12.8 and the other set mapped at 13.2. While actually riding on track I can switch back and forth with a switch in my bars. Even at NOLA where we are topped out in 6’th for a while I can switch between the two fuel maps and my butt dyno and my Speedo can not tell the difference. I’m sure a real dyno can.
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #17 ·
That's a nice capability you have there with the dual mapping! Makes me envious!

Jim G
 

·
Registered
Joined
·
1,005 Posts
It’s built into the stock ECU. In the USA both maps are the same. In some countries they might be a restricted hp map and a full power map. It’s simple to switch between the two maps.
 

·
Director of Moderation
Joined
·
2,725 Posts
I can’t. But I can tell you what that I have taken advantage of that second set of fuel maps in the ECU that I mentioned earlier. I have one set mapped at 12.8 and the other set mapped at 13.2. While actually riding on track I can switch back and forth with a switch in my bars. Even at NOLA where we are topped out in 6’th for a while I can switch between the two fuel maps and my butt dyno and my Speedo can not tell the difference. I’m sure a real dyno can.
Yeah you're not gonna notice any difference while riding. Before I starting tuning my ZX6R, it was running VERY rich. The person who flashed it before either made a mistake and loaded the wrong map on it or he had no idea what he was doing, but when I checked it on the dyno, the AFR was between 10-11 mostly, with the highest being right at 12.0 but only over the span of about 1000 rpm. I could tell it was rich based on the smell and popping, but had no idea it was that bad. After I subtracted a bunch of fuel throughout the most of the map, I brought it up to around 12.9-13.2 for the most part and there was barely any difference in the power and torque curves. They were within less than 1% difference throughout. After riding the bike I couldn't tell the difference at all other than it ran hotter.
 

·
Registered
Joined
·
521 Posts
Discussion Starter · #20 · (Edited)
Yeah you're not gonna notice any difference while riding. Before I starting tuning my ZX6R, it was running VERY rich. The person who flashed it before either made a mistake and loaded the wrong map on it or he had no idea what he was doing, but when I checked it on the dyno, the AFR was between 10-11 mostly, with the highest being right at 12.0 but only over the span of about 1000 rpm. I could tell it was rich based on the smell and popping, but had no idea it was that bad. After I subtracted a bunch of fuel throughout the most of the map, I brought it up to around 12.9-13.2 for the most part and there was barely any difference in the power and torque curves. They were within less than 1% difference throughout. After riding the bike I couldn't tell the difference at all other than it ran hotter.
THAT's an important finding. So, I guess the 400 engine is surprisingly tolerant of a rather wide AFR range. But the comment on running cooling with the richer AFR is also important.

Tonight, I rode my Z400 over the mountain pass to meet some friends on the coast for a ride. The approach to the summit is 5 km = 3 miles of winding road (almost switchback) with an AVERAGE 6% grade, posted at 80kph = 50MPH except for the sharp turns at the "almost switchbacks"! (That 6% grade is 20% steeper than the maximum grade allowed on the U.S. Interstate highway system).

The Z400 maintained just 3 bars on the coolant gage until I was within about 30 seconds of cresting the summit! It did flash to 4 bars for that last 30 seconds, but then quickly fell back to 3 bars. The ambient temperature was 29C = 84F at the time.

Jim G
 
1 - 20 of 55 Posts
Top