Tuning the RX4's Fuel Injection
 

Over in the RX3 subforum, user Lukas has made considerable progress in accessing and modifying the Delphi ECU's fuel injection map. For those interested, the thread is here:

http://www.chinariders.net/showthrea...d=1#post364185

Hopefully much, if not all of what he learns can be applied to the RX4's ECU for modifying its fuel injection mapping as well. There are definitely one or two areas where I would like for the RX4's fueling to be improved, with the light throttle, low speed stuff being at the top of my list.

Agreed I am also unhappy with the way my bike idles and the power just above idle
...as it is way way too lean !...
as with any F.I. system they always tune for economy and emission control not performance..... and I would be very interested on how to change that... the bike needs it !
....
Bob.......

Just an idea outside the box. I remember there was a thread on here about my type of bike. Not the same thing at all. However, somebody was able to turn a screw on the air intake to adjust it so that the EFI tendency to run way too lean at idle was improved.

I read that posting from Lukas and found it interesting, I have a Lifan expect and would like to learn more about the ECU system so will try to get the hardware from Amazon. The first problem is that Lifan has their own ECU which is a Hunier EFI 9.0 whereas the other China bikes use the Delphi and the port plugs are not compatible. I have downloaded ECU Hacker and when I get the cables from Amazon I will have a stab at printing a plug to fit the Hunier plug but don't really know if the ECU Hacker will recognize tis ECU.

...
thank you for that ! ....I haven't even seen a screw in there but I know there's gott'a be one somewhere ! LOL
I have to pull the rear shock soon so I will look into it then when I have it in the tiny shop... right now it's under it's cover being rained on and the occasional snow thrown in for good measure !
... part of the problem with the RX4 is getting all the stuff off of it so you can find the bike.... the valve adjustment is a snap once you get down to it.... but it'a a 2 hr job to get to that point ! .... reminds me of a Kawasaki Ninja !
you can't even SEE the carbs on my little 250R you have to pull the fairings to find them !
I guess most of the bikes are going that way though...personally I liked the old style where the headlight was attached to the front forks you could see around corners better at night ! oh well..... that's progress for ya !
....
Bob.....

The RX4 has a throttle body with a stepper motor to control idle speed by adjusting air bleed to the engine. There is no screw adjustment anywhere on the assembly. Some riders try to fool the ECU into thinking that the engine is running colder than the actual temp, just getting the ECU to inject more fuel. This 'fooler' is just an electrical resistor. Such a mod runs the real risk of over fueling across the board, washing the cylinders of oil needed for lube, and engine failure.

I just verified the lack of a screw by looking at the spare TB I have in my shop.

Bummer. It was just an outside of the box idea from another thread.

I wonder if guys who put on other carbs with bigger jets also run the risk of too much fuel and washing their pistons dry, haven't heard of any problems though as I think that would take a lot of fuel and I am sure that the bike would not run too well. I one is to alter the air intake temperature by a small amount would it make any difference before the ECU cut it back from the low reading of the O2 sensor. Mt thinking to alter the fuel air mixture one would have to alter the O2 reading. The whole input output relationship of the ECU needs a little more explanation, is it open loop at the start going to closed loop when all parameters are reached.

As I understand the system, you are running closed loop in the lower rpm conditions and open loop near full throttle. The O2 sensor will try to bring the mix back to a preset level but these O2 sensors are bargain basement parts with a very limited range of adjustment. That's common on motorcycles.

We have now exhausted my knowledge of how the ECU and its sensors work!

the ECU is the future, It is only a control instrument. Where once control instruments were linear devices they are now programable digital devices. In the case of the ECU it has taken the place of the carburetor and also the CDI so we need to take an in depth look at the EFI side of the controller which has many inputs and a few outputs. which of the inputs actually effect the fuel injection rate and how does it effect this rate. If we can determine the range of these outputs we may be able to make small changes to these signals so that we could alter the algorithm. What effect are we looking at? maybe increasing the pulse width of the injector and could we possibly see this if an oscilloscope is connected to the fuel injector. The other way of course is to get into the program and make alterations but unfortunately that is beyond my expertise. Can ECU hacker help to see the results of any changes to the inputs are made. I am surprised that some software geeks have not already broken this stuff down to down loadable apps yet. Maybe we should ask our grand kids whether they can help.

According to JerryHawk, you can use an adjustable potentiometer in lieu of the resistor on the intake air temp wire. Easier to turn the knob on the potentiometer than replace resisters. Take regular plug readings to prevent running too rich or too lean.

Adjusting the resistance leaves one with the same problem as before, the ECU will respond with more fuel across the entire rpm range. A good ECU map will try to maintain the mixture at a steady state but, in reality, many maps go from lean to rich and back to lean. This means that your 'fooler' might take care of a lean condition on parts of the map but a super rich condition elsewhere. The only way to know what's going on is to put the machine on the dyno. I have seen some very talented tuners who were able to build a map to flatten the mix but this is not something the average garage tuner can accomplish. If the Delphi ECU was used in large numbers of motorcycles we might see tuners stepping up to create a better map but that's not the case right now.

Personally, I do not have a problem with the fueling on my RX4. I'm keeping it stock, along with the intake and exhaust.

Does anybody know what the EFI sensors do, the input air temperature does it increase or decrease fuel with a rising temperature. What about the input air pressure is this just an advance signal to the EFI as to how much fuel to add with an opening throttle position sensor? and then what does the cylinder temperature sensor do, my thought maybe the sensor that closes the loop when the engine reaches operating temperature which then leaves the O2 sensor in control of the fuel levels. Anyone care to help here.

You might find this interesting...from Zongshen....https://www.youtube.com/watch?v=I6nzl8igRaY

Sadly, does not really answer your questions about how the sensors interact with the ECU. My experience is with Ducati and Moto Guzzi ECUs. There the temp sensor adds fuel while the engine is cold...much like the choke with a carb. The air pressure sensor will adjust according to air pressure...that is altitude. Engine temp...probably another fuel adjustment. The thing is, each of these sensors act in concert with each other. It's not a simple additive process. You might see the air temp sensor telling the ECU to increase fuel while the air pressure sensor is saying to decrease, and so on. That's why making a new map is so difficult.

“Adjusting the resistance leaves one with the same problem as before, the ECU will respond with more fuel across the entire rpm range. A good ECU map will try to maintain the mixture at a steady state but, in reality, many maps go from lean to rich and back to lean. This means that your 'fooler' might take care of a lean condition on parts of the map but a super rich condition elsewhere.“

Hmmm. Sounds just like many of my carb jetting experiences.

The reason I am horning in on your thread is because I may buy a Genuine G400C, and it has Delphi EFI.
I’d like to own an RX4, but the seat height concerns me since I have only a 29” inseam. I ride in a lot of stop and go traffic and tall seats make that more difficult.

Ah! A new weapon for the EFI/carb wars! Lolol. :lmao:

Great bedtime story.

The Delphi ECU is used on many small engines. They probably have similar (if not identical) sensors but the actual map is going to be modified to work with the specific engine model, exhaust and intake setup. As a purely academic exercise I'd like to know how the various sensors feed into the base map and how they interact. Check the throttle body size for that Genuine 400. On the RX4 it's a 45mm TB.

If you take a look at the IAP sensor on the x-pect it is between the throttle valve and the intake manifold so how does that give you the atmospheric pressure when it is working from almost a vacuum to near atmospheric pressure depending on the throttle position and the engine load. I sort of felt that that pressure sensor was sort of an advance warning to the EFI system as to where the RPM's were in relation to the throttle valve so as to supply more or less fuel. With RPM's rising on a flat road throttle valve will be closing and pressure in the inlet manifold will be dropping into more of a vacuum where Rpm's dropping on a hill with the throttle valve opening would be increasing the pressure to the IAP. So does this sensor really do anything about altitude or is altitude just handled by the O2 sensor in a closed loop.

x-pect?

The G400c makes a little over half of the hp of the Rx4, so it will have a much smaller throttle body. The rest of the EFI components should be very similar.

But the problem is that the bike Runs too lean from the get go !!! it could handle more fuel throughout the entire band ! .... the question is how much resistance is needed and where do I put it ? LOL I'm willing to add a resister in line to richen up my bike I don't think a bike should have a white plug in it...but that is how they are set up... !
.....
Bob.....

I own a 2021 RX4 and I lowered it 2" front and rear super easily
1.) i took out all the pre load and ran the nuts up to the top of the shock in the rear and locked them there... that dropped the rear almost exactly 2" (EDIT# I discovered that this is not the thing to do as it bottoms out the rear shock. I made a longer lower link instead to get it 2.5" lower)
2.) I loosened up the pinch bolts and raised the fork tubes in the triple trees up 2" and that lowered the front.... However to do that I had to buy some handle bar risers to fit the handle bars and move the handle bars back away from the fork tubes.... i got the 2" riser type that pivot in the original handle bar clamps
it worked real good I have a 30" inseam and can flat foot both feet at a stop.
.....
Bob.....

Thanks Bob Kelly. RX4 is back in the running for me.

When I get some time I will take a look at my air temp sensor and try to measure its resistance. Then raise the temperature and see if the resistance increases or decreases which will give me some idea where to aim for. Right now spring in the bee yard is keeping me a little busy. I would guess that the colder the air temperature the more fuel added.

And you know this how? Did you check with an O2 sensor or some other tool, or are you saying so because "Everyone knows"?

I removed and tested the Intake Air Temperature sensor and checked its resistance and at around 70 degrees F it read 1.8K ohms and at about 35 to 40 degrees F about 2.2K ohms.
So this thermistor has a negative coefficient and the higher the temperature goes the lower the resistance so to lower the temperature reading you will need to add some resistance in series with the thermistor. I would try no more than 100 ohms at a time or if you can find a 200 or 300 ohm potentiometer you could work from zero increase up to 300 ohms increase in steps. Measuring any changes is the challenge, however I just filled my tank with mid grade gas and feel there is a difference in power and acceleration on my Lifan e-pect so maybe the RX guys will be able to feel a difference. Looking at setting up ECU hacker I need to find some of those flat pins they use in the China Bike plugs so that I can make up a plug to fit the ECU port, any idea where I could find those.

LOL Your right I don't know for a fact ! but I trust my instincts and my 60 odd years of experience with motorcycles it feels lean throughout the entire power band..... and if I could add a resister to ritchen it up I most certainly would
to the gas mileages demise !
I have not seen a single Fuel injection system that gave you a tan sparkplug
they are all white.... and white is lean and hot.... you don't want that on a motorcycle period...even if it is water cooled tan is much better and a closer proximity to the desired 14:1 air to fuel ratio you should be getting
.... their probably set at 12:1 for all I know but it's certainly a ways away from the 14:1 sweetspot.
do a color check and see what I mean....the porcelain is white not tan or brown!
.....
Bob.......

I do not know if you have understood my thread about ecu is quite chaotic and I think to complete it with this content, http://www.chinariders.net/showthrea...t=29467&page=2 how to connect to hod ecu hacker, how to change maps and add what I did not describe, ie adding patch to the map which is quite convenient, because it allows you to change parameters without interfering in the map I changed the parameter of the cylinder capacity and maximum engine speed I think, But to tune the map I sent the map file to https://oldskulltuning.com/and for 30 euro they make tuning of fuel injection time and ignition I wanted to have straight-line map and it worked before changing cylinder I bought a camera and I would like to record how my bike runs.

These bikes do not have catalytic converter but catalytic converter mesh in the muffler which is an ordinary straight pipe and the mixture on the lower parts is set to very poor to make the exhaust as clean as possible which results in lack of power in the lower parts of the rpm my RX3 had power only from 6000k rpm now after map tuning by Italians the power is from 3000k rpm. At the beginning, it is quite a strange experience for someone who is used to riding this motorcycle, so the map must be tuned, i.e. injection time, fuel quantity and ignition timing, that all sensors make corrections to the amount of mixture and ignition timing in relation to the basic map and work together which means that a change in the signal of one sensor can cause something in the form of misinformation for the ecu which may result in a check engine light up.

Lukas by adding 100 ohms to the input air sensor all you would be doing is telling the ECU that the temperature is 10 degrees F cooler than it actually is. This does not change unless the temperature increases during the day and if it does the sensor will decrease in resistance and the temperature reading will continue to be 10F cooler than actual. So in effect the sensor will still increase and decrease but the ECU will just believe it is 10 degrees cooler. What the effect of this will be on the EFI is hard to tell but i think with just 100 Ohms in series with the sensor it will remain in range. Of course this is with the sensor on the Lifan e-pect not sure about the sensor on the RX's but that is easy to check with a multimeter.

The air intake temperature sensor influences the correction of the fuel mixture parameters and the ignition advance angle. Properly functioning power system and efficient sensor is very important to reduce fuel consumption, harmful exhaust emissions and improve engine performance. The intake temperature sensor is calibrated and its parameters must coincide with the parameters of the lambda probe in the exhaust manifold, if you change the value of the output signal, the computer on the one hand will get information that the air density is different and will give a different dose of fuel, but it will not be correct and the lambda probe will give a wrong result to the ECU please tell me what is your goal, do you want to increase engine power maybe I can help you.Do you have a picture of your ECU.

Lukas I am happy with my Lifan x-pect but would like to learn a little more about ECU,s and their operation. I will be setting up ECU hacker so that I can download information from my ECU which is a Hunier EFI 9.0. Also thought about putting a oscilloscope across the fuel injector and try to measure the pulse widths and by adjusting a 100 Ohm potentiometer in series with the IAS see if the pulse width varies by increasing the resistance. Just trying to learn by observation.

I think it is safer to run the tests from within the software, because you can change the parameters under the control of the ECU based on real data, by changing the resistance of the signal the data coming to the ECU will not match reality I think you can play with the AFR. On the internet you can find many products for cheating ECU by connecting various types of emulators that falsify the signal, but in fact it is just a bulshit, because real tuning always involves change of map calibration, because in case of damage to one of the sensors Ecu should go into emergency mode, which involves loss of power, and if any system such as change of cylinder capacity, change of exhaust or intake was changed then the emergency map can damage such an engine.Because the mixture will be too rich or too poor.

I would be very grateful if you would describe the observations you made during the experiment with the oscilloscope, because I have an old BMW from 91 and unfortunately there is no way to connect it to the computer and check why the engine runs unevenly and here the oscilloscope could help and any knowledge of this type is valuable to me and I hope that you will not pop out check engine during the tests.

Lukas as soon as my 100 ohm potentiometers and plugs arrive I will dig out my old scope and try out ECU hacker and see what the results are like. I think if the engine light comes on with a fault reading as soon as you correct that fault the light goes off. I ran the bike with the IA sensor unplugged and it ran with the fault light on and once I replaced the sensor the light remained off. The Hunier EFI 9.0 I guess is the same as the Lifan EFI 9.0 I could not find a plug to fit the EFI diagnostic female plug so will have to make one up to connect the 3 wires involved One is a positive another is ground and the last is the in and out data so we wll see how that goes.

I would recommend replacing the potentiometer with an actual fixed value resistor once you find an adjustment you like. Measure the resistance of the Pot when it's adjusted property and the wires are removed. Then permanently replace it with an actual resistor of around the same value. Potentiometers value change depending on temperature, wear and contaminates. Vibration could also wreak havoc temporarily changing the value or even creating an open circuit for a millisecond. Plus you don't want to make your motorcycle less reliable or add something else that can easily fail. Fixed resistors are far less effected by temperature, totally water, dust & vibration proof and almost never fail.

If you want adjustability then putting a fixed resistors in parallel with a Pot is a better idea because then if the Pot fails you don't have an open circuit condition.

Fitted the 100 ohm pot in series with the IA sensor thermistor and connected my 50 year old oscilloscope to the injector and ran the bike with the full 100 ohms added to the IA sensor. The x-pect now idles beautifully and revs up fine unfortunately the scope is a bust as all I can see are spikes from the injector that are so poor that I can not determine anything worthwhile from them. As soon as the weather improves a little I will take the bike for a long ride and see if I can feel any difference or if the ECU is going to complain. The pot is set up with 6" of twin core cable with a 2 way plug male and female either side so the IA sensor is removed one end of the pot plugged into where it was and the IA sensor then plugged into the other end of the pot cable, so the pot can be removed and a 100 ohm resistor can be easily added. The next step is to fit a new 4 way plug onto the 3 ECU wires from the ECU diagnosis plug

pyoungbl.... I hate to rain on your parade ...but in all my years changing jets and messing with bikes I have NEVER seen a condition where the cylinder is "Washed dry" of lubercant
that simply cannot happen and still run ! even with a badly sticking float where I could get the bike running by pushing it down a hill and dumping the clutch the spark plug will fowl out before any damage to the cylinder in such a wet condition....
so I think your fear is unfounded of washing the cylinder dry .... and sizing a piston or worse....
.... that being said I have to agree that the Resistor idea is not the best idea on the planet... like you said the fuel flow /Air ratio does waver alot, in a carbed bike. although not from the reading of the ECU in running condition...that fuel air ratio there is rock solid and does not deaviate at all with a carbed bike it does but as far as I can see the Fuel injected bike does not wonder at all with it's fuel air ratio ! and I was quite suprised at that..... but if you make it richer by changing the tempiture you will only make it very slightly richer.... probably not enough to even notice. and doing that will change how the ECU handles other things like Idle and perhaps deceleration.
.....
the ECU is a very precise instrument ! you don't want to be "Fooling " a sensor as almost every sensors has triple duty !
....Until I get to the point where I can change settings with no problem on my ECU I will leave it stock.... but I am sure already that there are a few settings where I can get a bunch more power from the thing.... doing that will be a major challenge !
....
learning to Run the HUD ECU HACKER program itself is the hardest part so far...there is a steep learning curve !
....
Bob........