Idle problem solved

[iwilson]

Could you reduce the hot idle down to around 1000rpm and install a simple throttle lock device for the cold idle. My bike doesn't take long for it to warm up from cold and settle to around 1200rpm, maybe 2 mins at most, which is about the time it takes to do up my chin strap and get my gloves on

The engine seems happy enough at 1,000 but the factory idle is 1,200 so that's what I'd aim for hot since that will give you a little more rpm headroom when cold starting. Only thing that concerns me with using a throttle stop is the TPS getting screwed up if it's set to auto-zero - but depending on how it's implemented it may sort itself out once warmed up. End of the day it's a simple enough to adopt a try it and see approach.

Actually managed to get out for a proper ride yesterday, although the idle was perfect the engine just wasn't pulling, so after a bit of head scratching I realised I had the factory Dominator ignition map and my own fuel map loaded. The two didn't get along - once that was sorted I had a great blast, I'd forgotten just how good this bike is when on song!

So having just enjoyed a very long and relatively rain free summer figured I'd take the bike to work today. Yeah right... There's always tomorrow I guess!

 

[iwilson]

Just a quick update on where I'm at with the idle. We left of with establishing the root cause of the engines inability to idle nicely being the charge robbing that occurs via the idle air pipe that joins both throttle bodies. Once this was plugged the air fuel ratio normalised between each cylinder and the engine just sounded a whole lot happier. No more fouled plugs and with the entire IAC system removed no odd idle behaviour hot or cold.

The only downside was in order to set the hot idle to around the factory specified 1250rpm the cold idle needed to be around 800rpm. However with the AFR now sorted the bike is perfectly happy to start and idle at this speed! No need to hang onto the throttle while the engine warms up. But I wanted to see if it was possible to solve the problem and have the ECU control the idle speed. So I got a local engineering shop to make this...

The idea was to block the tube and force the idle air to each cylinder to travel down individual tubes and reduce the possibility of charge stealing.

This is the plug that blocks the original IAC tube and feeds each throttle body with its own air supply.

Long story short it didn't work! The air lines are to restrictive to provide enough air to be able to alter the RPM by the required 300 or so needed to account for the engine going from cold to hot. Also it was proving difficult without sealant to block the balance tube so I was still seeing some charge robbing. So I had a couple of choices, spend more money on modifying the design to flow better and using a flexible sealant to glue the plug in. Or I could have machined new independent idle air lines into each throttle body and permanently sealed the old one. But as I only have one to experiment on, I decided not to!

The engine has never sounded better at idle so I'm going to leave it there as I've already spent more than I should! But if you want a fully reversible fix for the idle on your bike, then just get rid of the entire IAC system, block the tube and set the throttle position screw to achieve the desired idle speed. The result is quite spectacular, the engine is more than happy to idle at speeds it could only dream of before!



Rock solid start and idle from cold at 800RPM! Fully warm it sits at 1,200. I give it a couple of blips at the minute mark to see if it will recover back to idle without stalling (which it does).

So I'm going to leave it there, hoping that Norton/Jenvey will make a change to the design of the throttle body in the future (they are both aware of my findings).

 

[lcrken]

+1 on that. I'm also quite happy with the IAC delete and plugging the crossover tube.

Ken

 

[BritTwit]

The only downside was in order to set the hot idle to around the factory specified 1250rpm the cold idle needed to be around 800rpm. However with the AFR now sorted the bike is perfectly happy to start and idle at this speed! No need to hang onto the throttle while the engine warms up. But I wanted to see if it was possible to solve the problem and have the ECU control the idle speed. So I got a local engineering shop to make this...

Sounds like good progress.

Does anyone have an opinion about the oil pressure at 800RPM
Will this cause a potential issue, or will pressure be sufficient until engine is at operating temp and normal RPM?

 

[Fast Eddie]

I still don’t understand how one cylinder can rob mixture!

I understand how it could rob charge (fuel AND air) but I don’t understand how it can only rob fuel and leave the other side weak.

Can anyone shed any light on this?

 

[BritTwit]

I still don’t understand how one cylinder can rob mixture!

I understand how it could rob charge (fuel AND air) but I don’t understand how it can only rob fuel and leave the other side weak.

Can anyone shed any light on this?

Being a carb guy at heart, I am a little confused too.
But from another iwilson thread:

Here’s the crazy vid. In this one I switch both o2 sensors on and watch the ensuing chaos. Cylinder 1 screams for more fuel and you can see the fuel trim number rapidly increasing as the ECU adds fuel. This fuel of course is ending up in cylinder 2 which the ECU attempts to trim (negative number).......

I think what iwilson is saying is that the ECU misinterprets the O2 data from the sensors if ECU reads a rich AFM in cylinder #2 it immediately trims the mixture for the next ignition cycle - cylinder #1.
This would lean out the mixture that goes to cylinder #1.

What I don't understand is why can't the ECU keep straight which cylinder's AFM is lean/rich and correct only that particular cylinder's mixture on the next ignition cycle?

 

[Fast Eddie]

If that’s correct, at least it makes sense.

But it’s not ‘robbing’... it has nothing to do with one cylinder robbing from the other. It has to do with the ECU giving the cylinder the wrong ‘dosage’ in the first place.

Wouldn’t happen with carbs now would it !?!

 

[kommando]

What I don't understand is why can't the ECU keep straight which cylinder's AFM is lean/rich and correct only that particular cylinder's mixture on the next ignition cycle?

Its being confused by the pipe connecting the 2 intakes, it's a 270 degree twin so unlike a 180 degree twin the pipe is biased to one cylinder due to the different gaps.

 

[BritTwit]

If that’s correct, at least it makes sense.

But it’s not ‘robbing’... it has nothing to do with one cylinder robbing from the other. It has to do with the ECU giving the cylinder the wrong ‘dosage’ in the first place.

Wouldn’t happen with carbs now would it !?!

The alternative's looking pretty good about now, aye?

Hosted on Fotki

Hosted on Fotki

 

[BritTwit]

Its being confused by the pipe connecting the 2 intakes, it's a 270 degree twin so unlike a 180 degree twin the pipe is biased to one cylinder due to the different gaps.

Interesting, even with a camshaft position sensor, the ECU can't accurately control 2 separate cylinders?

 

[kommando]

It could if they are not connected as Iwilson in has found out, it's not really the ECU's fault as it's just reacting to what the sensors are telling it. You could add a routine so below certain revs the oxygen sensors are ignored but then it would fail emissions. Isolastics rule ok.

 

[BritTwit]

Isolastics rule ok.

And FCR's!

 

[iwilson]

If that’s correct, at least it makes sense.

But it’s not ‘robbing’... it has nothing to do with one cylinder robbing from the other. It has to do with the ECU giving the cylinder the wrong ‘dosage’ in the first place.

Wouldn’t happen with carbs now would it !?!

Eddie, one cylinder is definitely robbing from the other. It's a case of air taking the path of least resistance. The IAC valve acts as a restriction on the flow of air into the cylinders - that's how it modulates the idle speed. This restriction means that at certain times charge laden air is drawn through the IAC pipe from one cylinder to the other as it presents the path of least resistance. No doubt there is a combination of factors that contribute to the phenomenon, but I can state with 100% certainty that it is occurring! The fix is to run independent IAC lines to each throttle body and ensure the supply from the IAC valve is always the path of least resistance. Or for a no cost fix just remove the lot and plug the tube.

Regarding the o2 sensors the ECU is being presented with data from the o2 sensors that cyl. 1 is lean and cyl. 2 is rich. So it takes the appropriate action adds fuel to cyl. 1 and removes it from cyl. 2 which just makes matters worse. If you're lucky the ECU should realise all is not well and ignore the input from the o2 sensors. But you'll still have an engine that exhibits poor idle characteristics due to the charge robbing. Once the throttle is opened the problem disappears as air is feed via each throttle body butterfly.

 

[iwilson]

Being a carb guy at heart, I am a little confused too.
But from another iwilson thread:



I think what iwilson is saying is that the ECU misinterprets the O2 data from the sensors if ECU reads a rich AFM in cylinder #2 it immediately trims the mixture for the next ignition cycle - cylinder #1.
This would lean out the mixture that goes to cylinder #1.

What I don't understand is why can't the ECU keep straight which cylinder's AFM is lean/rich and correct only that particular cylinder's mixture on the next ignition cycle?

The charge robbing effect is huge. Cyl. 1 is running extremely lean and cyl. 2 extremely rich. Any additional fuel the ECU adds to cyl. 1 is promptly stolen by cyl. 2. So there is nothing the ECU can do to correct the situation. The ECU is getting good data, it's just that it cannot correct the problem as it's a mechanical problem not a software correctable problem. Cyl. 1 is always going to be lean, removing fuel from cyl. 2 will result in a better AFR, but you'll have this crazy situation of cyl. 1's fuel injector actually feeding cyl. 2. Cyl. 2's fuel injector would barely be operating at all!

Bear in mind that the fuel injector does not fire on the intake stroke, what happens is that it builds up a thin film of fuel on the back of the valve this is constantly being depleted as it vaporises and is drawn into the cylinder. You can tell the ECU when to fire the injector relative to the 720 degrees the engine turns for one cycle. Norton actually suggested I try this when I was trying to figure out the problem. I have a very long and boring video where I change the injector timing a degree at a time for a full 720 degrees while observing what effect it had on the individual cyl. AFR's. It did have an effect but not enough to come close to solving the problem and the injector timing was miles away from where you would normally expect it to be set (with the engine not sounding very happy).

This is the injector timing map as you can see injector timing is changed relative to RPM and throttle position.



This is the boring vid of me adjusting the injector timing through a complete engine cycle.

 

[richard-7]

We think some problems are from joined exhausts. When you replace with the catless pipes that are completely separate the bikes runs a lot better.

My dad and Tony separately made a manual choke. Works great. More to come.

 

[BritTwit]

The charge robbing effect is huge. Cyl. 1 is running extremely lean and cyl. 2 extremely rich. Any additional fuel the ECU adds to cyl. 1 is promptly stolen by cyl. 2.

So if I understand what you are saying:

The amount of time that the injector in cylinder #2 is providing fuel to cyl #2 is extended by robbing fueling time from the injector in cylinder #1's?

 

[iwilson]

So if I understand what you are saying:

The amount of time that the injector in cylinder #2 is providing fuel to cyl #2 is extended by robbing fueling time from the injector in cylinder #1's?

Not quite sure I understand what you're saying but....

When cyl. 1's injector fires (or at some point in the engine cycle after it has fired) a large portion of that fuel passes through the IAC connecting tube into cyl. 2 causing the mixture to run very rich. The o2 sensor in cyl. 2's exhaust pipe reports this to the ECU and the ECU reduces the fuel coming out of injector 2 (wrongly since it's injector 1 that is causing the mixture to go rich). Meanwhile cyl. 1's o2 sensor is reporting it is lean. The ECU increases the fuel coming out of injector 1 to richen the mixture - but this fuel of course is just ending up in cyl. 2 instead. This causes an endless loop of corrections which ends when the max fuel trim percentages are reached.

You can see the maximum fuel trim setting circled - which is +/- 25% of the value in the fuel table for that RPM/TPS setting.

Hope this makes it clear.

 

[Fast Eddie]

Brilliant stuff iwilson, that’s certainly ‘adding to the body of knowledge’.

You should wrap this up and submit it as a phd and you could become the worlds first Dr of Norton’s !

 

[TonyA]

We think some problems are from joined exhausts. When you replace with the catless pipes that are completely separate the bikes runs a lot better.

My dad and Tony separately made a manual choke. Works great. More to come.

Thats right ! Just waitng for the go ahead from base camp. Don't know what to name it yet either NAIS ( North American Idle Solution ) or One-07 Dial a Idle . Either way its real , its simple and it works .

 

[BritTwit]

Not quite sure I understand what you're saying but....

When cyl. 1's injector fires (or at some point in the engine cycle after it has fired) a large portion of that fuel passes through the IAC connecting tube into cyl. 2 causing the mixture to run very rich. The o2 sensor in cyl. 2's exhaust pipe reports this to the ECU and the ECU reduces the fuel coming out of injector 2 (wrongly since it's injector 1 that is causing the mixture to go rich). Meanwhile cyl. 1's o2 sensor is reporting it is lean. The ECU increases the fuel coming out of injector 1 to richen the mixture - but this fuel of course is just ending up in cyl. 2 instead. This causes an endless loop of corrections which ends when the max fuel trim percentages are reached.


You can see the maximum fuel trim setting circled - which is +/- 25% of the value in the fuel table for that RPM/TPS setting.

Hope this makes it clear.

iwilson, seems to make sense now so if you can just indulge this EFI dolt one more time:

You say ....a large portion of that fuel passes through the IAC connecting tube...

I was under the impression that only air passes through the IAC tube from the airbox, and that the separate injectors for each cylinder provided the fuel directly into the cylinder intake port.

Is this not the case?
How does fuel get into the IAC connecting tube?