Tuned S4Rs stutters below 3K, rough in steady cruising: not the coil, after all

[Moronic]

Thanks guys for the inputs.

CDawg, I'm not sure what you mean by the fuel sending unit? Do you mean the fuel pump? The injectors? The hoses connecting pump to injectors?

Ak1nza, it's very interesting that you are happy with the way your full Arrows bike cruises. Have you ridden it with the stock system in place, or did you purchase it with the Arrows already installed?

Yes, the shop has suggested I consider having them refit the stock exhaust and map to test their diagnosis. A lot of work but in the end I might go that way.

Meantime, I will shortly email an inquiry to Arrow in Italy. Perhaps they can shed some light on the problem.

[CDawg]

CDawg, I'm not sure what you mean by the fuel sending unit? Do you mean the fuel pump? The injectors? The hoses connecting pump to injectors?

That's what the shop called it.  It's the housing including the fuel pump, filter etc...

[Moronic]

Thanks CDawg for the clarification. I'll keep that one in mind, although I'd hope the shop would have found something like that, if it were present, when they tried it with a substitute fuel pump and filter.

[Ak1nza]

Ak1nza, it's very interesting that you are happy with the way your full Arrows bike cruises. Have you ridden it with the stock system in place, or did you purchase it with the Arrows already installed?

They were installed when I got the bike but did not have the DB Killers installed.  Definitely a lot of popping on deceleration without them in, and I much prefer the noise levels with them in than without.  Hope you get to the bottom of this!

[Moronic]

Thanks Akinza. No response yet from Arrow and it's been a week now. I may have found a similarly equipped bike to compare mine with, which will be interesting. I will keep you all informed.

A couple of other developments.

I've noticed that the "coming good" I reported earlier is still there with the fresher throttle bodies. It is just that the bad isn't nearly as bad, and the switch from good to bad is, if anything, better defined.

A tell-tale is the crispness off the bottom when, say, changing down for a slow corner, or engaging the clutch to pull away from a traffic light. When it's bad, it's quite obvious the bike is relying on one cylinder when responding to those first few degrees of throttle rotation. I need to exaggerate my throttle movement to bring in the rear cylinder and get the engine to spin up properly. In contrast, when the bike is good she is quite crisp off the bottom, with the usual small squeeze on the throttle doing the trick as expected.

The second thing is that I've bought a spare set of throttle bodies off eBay from a bike only 6000 miles old. (Great service and value from rubbersideup.com BTW.) Given the wear in mine at 35,000 miles I thought a spare might be handy, but I also wanted to look at whether that leaky spindle could be sealed in some way.

However, the cover-it-with-your-mouth-and-blow test tells me that on this unit anyway, there is no way air can get past the existing seal on the non-actuator side of the rear throttle body. The actuator side I can't test, and it makes sense that this is where most of the wear occurs. It looks like there is a seal of some sort on the spindle housing that could certainly wear. And there is no other way I could seal that.

However, it is hard to see how wear in that part could produce the sharp transitions between running smoothly and running poorly that the bike exhibits. Of course, it is possible that it could. 

[Moronic]

Okay I've just had the bike away solo for a 1000km long-weekend and those transitions between great and okay were fairly frequent and obvious.



Great also lasted a hundred km or more on various parts of the trip. Way too long for it to be attributable to that throttle housing leak having closed up, IMO. A leak of that kind should be fairly steady, you would think.

As well, when it was great the bike was running so damned well that it's very difficult to attribute problems to the Arrow exhaust. It's bloody wonderful. Produces a range of notes from the bottom to the top of the rev range and it is difficult to want any other exhaust. And the motor feels so smooth. (Thanks, Pro Twin.)

When it's just okay? We are back to the cruising speed stumble, but the not-so-good-ness is most obvious down low, where as noted in earlier posts the bike runs quite roughly. What is also obvious then is a change in the exhaust note: when the motor feels good, the exhaust note down low is richer and more rounded; when it's bad, the note is sharp and blaring. Leanness for sure.

What else? Well, the transition from okay to great often occurs after I've stopped the bike. As in, switched the ignition off and then on again. Given that on an S4Rs that mainly happens at fuel stops, it is easy to see how I was happy to connect the "good" with refuelling.

So what has occurred to me with a great deal of insistence is that the issues I have been having are very likely connected with the stepper-motor fast-idle system. Either it's developed a fault all by itself, or it's not working properly with this exhaust.

The valve is supposed to cycle automatically whenever you switch the ignition on. The possible volume of an air bleed through the valve would likely dwarf that through the spindle housing. Transitions could be sharp-ish if the valve was opening and closing. It would be impossible to tune out the effects of the air bleed by adjusting the fuelling, because the volume of air would be relatively large and changeable. And the valve now on the bike is the valve that came with the substitute throttle bodies, which alone could have produced an improvement in the way the motor behaved.

Pro-Twin tried running the bike after closing the valve and unplugging it from the electrical system, and also attempted to clamp closed the connected hoses, with no improvement. I assume this was on the dyno. They said they did not disconnect the system and plug the ports, and try riding the bike like that.

So that is the next step I'll be trying, and I have a feeling this one will do the trick. Can't see what else it could be.

Comments welcome.

[Moronic]

Okay, some fresh info on this. And still no solution.

1. Disconnected the stepper motor and fast-idle hoses, plugged the spigots. No improvement. Damn. 

2. Spent a day on the road with an acquaintance who has an identical model S4Rs to which he had fitted an identical Arrow exhaust. His bike was purchased used, on stock exhaust but with no O2 probe fitted (I am guessing a reflashed or DP ECU, owner doesn't know). The Arrow was simply bolted on, with no retuning. I rode his bike. Findings:

a) His bike was beautifully crisp off the bottom and stayed that way.

b) It showed none of the switching between good and not-so good that mine shows. (None reported either.)

c) It misfired a bit in the 3.5-5K rpm range. Never felt as laboured there as mine can, though.

d) It used much less fuel (about 20 per cent less, identical use, range of speeds).


All this supports my thought that it's not the pipe, and it's not just the throttle spindle. But if it's not the fast-idle system either, then what?

Leak at the vacuum gauge port? Cracked rear manifold? Anyone ever seen these?

[Bill in OKC]

2. Spent a day on the road with an acquaintance who has an identical model S4Rs to which he had fitted an identical Arrow exhaust. His bike was purchased used, on stock exhaust but with no O2 probe fitted (I am guessing a reflashed or DP ECU, owner doesn't know). The Arrow was simply bolted on, with no retuning. I rode his bike. Findings:

Are you saying your bike does still have the O2 sensor connected?  I *think* that could be a problem if so. My bike never ran right with the O2 sensor connected, ran better with it disconnected (default map I think).  When the DP ecu was installed, I was told that if the O2 senor was present the DP ecu would use it.  With the stock ecu (non-flashed) you will get a check engine light when the O2 sensor is disconnected.  Worth a try for testing purposes anyway.

[Moronic]

Hi Bill, no, my bike has been dyno tuned and reflashed specifically for this exhaust (using Tuneboy software).

That included removing the O2 sensor and switching off that part of the ECU that would look for it.

Good thought though. Thanks for your input. 

[Moronic]

Quick update to say that I have finally sorted this problem - I think about a year after I got the work done.

It was the f*%@ing rear cylinder coil.

Yeah I know that was supposed to have been checked. But there's many a slip, etc ...

Since my previous post on this I had fitted a parallel earth cable to a new earth point using the Motoelectrics kit, and also cleaned up the stock earth point. That helped, but the switching between sweet and laboured running remained.

On an overnight run to test this, there was a period soon after start-up when she dropped a cylinder. After a minute or so of running on one she came back onto two. And had run on two ever since.

But as I had run out of new options, it appeared the next step was to try stuff that had already been done. And as one obvious source of a drop onto one cylinder was the coil (plugs being quite new), I decided that was the next step.

The front coil was also new. So I replaced the rear, thinking to myself that I had just thrown another $250 into the wind.

Well whaddaya know. She started better, idled more smoothly, produced a quieter, richer note, ran nicely around town, and on a 350 mile country run completed a couple of days ago ran smoothly and eagerly at cruising speed - pretty much like she had at the best of the "good" times.

Yes I am pissed off that such a simple problem led to this much heartache, grief and expense - and not even just for me.

Yes, that is very much overidden by delight in the bike now running properly.

I'm not sure whether the coil failed on the dyno, or whether it had been on the way out but still doing the job with the stock exhaust. It could be that small-throttle combustion conditions are significantly more challenging with the 50mm Arrow headers than with the stock pipe.

I think she is now running a bit rich, so will look into that.

Thanks to all those who offered help and support. 

[Howie]

A completely failed coil is easy to diagnose.  A weak or intermittently failing coil not so easy.  There are two ways to pinpoint a weak or intermittent failure.  One is what you did.  Replace it.  The other requires an oscilliscope, something rarely seen in a motorcycle shop, not sure why or a coil tester, which seem to have disapeared from the industry decades ago.  Oh, a third.  If the failure happens often and long enough, swap coils from cylinder to cylinder.  Glad you got it fixed

[Moronic]

Unfortunately I need to extend this thread and reveal that after more riding the intermittent poor running has returned. It appears the fitting of the fresh coil merely coincided with a period when the bike was intermittently good. 

Since then I have fitted a fresh ECU (thanks Brad Black) and determined that the ECU is not at the root of this either.

I also tried disconnecting the coolant temp sensor that supplies the ECU, to see whether it was intermittently failing to tell the ECU that the bike had warmed up. I did this at a time when the bike was hot and running poorly. If it hadn't run any worse, then that sensor would look like the culprit. It ran much worse. So, not the coolant sensor.

What is left? The air pressure/temp sensor I replaced early in the piece. Pro Twin regapped and then replaced the crank position sensor, to no avail.

An injector might be faulty, but it's a funny sort of faulty injector that displays the same low and mid-speed fault for 10,000km and still allows the bike to rev out and make quite good power at the top.

Short in the wiring loom? Possible. I am hoping not.

Next thing I am going to try is the throttle position sensor. An intermittent fault there seems to match the symptoms I am experiencing, as far as I can tell from reading around on the net. There was a TPS glitch when the new ECU went in: initially a fault wouldn't clear, and then after a few tries and a swap back and forth with the old ECU it did clear. I suspect also that the TPS voltage signal can drift without triggering a fault report on the ECU: it would just think the throttle was moving.

So a fresh TPS is going in around the end of next week. Here's hoping ... 

[Bill in OKC]

TPS sensor sounds like a good candidate, can you take the connector off and see if the pins look shiny and clean?.  I have an S4Rs with Arrows and it runs pretty good.  I recently re-balanced the throttle bodies, they were out a lot.  It improved my idle and got rid of some popping on deceleration I was having.  The balance was good ?3? years ago the last time I checked them.  I also found that there is a tiny screen filter in the top of each injector.  There was not much there but some fuzz but I guess it didn't hurt to clean them.

[thorn14]

Is your tps the linear or non linear? I know the manually adjustable one can be checked with a multimeter, not sure about the other one.

[Howie]

They can all be checked with a multimeter.  Remove the electrical connector from the TPS.  Middle terminal in is power in, 5 volts.  Since the bike runs, you have that.  Figure out a way to back probe the two outside terminals at the connector  Attach a voltmeter to the outside terminals.  With the key on, move from closed throttle to open throttle at a very slow, steady pace.  Your voltmeter should read about 1 volt at closed throttle and smoothly increase to about 4 volts at open throttle.  Problem is your symptoms are intermittent, so you might test good when the bike is running well.