The ignition switch is a good thing to check.  I had much trouble with mine last year on a bike that had always been garaged and was in otherwise pristine condition. 

I suspect that the problem is due to a design that puts very little contact force on the internal electrical contacts, making them somewhat sensitive to the slightest bit of dirt or corrosion. 

I suspect the reason you see it more at low rpm's is that the vibration of the engine is less as the rpm's increase, shaking the switch internals less.  As the problem progresses you will have the pleasure of having the electrical system shut down momentarily, usually on the highway at speed, which is what finally convinced me to go in and clean the switch.

Point-Seven-Five wrote:As the problem progresses you will have the pleasure of having the electrical system shut down momentarily, usually on the highway at speed

.... and its even more pleasurable when that happens at night time. Throw in a corner and you'll be having a real hoot.

Don't delay. Its a common problem and well documented here on the forum.

that curve looks pretty good to me (although there is a slight dip at the end , but well out of the working range expected )
and about the right starting volt measurement

I wouldn't be going in there at all.... yet !

I would suggest there are other things that need to be sorted before even thinking about it .

just a  question , did you notice if there was any movement of the barn door up or down , along the axis of the pivot pin ..?

I have seen slight lateral movement of this , just a tiny bit but it did not affect the wiper readings ...just wondering

btw  did you also measure the temperature sensor on the afm ?  forgotten the pin numbers at this moment
its just a test to prove that it is in circuit and works ...(they are quite sensitive , just blowing across it can produce good swing results )

good luck

Thanks for the feedback, Charlie

I did not notice lateral movement on the vane.

Engine Temperature Sensor

I took two readings (Pin 10), 0.9K after running the engine for a while and 2.6K after leaving it overnight.

Referring to page 18 on that wonderful TroubleShooting document, the .9K suggests about 50 degrees C and the 2.6K about 20 degrees.

The 20 degrees indicated by the 2.6K looks right. I measured the room temperature at that time and it was just short of 20 degrees.

I could not verify the 50 degrees. It looks a bit on the low side; I'd expect the engine to be warmer than that - could it be if the EFI thinks the engine is less warm than what it actually is, actual temperature say 65 and the sensor indicates 50, it would send more fuel, consequently faster idle?

Resistance measurements (AFM)


The Typical Values are from, again, that TroubleShoot document, page 25
My values, the measured values on my bike, seem okay except the 76 ohm value is a smidgen lower.






Does this information in anyway change the feedback you have given me up to now.

Have a super weekend

don't worry about it ...the temp sensor is deep in the airflow path ..which runs considerably cooler than the ambient temp of the hardware surrounding it given that air is flowing (and cooling )


the temp sensor in the afm is operating as a compensation circuit to make minor adjustments to the readings of the afm back to the efi computer ...considering the temperatures experienced by the air flow temperature ..its a feedback circuit 
so far the system seems to be showing all the standard expectations

keep at it  and of course... good luck !

Just to expand slightly on Charlie's excellent information.  The temperature probe in the air path is giving the ECU information that it uses to determine the density of the air.  Cooler is more dense than hot and thus needs(or can handle) more fuel.  This is how the MAF sensor is able to accurately tell the ECU how much oxygen is coming into the engine for combustion. 

As the AIR temperature INCREASES, the ECU will inject LESS fuel for a given flap position. 

This has nothing to do with engine temperature.  The mixture adjustment for engine temperature is done via the water temperature taken by the sensor in the radiator line at the top front of the engine block.  The enriching controlled by this sensor is the reason why the engine doesn't need a conventional choke.

The AFM vane, the little "barn door", oscillates while the engine is running - is this right? From what I've seen in photos of the inside of the AFM, I'd think that sweep arm will wear a hole in the contact surface. I'd say it moves +/- 3 mm from the average position, say a swing of about 5 mm.

At idle the oscillation frequency is probably lower. My guess is at high revs it would perhaps not happen.

How did I find this out? I removed the air filter and the lower part of the air filter box-AFM still connected to the air delivery. Then I stuck my finger into the AFM opening - don't ask what I thought I'd discover.

Besides the oscillation I made a further discovery: If I disconnected the AFM from the plenum and activated the vane with my finger I could position the vane so the engine revs remain at 1000, not the above 1250 it does with the AFM connected. The engine now sucks as much air as the closed butterfly valves permit - I'm guessing - but the amount of fuel injected is determined by the manually operated deflection of the vane.

NOTE: Throttle cable is disconnected. Fast idle is disconnected. TPS is removed - can't obstruct movement. I checked, the hose clamps do not touch the mechanism anywhere.

Based on the above, my conclusion(or maybe confusion is a better word) regarding the fast idle is:
Is it caused by the engine sucking too hard?
Maybe the butterfly valves don't close completely?
Or maybe the butterfly valves are shut as far as they can go, but are worn and no longer restrict air flow sufficiently to permit a decent idle speed?
Or all the above?

Who can throw light on this?

My conclusion about the vibrating/oscillation of the vane is:

Surely the volume of air in the plenum chamber should act as damper and prevent the oscillation?

Can anybody throw light on that?

Have you tried disconnecting the throttle cable to see if its just the cable holding the TBs open slightly?

My RT had a (new) cable where the cable outer was a litle too long holding the throttles open.  Some very careful trimming back fixed it.

Al

I have modified the original post. I should have said in there that the obvious stuff I've taken care of as much as I know how - I added this note to the post:

NOTE: Throttle cable is disconnected. Fast idle is disconnected. TPS is removed - can't obstruct movement. I checked, the hose clamps do not touch the mechanism anywhere.

The post is a little long, so I probably missed some points.

Your only option now is to balance the Throttle Bodies to achieve a lower idle speed.

The flap should not move when at idle. There is a bypass around the flap so as to supply air at idle.
Open the bypass at the MAF up quite a few turns then balance the throttle bodies using the big brass screws (approx 11/2 turns out from stop) with the butterflys closed completely then when the idle is even with the same readings on each cylinder start closing the air bypass on the MAF until it falters then back about 1/4 turn. That way the flap should start to open as soon as the throttle is opened providing all the butterflys are evenly set.

RicK G wrote:The flap should not move when at idle. There is a bypass around the flap so as to supply air at idle.
Open the bypass at the MAF up quite a few turns then balance the throttle bodies using the big brass screws (approx 11/2 turns out from stop) with the butterflys closed completely then when the idle is even with the same readings on each cylinder start closing the air bypass on the MAF until it falters then back about 1/4 turn. That way the flap should start to open as soon as the throttle is opened providing all the butterflys are evenly set.

Rick that sounds like a real possibility. It may be that the bypass passage on my AFM is blocked. What you say makes a lot of sense

Scenario: bypass is blocked or somewhat blocked - the vane will tend to open - more fuel because the EFI gets a "vane open" signal - even with closed butterflies the motor will tend to rev higher

I'll be taking out the AFM and see if that passage is clear

Rick


Where is that bypass air passage?

I took out the AFM bypass adjuster screw and I can't see what's at the bottom of the hole.

Here's my thinking; bypass air must be pulled from a) upstream of the vane or b) from outside of the body of the AFM, right? I cannot see any opening that would let in "unmeasured" air.

I did a very crude test as well; I rigged up the AFM, a resistance meter and Hoover nozzle on the bench. I measure resistance between pin 7 & 5, 0.076K with the vane in closed position. Then with the bypass screw seated(closed) I moved the Hoover nozzle to where the meter registered a small increase( i.e. the vane just starting to open), 0.085K, a bit erratic, but a very low increase. Then I gradually opened the bypass screw many turns. No change in resistance.

I don't consider this conclusive, like I said it is very crude.

I was reluctant to claim "fixed" before I'd been for a test run. I went for a 15 minute ride. The motor was nice and hot and it idled on the 1000 rpm mark. Those who have been plagued with warm engine idle revs around 1500 to 1700 rpm and that you just can't adjust down will understand my joy

Thank you to all who offered suggestions and ideas.

It has been a long search. From the start I suspected the throttle valves were not shutting properly. With throttle & fast idle cables removed, idle switch aka throttle position switch removed, no physical interference from those TB hose clamps and the stop screw not touching the stopper, idle revs would go higher as the motor warmed up towards that very noisy 1700 level.

I removed the TB assembly and, on the bench, adjusted those never-to-be-touched inter-TB adjusters. I will post a separate thread on how I did that. It may be that one of the throttle valves was bottoming out and preventing the others from closing or maybe just wear - it's an old bike.

Knowing what the problem is, is important, but equally so, in my opinion is knowing what didn't cause the fast idle.

You will be familiar with the drawing from Bert's document below.  Except for the motor and the Hall sensors I swapped out every component shown in that drawing - I have another K100. I also tested the AFM, the coolant temperature sensor and air temperature sensor. I replaced all rubbers between the TBs and the AFM. The breather pipe was broken; replacing that helped some, but the rest made no difference to the fast idle.


So that's the short part of the story. I'll do another thread later on how I did butterfly valve adjustment. I will also post stuff I discovered about my AFM

Thanks again for your advice and ideas.