K75 Throttle Response Hesitation Off Idle

[waterppk]

Hello,

I've got a 1990 K75 that was recently purchased.  It wasn't well cared for by its previous owner, having been crashed a few times and all of the maintenance was neglected on it (and it lived outside for the last 5+ years) so please don't assume that anything has been assembled correctly on it  :eek:

I've gone through it and replaced:
fuel filter
air filter
engine oil filter/oil
Breather hose (completely gone and open)
Spark plugs
Radiator fan, stock was rusted into a ball

And cleaned:
Entire intake air path, MAF was a mess, throttle bodies and hoses were very dirty inside
Fuel tank was pulled open and poked around in, hoses are good, pump whines but pumps fuel
Ultrasonic bath for the injectors, then 40psi compressed air and exercised the coils
Plug wire boots, one of the plug ends is pretty bad but doesn't seem to make any difference if it's pushed/twisted/wiggled while running so I skipped replacing it for now
Rebuilt all bodywork, screws were mixed and matched everywhere and mostly held together with wire.

Things I haven't done:
Check fuel pressure
Check injector spray pattern
Drilled a hole for an O2 bung to check mixture, smells rich but haven't checked it beyond my nose
Fixed the *small* leak in the exhaust where the header meets the muffler pipe
Balanced throttle bodies or verified mixture screws, everything is still covered in loctite/covers so I don't think they've been touched
Checked sensor values on coolant/MAF/etc

So here's the problem, from idle (~1000 RPM, lower tends to have it stall and die), if I open the throttle quickly it'll really hesitate, then pick up and rev.  If I let it return, and open the throttle it'll have the same problem.  At low idle it likes to die and/or stall.

I posted a video on YouTube of the behavior:


Any ideas are helpful, I have lots of tools at my disposal if necessary.

Thanks!

[TimTyler]

Your poor bike!! The previous owner should be punished!

I'd start with balancing the throttle bodies, adjusting the throttle position switch, and then check and adjust the valves for good measure.

[mystic red]

Nope, first thing I'd do is clean the injectors...then if that didn't work, what Tim said.

[waterppk]

Where's the throttle switch live/adjust at?  Cursory googling doesn't show much, but it did reveal what the big knob is for under the brake lever - cruise control!?!

[Photog]

There is a pdf of the BMW shop manual linked in this thread;

http://www.motobrick.com/index.php/topic,2252.0.html
The Haynes manual is also around somewhere online, but I don't have a pointer just now.

The throttle switch lives at the top of the throttle bodies at the back (towards the battery)
You should hear it click when the throttle is moved off idle. There is a thread in here discussing adjustment.
the knob at the grip is for holding the throttle open for whatever servicing where that would be needed (not sure what that would be, but there you go).
It's not meant as a cruise, and is really difficult to manipulate wearing gloves while moving.  :eek:

[JamesInCA]

Really, injectors and TPS? Doesn't it kinda sound more like an air leak?  Especially since the injectors have already had at least a cursory cleaning.

By "breather hose," do you mean the z-shaped crankcase vent hose near the back of the throttle bodies?

The big elbow coming off the air box can also crack and allow extra air in, as can the rubber/plastic bits above and below the throttle bodies.

Given the condition you've described, replacing all the fuel supply hoses is probably justifiable. Including the ones in the tank.

You did put the engine oil fill cap back on, right? It runs lousy without it; another place for air to come in.

My first guess, though, is rubber parts.  The hard parts on these machines seem pretty much indestructible, but when the rubber goes, all hell breaks loose.

[waterppk]

Alright, I've:

Checked and adjusted the throttle position sensor.  It now clicks on when the throttle is closed, and opens when the throttle is anything off closed.  Curiously, while it was at idle I unplugged and plugged in the TPS and got absolutely no change in idle.  Not sure if that's correct or not, but didn't seem to make any noticeable difference.

Checked the fuel pressure at the rear end of the fuel rail (hose going to the regulator).  I cranked the engine and it immediately hit 38PSI and stayed.  Seems to be within specifications.

I tried unhooking the crank case vent, I left it open on the engine side, and taped up the intake manifold end.  No change in any issues.  I also checked that the air filter is seated and the elbow from the air box to the intake manifold is free of any cracks or damage, it appears to be seated correctly.  The hose running from the throttle body to the fuel pressure regulator is a little surface checkered (I'll replace it ASAP, but once I get this hesitation issue figured out), unhooking it at idle didn't seem to change the response at all.

Oil cap is on the engine.

I replaced all fuel hoses previously except the ones in the tank, I didn't have submersion rated hose and it isn't flaky/checkered at all.


Questions:
There is a hole in the front of the air box above the filter and MAF, I taped it up with foil tape previously, is there something that's suppose to go in there?

There's a small vent port in the very front of the engine below the radiator, is that just suppose to vent to the ground through a hose or does it run somewhere?

Any way to easily check the MAF?  I had it out and cleaned it when I was cleaning the intake path previously, seemed to snap closed as I would have expected.



Next steps seem to point to throttle body air balance, mixture screws, and poking at the MAF.  Any other thoughts?  I'm not opposed to changing the injectors but it seems like they're working.

[TimTyler]

You should include some pictures in reference to your questions.

[waterppk]

Hmmm, likely found the culprit.  The fuel tank looks like apple cider, looks like the tank filler-port gasket has been letting lots of water in :-(

What's the proper way to install and seal the tank to prevent water infiltration?  Looks like a dry rubber seal.  Anybody sell replacement seals or?

[2000rsv]

http://www.maxbmwmotorcycles.com/fiche/DiagramsMain.aspx?vid=51739&rnd=08102012

Look here for parts.  Everything is here including the in-tank fuel line.

[WayneDW]

Hmmm, likely found the culprit.  The fuel tank looks like apple cider, looks like the tank filler-port gasket has been letting lots of water in :-(

What's the proper way to install and seal the tank to prevent water infiltration?  Looks like a dry rubber seal.  Anybody sell replacement seals or?

Make sure the gas cap mechanism is not turned around.  When you open up the flip lid there is small drainage hole on one side. That hole should be on the left, if it's on the right it's turned around 180degrees.
The water drain then runs through the tank and exits at a rubber hose, right rear of the tank.
The rubber gasket is available at BMW, but I'm betting its the drain thing.  Also, if the gasket was installed 180degrees off it would block the drain even if the cap is on correctly.  You should be able to see through the hole, through the gasket and into the drain pipe.

[waterppk]

Make sure the gas cap mechanism is not turned around.

This ended up being exactly the issue.  The prior owner had replaced the fuel pump at some point and must have installed the fuel filler backwards (which I carefully reinstalled the same way when I changed the filter).  It's funny they didn't offset the bolt pattern or something to prevent reverse installation given the importance of the drain.

I cleaned out the tank and drained everything else, fresh fuel fixed it up.  It still hesitates when I crack the throttle quickly but I might be use to those Japanese engines, I'll continue to work on it (and post any progress along the way).  It's now rideable!

[JamesInCA]

Good work!

Some people disconnect or remove the throttle position switch to reduce that bit of hesitation when opening the throttle quickly. The FI computer reduces fuel flow when it sees the switch is closed, so there's a brief lag after the switch opens when it's increasing flow. The trade-off is more backfiring on deceleration.

[rbm]

There is a hole in the front of the air box above the filter and MAF, I taped it up with foil tape previously, is there something that's suppose to go in there?

It's a port to get access to the CO2 adjustment on the MAF.  Normally there is a plastic cap that covers that hole.
Adjustment Procedure

[Kyle10]

"By "breather hose," do you mean the z-shaped crankcase vent hose near the back of the throttle bodies?"

That's prob it. Letting air in that's not 'noticed' by the oxygen sensor.
I found mine to be rotted, cracked. Replaced it and the hesitation was no more.

[MikeisonTV]

Waterppk,

I have the EXACT same prob on the 91' K100LT I just purchased last week!!!!! I've begun changing the same things you have, but u are a lot further along the list than me.


This was very helpful for me!
http://hrsbstaff.ednet.ns.ca/bvogel/K100/download/bike-wont-start1.htm

NEiTHER OF US HAVE MESSED WITH THE FUEL PRESSURE REGULATOR !!!!! This may be the answer!

[Grim]

I am having this same problem intermittently when not up to full operating temp on my 95 1100LT. Last week I found a fouled plug. New plugs and no other problem till today. Today it was acting like the vid but cleared up as it warmed up. I need to pull the plugs to see if I'm rich on the mixture.

I'll toss in my findings so collectively we might solve this.


The FI on these bikes has been around in one variant or another since the 60's. it's not a whole lot different then what was on Porsches and VW in the early 70's. Even my 86 Toyota 4runner (ND builds it under license of Bosch) has this same basic injection set up as did my first car,73 VW wagon. 

Some basic info on how the FI on these bikes function for reference. The FI on these bikes use the MAF to adjust fuel mixture. ANY air/vacuum leak after the MAF is going to cause a problem.

The TPS main purpose is to let the ECM know that you are decelerating. It backs way off the fuel to prevent back fires on deceleration.

At idle or when accelerating the O2 sensor (if equipped) does nothing. It only comes into play at steady cruise when the measurements on the TPS and MAF equalize.

The fuel pressure regulator: At idle or steady cruise manifold vacuum will be high and will cause the regulator to drop pressure on the fuel rail. When the throttle is opened manifold vacuum drops and the regulator increases fuel pressure on the rail causing the injectors to supply more fuel.




To the OP.

knowing the above the first thing to always check for is vacuum leaks.

There is some lines you may not know about.

On the bottom of the fuel tank is two pipes near the middle (in addition to the main fuel lines near the throttles) where the sender for the gas gauge mounts.  The rear line is the drain for the hole up around the gas cap. It connects to a hose that drains by the right foot peg. The forward hose is a EVAP connection. It connects directly to the crank case and may have an inline valve. They are often forgotten when installing the tank. If that line is leaking or not connected its sucking air in after the MAF and will throw off the mixture.

 That EVAP line is often taken off and plugged at the engine. The reason is an over filled tank can dump raw fuel into the crank case if it decides to vent.  :yow

Search "smell fuel" or smell gas" and there is info on the" AIR ACCUMULATOR CUP". Worth doing. I'll be ordering the cup ($3) soon.
Here is a couple good posts about that line
http://www.motobrick.com/index.php/topic,4117.msg25144.html#msg25144

http://www.motobrick.com/index.php/topic,989.msg3961.html#msg3961

[JamesInCA]

At idle or when accelerating the O2 sensor (if equipped) does nothing.

What O2 sensor?

It only comes into play at steady cruise when the measurements on the TPS and MAF equalize.

The throttle position switch only has three states: On (low) - Off - On (high). The air flow meter measures air flow along a continuous range.  How can the TPS and air flow meter "equalize"?

[Grim]

At idle or when accelerating the O2 sensor (if equipped) does nothing.

What O2 sensor?

It only comes into play at steady cruise when the measurements on the TPS and MAF equalize.

The throttle position switch only has three states: On (low) - Off - On (high). The air flow meter measures air flow along a continuous range.  How can the TPS and air flow meter "equalize"?

First let me explain how I treat tech posts. My goal is for it to be a resource that has a completion. Nothing pisses me off more is a bunch of dead end posts that dont list the ultimate resolution. So I really try to end them.

This is going to be up there forever. So the next guy that finds this may or may not know the basics so I put them up there to get them some basic understand how the different parts related to one another.

O2 Sensor:

Note I put in"if Equipped" and 90's Bikes like my 95 1100 they are so equipped. I put that up there so that people with O2 Sensors will understand when it comes into play and in this instance to ignore it for these symptoms ;)

The TPS on the variants used on cars its a potentiometer not on/off but it does serve the same purpose to cut fuel on Deceleration.

The manuals for this bike (both the Clymer and factory) SUCK! I've been spoiled by the Manuals for my Toyota's. Example : Testing a AFM http://www.4crawler.com/4x4/CheapTricks/AFM/index.shtml
Testing the TPS
http://www.4crawler.com/4x4/CheapTricks/TPS/index.shtml
There is full diagnostics instructions with a VOM with the values so that you can check every aspect of the AFM, TPS, and all sensors. I'm not seeing this in ether manual for this bike. :(   

I will dive into this weekend on mine. Maybe we can do some extended diagnostics cross referencing better written manuals for other vehicles since variants of this FI is used on so many vehicles.

The TPS can be unplugged if this operates like it does on other vehicles and if the problem doesn't go away it may rule it out.

Back to the problem at hand:

Today was cool like it was last week when mine did this the first time. In my case my bike started and idled fine. Being the bike was cold I was being easy on it as I crept out of the lot . As I attempted to pull out onto the road I gave it a decent bit of throttle when it fell on its face and then it started acting exactly like the OP's video.

Last week it was even colder and I got a lot farther down the road before it did this. Today I got it to rev up and held it at about 2500 rpm till it started to clear. Then it came back around and got me home seeming to run Ok the rest of the way.

 If I cant find an issue with vacuum leaks my next plan of action is to test the Ambient air temp sensors since both time I have had this happen it was cool and this happened before it achieved full operating temp.  Looking at Clymer they show temp sensors in the front of the AFM just like my Toyota and one in the side of the air box.  Hopefully its the same values. See where that gets me.

I haven't had the air filter out...sure hope I dont find a K&N. That oil is known to foul AFM's. I hate them damn things.  It looks like the AFM hang electronics down. That oil could wick into electronic workings and foul it.

[JamesInCA]

I heartily agree about unresolved mechanical posts.

Your K1100 has the Motronic fuel injection. The K75 bikes -- all of them -- have the earlier LE-Jetronic.

Different beasts. The LE-Jetronic has no O2 sensor. Three-position TPS, not a potentiometer.

Yes, there are common principles. But by mixing up diagnostics for different systems in a single thread, I'm not sure clarity is served.  A K75 owner will never have bad input from his O2 sensor. Likewise, a K1100 owner might or might not find the resistance for his thermostat on pin 10 of the FI computer connector. I have no idea.  The point is the diagnostic pathways are different.

[Grim]

Had my Puzzler on this the last couple days.

Couple things that we have in common. Both our bikes had some neglect and been in the elements for a while.

MikeisonTV what about yours?

Got thinking about some known problems I have read about. One of them is an issue with the Kill switch going bad.

See the connection to being in the elements?

Story time

Once upon the time my mother had a car I refer to as the devil. 1986 Jetta, it was stolen before it got its tag in the mail with less then 100 miles on the clock and unfortunately it was recovered. 
Working on that car I have assured my place in hell for the curse words I have made up and taking the lords name in vein.  So now I side with the atheist hoping that it will mean I wont sped eternity in hell working on that car.
The car had a intermittent problem were it would just die. Take off from a light and it would fall on its face. 2 minutes later its running fine. It would go for days and not do the problem and then do it 5 times in 10 minutes. I couldn't duplicate it with any consistency. I couldn't solve it.  I did everything I could think of to fix it short of throwing endless parts at it. I gave up....considered insurance fraud. 

So mom started paying other people not to solve the problem. Fuel pump...didn't make it out of the parking lot after that mechanic declared it fixed and charger her $400 for that. Half of the shops in Marietta GA worked on this car.

To make a long story short my poor mom spent more on that car trying to get this fixed then it was worth. Finally found a reputable guy that specialized in VW's handed him the keys and went and got a rental. He kept it for 2 weeks and used it as his parts runner trying to track this intermittent problem down. Every time it would mess up he would swap a part from a couple shop parts cars he had. When it would mess up again he would put the original part back on and swap something else. He was going bald ripping his hair out from this car. Finally swapped out the fuse box and the problem went away.
Sent mom home with the fuse box and only charged her $200 for the work but wouldn't promise it fixed because of how intermittent it was.  I took the fuse box apart and what I found was the contacts for the fuse for the pump had corrosion from electrolysis due to the high current draw of the pump and barely adequate design. The contacts would heat and the connection the the fuse would go away.

So here is what I am thinking.

We know that there is a known issue with the kill switch and I think it may be the problem and being in the elements contributed.

What I'm thinking is its not allowing enough current to the fuel pump. When you start to rev the engine the fuel pressure regulator sees the drop in manifold pressure and increases fuel pressure. That increases the load (current draw) on the pump. The pump cant handle the demand because of lack of power(amps), rail pressure drops and it stumbles.

Now go watch your video again and pay close attention to your battery light. Even when the RPMS are above 1000 that light starts coming on when you blip the throttle. Here's why once you get the RPMS up it keeps up, The Alternator has a HUGE swing of RPM it has to work at so its geared to not be over driven at high engine RPM. Well at 1000 RPM its not capable of making the full 50 amps. When that pump takes the load for extra fuel pressure the voltage will drop meaning it needs to make even more power (amps)....hence you see the light flicker. I'm also betting the Alternators field circuit is run through that switch.

Since your is doing it consistently (mine has done it once in a week so I'm not able to replicate consistently) here is what I think will prove or disprove this. Need by pass the kill switch and get power directly too the pump. Find an inline fuse of what ever the pump is fuse at. A switch and some wire. Run the pump (both hot and ground) straight off the battery bypassing the harness any relays and the switch.

 If I'm right the problem will go away.

[JamesInCA]

The fuel pump power does not run through the kill switch (though its relay is powered through the kill switch). The switched side of the relay gets power directly from the battery and passes it through fuse 6 to the fuel pump.

If the battery is in good health, it should be able to power the pump through alternator brownouts.

[JamesInCA]

If you want to power the fuel pump directly, to assess whether there's a problem in the wiring/relay leading up to it, you can get at its power either at its relay in the relay box, or by unplugging the tank connector on the right side of the bike and putting power there.

[Grim]

James:
Does the Field for the alternator also go through that same relay?



Had a chance to mess with mine today. Started fine. Not once have I had a problem starting just running when I give it gas.

Went to take it around the block and only got to the corner and then started exhibiting the problem.  I could hear the pump changing pitch and I also got the Charge light flickering on.

Limped it back home and almost didn't get it to pull up the drive back into the garage.

 Pulled the plugs looking for a obvious sign of an ignition issue and they had a brown cast but not wet like they had fouled. Changed a week ago. Darker then I would have like to have seen.

At that point we had party to run off too. 

Got home and decided to start looking at connections. Pulled the ECM plug. Took a Q-tip and cleaned all the pins on the ECM. took a fine jewelers screw driver and made sure I didn't feel any burned spots on the plug side. I did find one connector that didn't seem seated correctly.

Checked the connection on the oil and temp sender. One had the wire jammed against the housing in and odd way. Made sure that wasn't rubbing against anything. I as of yet have not had a chance to pull the fairing. I wanted to check the connections on the Hall sensor.

So I decided to fire it off and see if the problem was still there.

It disappeared again. I wiggled everything i had touched trying to get it to come back and it ran fine. Did a couple laps of the block and it didn't miss a beat.


James:
Is the sender for the temp gauge, Idiot light and ECM shared? My gauge seems to be functioning fine. I know on my cars its a separate sender for the gauge.

Like I have said its intermittent on mine. Its done it 4 times in 4 weeks. I was surprised it did it today.

So either it was some corrosion in the ECM plug  or Temp and pressure plugs and its fixed or I'll see it again in a few days.

With no symptoms I will have to wait.

[MikeisonTV]

Grimm,  sorry for being MIA as we hack at this beast.

Here is a video of my bike, with the same problem. I will note that once the bike is warmed up, the battery light stops coming on when it starts to choke as throttle is applied.



The original plugs were a little wet on the threads. So I purchased the NGK's as noted in the video. NOTE: I HAVE SWITCHED TO THE CORRECT BOSCH SPARK PLUGS, Please disregard the NGK one's listed at the start of my video as I learned they are for not for the 4 valve k bikes.

Since the video, I installed your most hated K&N air filter yesterday as I got one for $40 and don't share the same distain for them.  I also sent my fuel injectors off to be tested and cleaned at $15 a piece. If that is the culprit, I'll post the contact for info the servicer as thats cheap.

I also received this info on the k100 forum after a senior member watched the video:

I'm getting in to this conversation a bit late so forgive me if something I put here is redundant. My guess is that you are lean. I would disregard the TPS for the moment.

The video of the fuel movements in the tank may look suspect.  When fuel exits the filter, it should flow on to the tank outlet fitting down to the forward end of the injector log rail. Through the fuel rail to supply pressurized fuel to the tops of the injectors, excess fuel out the rear of the rail, around the rear end of the throttle bodies and into the fuel pressure regulator. The pressure regulator provides the needed flow restriction to maintain the line pressure at about maybe 38 psi. Once pressure is built in the rail, a spring in the pressure regulator lets excess fuel to... then leave the pressure regulator, through the hose back to the tank return port. I'm not sure if your 91 k100 contains a check valve to prevent fuel back flow when the hoses are removed for service or not. If it does, you may hear a slight buzzing or rattling sound at that area while fuel is flowing through back into the tank. (normal). Do not remove the check valve.

You must insure that there's no fuel squirting out at any of the fittings in the tank from the hoses for the filter, the filter itself, and that it's correctly plumbed in the correct direction of flow to the proper ports.

You need to install a (temporary) fuel pressure gauge in the hose between the fuel tank outlet pipe and the front of the fuel rail. Checking to see that you have about 35 psi - 50 psi. When you "Crack" the throttle.... vacuum in the sense line from the throttle body to the pressure regulator drops, allowing the regulator to raise the pressure. Hence, disconnecting the vacuum line to the regulator should cause a rise at the gauge. If all of that checks correctly, you could move on.
At Idle, You could spray something like carb cleaner all around all the rubber connectors between the airbox and the throttle bodies and between the throttle bodies and the cylinder head below them. I expect you have already tested the crankcase breather hose behind the rear of the throttle bodies for vacuum leaks. If all that checks out, You need to insure that the integrity of the large Elbow hose between the airbox flow meter outlet and the air intake plenum inlet are secure and leak free. There can be NO air leaks between the outlet of the Airflow meter (located in the top of the air filter box) and the manifold flanges at the top of the cylinder head.
Last, you should be sure that the oil fill plug is correctly installed & tightened so no air can enter there. If all that is correct, You should inspect the airflow meter itself to see that the rotating flapper door inside moves through it's entire range of travel open & closed with a nice light SMOOTH action with no tight spots or binding. I expect if all that can be checked & corrected the engine will be well.

As for the Ignition, I doubt if that is the problem. Just as a routine matter, You need to have the correct (non Resistor) spark plugs with the little screw on tips on the top. Beyond that the plug leads need to be VERY firmly twisted/pushed on to the plug tops with pliers and a LOT of hand pressure. These go on & come off really quite hard.
I'll be interested to see if you thoroughly re-check all these items and what is found.
Since it runs quite smoothly, I doubt if injectors are the failure but it's not inconceivable.
My number 1 bet is air ingestion somewhere between the air flow meter and the intake ports on the head. This would cause the airflow meter to not react correctly.
My #2 bet is low fuel pressure.
One should also verify that the Fuel injector main connector at the FI computer is getting the correct resistance readings from the coolant temp sensor. There are other portions of the forum that cover this.
I'm sure that others here may have a different view of what I've posted. Just trying to help.
Regards, ibj...
------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------
   
From RICK G:
+10 on that ibj
The fuel pressure is 32 -33psi and the pump should be capable of pushing 65 psi, you can test the pump pressure by clamping the outlet of the fuel rail.
I have seen a badly clogged filter cause big problems, it will idle without very much pressure at all as the fuel will get sucked out of the injectors with the high vacuum but as soon as you rev it dies. That plastic filter is good dumpster food and they usually explode. dry the plugs off with a flame or put them in the oven if your game.
The check valve was only up till 85 then the return goes right to the top of the tank.
If you put a piece of hose on the top of the pump clamp it then turn the ignition on and when you push the start button the pump will run for 1.5 seconds if the engine doesnt start but in that time hold your thumb over the hose end with it inside the tank and the pressure should be high enough that you have a very hard time stopping the flow and it should be a very healthy flow not a dribble, that is a good indicator that the pump is OK.
MAKE SURE THE HOSE IS POINTING INTO THE TANK it is very supprising just how much fuel it will pump.
also check the tube that goes from the lower rear of the plenum (black plastic box above TBs) to the engine breather if that is cracked and falling to bits a big heap of air gets in there.