'98 EV Engine Dies At Speed - A Dope Tuned My Bike!

[Bill Havins]

Last year about this time Judy and I rode our '98 EV to the Albuquerque Balloon Fiesta - I posted a ride report after we got home and, in that report, noted that the TPS seemed to be acting up.  Over the winter I replaced the seals on the throttle body butterfly shaft and felt everything would be fine.  That's what I get for thinking!

Today we rode to Alejandra's in Ballinger, Texas, for lunch.  It was pretty warm, muggy, and there was a bit of a crosswind.  This was the "shake down cruise" for the new brake pads I had put on, and we were trying out our new Sena SMH10R headsets (they're great, by the way).

As we were riding home we slowed to go through a small town.  On the other side of the town I accelerated through the gears and, once in 5th and up to about 3750 RPM the engine died - it was like I had thrown a switch!  A fraction of a second later there was a loud backfire from both mufflers (Judy said it sounded like a rifle shot).  The engine did not "re-catch" when I rolled off the throttle.  I eased to the side of the road and switched off the ignition.  "Dawg!"

I waited for a minute and we discussed the situation.  We were 40 miles from home and the forecast called for rain (it's raining now as I type this - Yippee!).

I then tried to start the engine and it caught immediately and idled pleasantly.  Huh?  I set off on the paved shoulder and things seemed to go well.  I tried to keep the engine speed under 3200 RPM and we went on for ten more miles, all the while on the shoulder.  I finally said, "Okay.  Here we go!"  I accelerated and merged into the traffic lane but any time the RPMs made it up to 3750 the engine began to buck so I quickly eased off the throttle and kept things just at 3000 RPM.  We finally got to Tuscola and I pulled into a convenience store to fill up.  What the heck, maybe it was a gas thing.  The tank took 2.56 gallons.  I turned onto a side road and went cross country on paved county roads for the next 20 miles, finally arriving home.  We only experienced one more cut-out/backfire during the last 20 miles.

We are scheduled to leave for another back roads trip in New Mexico in October.  I told Judy we just couldn't go on the Guzzi (have to take the Nissan) if I can't get this figured out.

Some of you may remember me futzing with removing the Power Commander from this bike about two years ago.  When I was data-logging the performance of the bike with a multi-channel device I captured several instances (maybe three) where the RPM log would suddenly drop to zero; engine speed was inferred from the left cylinder injector.  Here's an excerpt from one of the data runs.  The dark blue line is engine speed, the red line is AFR, and the maroon line is throttle position (measured from the TPS itself).  The link to the .pdf under the image provides an "enlargeable" look at this chart excerpt.



http://havinsdesigns.com/images/datarun.pdf

I think I have a problem....

Over the winter I rebuilt the ignition switch and the kill switch.  I'll replace them if the general consensus is it needs to be done.  The relays were replaced on this bike with Panasonic units about two years ago - I think I trust them, but how do you know.  I have another TPS I can put on, but the fact that the engine died completely and would not catch when I rolled off the throttle (see above) causes me to believe I'm dealing with something else.

I'm thinking I have a slowly dying crank position sensor; there are two of them on this scooter.  Or, it could be the coil energizer box (or whatever the unit is called that sits under the left side cover).

I would appreciate your thoughts.  If we go on the scooter I have to have this done in less than two weeks.  Grrrr...

Thanks!

Bill

[nbags]

you may want to pull out crank sensor and see if it has metal shaving attached to it good luck

[Kiwi Dave]

Did you notice any tank suck when you filled up?

[fotoguzzi]

battery terminals clean and tight?

[Bill Havins]

Oh....


I cleaned this thing three(?) years ago?  From now on it's set the valves, clean the sensors, etc.

The front crankshaft sensor was clean and appropriately gapped.

Tomorrow after work I'll try to put 80 or so miles on the bike.  If things are good, great.  Otherwise I'll know to dig deeper.

And, yes, the battery terminals are swell, and there has been no tank suck.

This was obviously electrical - 70+ MPH, then , [insert sound of crickets chirping] followed by a remarkable backfire!  Just like throwing a switch!

It's hard to believe that the above can cause such a thing.  But I'm sure the ECU's software has a "engine not spinning -> cut ignition" line of code.  Perhaps I'll know tomorrow.

Thanks for the thoughts!

Bill

[Toystoretom]

I know nothing about Guzzis but if the crank sensor cut out would you get any RPM readings out of your tach? You would have to ride it until it fails and see if the tacho just drops to zero. OR... if you crank it does the tacho bounce and when it fails no bounce? I don't know if any of these are valid tests but what the heck its free. I'm not sure how you would test a crank sensor on a Guzzi other than to replace it. Does the bike's ECU allow for real time data on different sensors that you could do a little freeze frame action on and graph it out, or am I just dreaming here. How do the Guzzi techs do it? Do they have a scanner? Is this thing OBD II or some weird one off deal?

Sorry for that, just thinking out loud.

Tom

[Madtownguzzi]

Check the wire harness plug to the ECU to make sure that it is not loose. Mine would cut out over bumps and rough roads. I cured it be wrapping wire ties all around the ECU to hold the wire harness plug in tight.

[Kent in Upstate NY]

Funky sidestand switch?

[Kiwi_Roy]

To rule out electrical monitor the petcock fuse with a small lamp. if there's power there chances are the bike will run.
The petcock is connected after the ignition switch, kill switch, sidestand circuit and several connectors.
http://www.thisoldtractor.com/guzzi007/schematics/1998_EV.gif
If it cuts out again glance at the lamp, it will halve the possible trouble points.

[Bill Havins]

Thanks, Roy.  Will get that wired in tomorrow.

Kent - sidestand switch was disconnected many moons ago.

[fotoguzzi]

Did you know the 98 EV can display trouble codes? procedure noted in the FI supplement manual. 

[Bill Havins]

I'll take a look at the trouble codes.  Easy enough to do.

EDIT: According to the manual the trouble codes are erased as soon as the ignition is switched off.  So...  if it happens again I can check it.

It struck me this morning that this has only happened when we have been riding two-up and over an hour on the road.  And it only seems to happen in fifth gear.  Coincidence?

I moved the ECU to the battery compartment about three years ago.  I'll check the wiring loom.

Onward and upward.

EDIT:  Oops!  Just checked the battery connections - ground wires slightly loose.  What a dope!  But I need to keep looking....

[normzone]

I've nothing to contribute to your dilemma, but I just wanted to add that I cherish threads such as this one.

I own a similar bike (1999 Bassa) and know very little about it yet. I always read these threads and look forward to the solution, then file that knowledge away in the " if this ever happens to me, try this " lobe of my brain.

Thank you for sharing your learning curve here.

[Lee Bruns]

Oh....


I cleaned this thing three(?) years ago?  From now on it's set the valves, clean the sensors, etc.

The front crankshaft sensor was clean and appropriately gapped.

Tomorrow after work I'll try to put 80 or so miles on the bike.  If things are good, great.  Otherwise I'll know to dig deeper.

And, yes, the battery terminals are swell, and there has been no tank suck.

This was obviously electrical - 70+ MPH, then , [insert sound of crickets chirping] followed by a remarkable backfire!  Just like throwing a switch!

It's hard to believe that the above can cause such a thing.  But I'm sure the ECU's software has a "engine not spinning -> cut ignition" line of code.  Perhaps I'll know tomorrow.

Thanks for the thoughts!

Bill

That blasted crank sensor was one of my regular maintenance items after I found out that the clutch crud likes to hang on to the sensor. JUST like in your photo, but the first time I checked it, it was far worse than that. Now its one of the first things I suggest owners check.

[Sasquatch Jim]

  Don't put the sidestand down at cruising speed.  Or just check it for normal functioning.

[Bill Havins]

A little update.  I cleaned the flywheel sensor, and cleaned/torqued the battery terminals/cables.  I pulled the ECU plug and did the DeOxit treatment on all of the terminals.

The bike started easier (almost instantly once the engine had warmed).  I had occasion to do about a forty mile high speed run this afternoon and the bike ran very well.  No evidence that the engine wanted to quit at any sustained engine speed.

I'll do a routine service this weekend (fluids, set valves, check TPS and throttle sync) and I'll take another look at the flywheel sensor to see if it has collected more swarf.  On Sunday Judy and I will retrace our route from this past Sunday and see what happens.

Fingers crossed....

Bill

[Bill Havins]

Okay.  So I’ve changed the title of this thread to include the confession that “A Dope Tuned My Bike!”  Oh boy!  It’s a wonder I can brush my teeth by myself…

I have been really frustrated that I haven’t been able to tune my EV to my satisfaction.  This is after removing the Power Commander, trying stock mufflers in place of a pair of Bubs, and then finally attaching a stainless steel crossover and pair of mufflers from Keihen Systems.  My gas mileage has continued to be bad – that’s B-A-D, bad.  And my spark plugs have read really rich.  Then there was my apparent problem with the TPS on our trip to New Mexico last year (gas mileage was bad), and the recent incident of the engine dying while at speed (see original post).

Now, I’m no mechanical neophyte.  The Guzzi I had before this one was a 1987 SPII that Harper’s had just taken in on consignment.  When I brought it home it wouldn’t stay running at idle.  It didn’t take long to figure out that the original owner had tried to hot-rod it and had jetted it way too rich.  Really rich!  Rich enough to…  you get the picture.  In just a few days I had it running like a top and averaging 47 MPG on the highway.  I like carburetors.  I understand them.  But this fuel injection thing.  It has taken me a while to even begin to understand it.  So here’s where I am with my EV at this point.

I spent the last two days changing all of the fluids in the scooter.  After cleaning the flywheel sensor (see earlier post with photo) I began to “retune” everything.

First, I set the valves.  Now, over the past couple of years I have stopped trying to see the timing marks on the flywheel and have used the popular “find top-dead-center with a soda straw” technique.  Gong!  That was my first mistake.  This time I used the timing marks on the flywheel and, once I was sure I had them aligned exactly, I found all of the valves were off.  The intake valve on the right cylinder had no measureable clearance.  Holy cow!  None!  I asked myself, “How could that be?”

Here’s how.  Using the soda straw technique it is possible to be off of TDC by several degrees of crankshaft/camshaft rotation.  The movement of the piston is not linearly related to degrees of crankshaft rotation – the piston moves less when the connecting rod big end is at either the top or bottom of its stroke.  The piston moves more per degree of crankshaft rotation when the connecting rod big end is in the middle of its stroke.  This can lead to valve settings that are way off.  So, I gave my soda straw to the kid across the street who was drinking a soda out of a can – I won’t be using it anymore.

Next, I reset the throttle bodies idle.  Per Guzziology I set the throttle bodies bypass screws to “one-half-turn out” and, with the throttle linkages removed, set idle to 1050 RPM with intake manifold vacuums equal.  I then adjusted the Throttle Position Sensor to 375 mVolts (measured across pins 11 and 17 of the ECU connector) and then readjusted the idle set screws to achieve an idle of 1050 RPM.  I continued to adjust the TPS/idle set screws until I had a “perfect” idle of 1050 RPM (+/- 50 RPM) with the TPS set at 375mVolts.  It took three “rounds” of adjustment to achieve this “perfect” idle.

I then re-connected the throttle bodies linkages and synched the throttle bodies at 2800 RPM without touching the idle screws or the TPS.  I then adjusted the “choke” lever to achieve a “fast idle” of up to 4500 RPM when fully engaged.

Things were way off when I started the TB synch.  It was W-A-Y O-F-F!  Don’t know how.  But it was.

Near dusk I took the EV for a very spirited run up the highway at speeds in excess of 70 MPH.  When I got home I checked one of the plugs.  It had burned off most of the black that had been there.  White was beginning to show through on half of the ceramic insulator; the other half was a “deep caramelized brown.”  I’ll take that until I can install new plugs on Wednesday and try another spirited run.

Regarding the chart I included in my original post – I found this entry in the Guzzi Fuel Injection Manual for the ’98 EV:  “When the throttle is closed while the engine is working at high level, the fuel supply is cut off for a period.  This increases engine braking and reduces fuel consumption and emissions.  Deceleration is detected by both throttle position and engine speed.”

Okay, then.  That may be the explanation for the sudden “dip” in the blue line in my chart.  The chart may have nothing to do with my engine quitting at speed.  It may reflect an instance where the ECU simply stopped the injector for a moment in response to deceleration.

I’m glad to wear a “dunce hat” for a bit.

[fotoguzzi]

 

First, I set the valves.  Now, over the past couple of years I have stopped trying to see the timing marks on the flywheel and have used the popular “find top-dead-center with a soda straw” technique.  Gong!  That was my first mistake.   

could you have set them on the exhaust stroke? the straw method works fine as long as your on the compression stroke.

[rocker59]

In my experience, "the straw method" will get you a lot closer than the stamped "D" and "S" marks on the flywheel.

[rocker59]

Did you mean the small line marks next to the D or S???

Just curious

mike

I don't recall seeing the lines. 

[Kiwi_Roy]

So Bill, I read through the post again looking for what was the problem.
Being a little out of tune wouldn't cause it to suddenly cut out would it? Or do you think the plug sooted up so bad it just wouldn't spark?
IMHO the weak part of these bikes is the wiring leading up to the ECU (Ignition Switch, Kill Switch, Stand circuit and various connectors) once it gets to the ECU the wiring is pretty robust, the ECU receives power directly from the battery through a pair of dedicated fuses. 
An easy way to monitor the safety circuit would be a lamp across the two outside pins of the ECU relay.

[Johnny Moto]

I had similar problems a month ago on my 98 V11 EV. Temperatures were in the mid to high nineties, and humidity typical of south Texas. Traveling at 75 MPH, 4,200 RPM, the engine died like a cut-off switch had been thrown. It happened twice, restarted easily, and I got home safely. The problem reoccured two days later. Not knowing where to start diagnosing the problem, I cleaned every electrical connection I could find using DeoxIT D5 and Sheild S5. I started with the battery, grounds, fuses and relays, and worked out from there.

While I did not find any obvious culprit, in fact the connections all looked clean, the bike seemed to fire up with a bit more gusto, accelerated from a stop with significantly less throttle, and now -nearly a thousand miles later- no reoccurance of the sudden death. Go figure.

[LowRyter]

My EV has been doing the TPS cough, skip, stumble and backfire at about 4k rpm.  The TPS was changed out and it ran fine for awhile and then got progressively worse all over again.  Checked the new TPS and it has a spike in the voltage curve.  Contacted the vendor and replacement is on the way.  I think these replacement TPS parts are cheap made in China quality stuff.

I am still afraid the TPS is only part of the story.  I've been checking connectors and fuses and replaced all the relays.  But the first step is change it out.  Hopefully the new part will be delivered in the next few days. 

[Bill Havins]

Roy,

I don't have any idea what caused the engine to stop.  I'm going to chalk it up to a really dirty flywheel sensor combined with a very slightly loose ground on the battery.  Or maybe my Windows 10 upgrade didn't have the right driver for the ECU...or maybe it "obsoleted" the ECU.  Surely Bill Gates is responsible. 

I just picked up new spark plugs at my favorite motorcycle parts counter.  When I install them tonight I'll shoot photos of the lines on the flywheel and double-check that they match TDC (if the kid across the street will give me my soda straw back).

More later.

Bill

[jim_W]

My TPS started acting up on my way back from Iowa to AZ . Middle of no-where. would idle ok and run well at full throttle but would not cruise at 3500 to 4000 rpm. First I thought I had bad gas as I had fueled up at some way out of the way places.
Found out a TPS from an older  Road King is the same or nearly the same as my CALVIN and probably other EV's.
Part number 27271-91 at the Harley store got me right back on the road and it runs better than it ever has.
  Read the voltage (600mv) at the idle stop and set new one to same voltage.  Screws are hard to get at but a modified 20T bit got it off and back on.
  Probably not the correct way to set it up but it got me back on the road and is running great.
 

[Bill Havins]

Here are the flywheel marks for TDC.  This is the right cylinder mark seen from about a foot away:


Close-up of the right cylinder TDC mark.  Notice how sharp the line is.


And here is the left cylinder TDC mark (gotta' stop drinking those double espressos before operating the laser, Luigi....).


The "puffy" look to this line is actually paint I applied some time back - this one was really hard to find before I added the paint.

There is another mark on the flywheel that is found maybe 10 to 15 degrees before you get to the right cylinder TDC mark.  It should not be used for setting the valves.

Bill

[Kiwi_Roy]

I don't believe TDC can be that critical for setting the valves, they are closed for the major part of the cam, a few degrees either way should make no difference.
Aren't the lines more for setting the static ignition timing?

Perhaps I have been setting my valves wrong all these years

[Bill Havins]

Finally!



Time to stop bellyaching and go for a long ride.

Bill

[fotoguzzi]

I don't believe TDC can be that critical for setting the valves, they are closed for the major part of the cam, a few degrees either way should make no difference.
Aren't the lines more for setting the static ignition timing?

Perhaps I have been setting my valves right all these years

your right Roy and your valves have been happy all along..

[Bill Havins]

Re: the soda straw method.  I haven't been able to get it to work well.  I'm sure it works fine for most.  And Roy, I would never suggest your scooters aren't tuned well.    I'm just really concrete and feel more comfortable with an indicator rather than trying to perceive the moment just before a soda straw starts to fall.

Bill