Power cutting out, advice wanted please

Kiwi_Roy wrote:

GuZeee wrote:

Kiwi_Roy wrote:

I asked about the kill switch, if you operate that while riding does it look the same, lights etc.
As for "tank suck", I did clear a plugged breather hose and it didn't make a difference. I never feel any pressure release when I open the fuel cap either. But it does explain the symptoms better than some of the other possibilities... I wonder if there is an easy way to test for that while riding? (I could test if the problem goes away when the fuel cap is not snapped shut, but I don't really want to ride long distances in that state.)

Do you have a spare key, just crack it open when the symptoms occur

Yes, I forgot, I do have a spare key. Since the last test I've changed the coils and re-attached the main ground wires with lock/wave washers. This weekend I will go on a ride with the spare key in the gas cap. If it starts happening I'll pop the cap. If it immediately works again, we've found the problem! (And if it doesn't happen, it means: maybe it's fixed and maybe it isn't)

motor-timothy wrote:

I must say you do seem to have the worst luck with your GRiSO  Hope it turns out to be something cheap and easily fixed!

Thanks for the good wishes! I hope so too

Yes, I have had a my share of things to deal with since buying this bike 6 months ago:

A couple of things I would consider design flaws: flat tappets (replaced with rollers), rear main seal leak (had to drop engine, remove gearbox and clutch, etc.)

Some problems were self-inflicted: multiple issues with the clutch slave cylinder (now all resolved) were due to incorrect reassembly on my part.

There were some general wear things like the burned out headlight, that are expected to happen with any bike.

Then there are the two mystery problems: the alternator suddenly dying and shorting out (fixed by replacing the alternator), and then this engine cut-out problem, which is the worst thing so far because I can't figure it out and it's positively dangerous and scary when it happens at the wrong moment.

This is everything I've done to it so far (all self-serviced):

9/7/2017: 6980 miles
* New spark plugs and caps (NGK)
* Roller tappet upgrade "C" kit installed
* Adjust valves
* ECU flash with “beetle map”
* Change engine oil and filter
* Clean out sump
* Remove charcoal canister

9/22/2017: 7240 miles
- Throttle body balance
- Change gearbox oil
- Change shaft oil

Mistral high-pipe exhaust / new ECU map

12/20/2017: 7610 miles
- New alternator and belt
- Re-install rear main seal to fix oil leak
- Replace steering head bearings
- Replace swingarm bearings
- Rear tire (Pirelli Angel GT)

1/13/2018: 7700 miles
- Replace fuel lines

1/21/2018: 7900 miles
- Replace fuel filter

Also did some cosmetic mods along the way:
- Small windscreen
- Bullet rear turn signals
- Powder-coated gloss black frame
- Painted black front fender
- Removed rear foot-pegs
- Aluminum clutch and front brake levers
- Blacked-out tank logo
- Rizoma black foot-pegs
(The bike already had a tail tidy and bar-end mirrors when I got it.)

Avgpetro  mentioned the grounds, the schematic shows two separate grounding points, I would be tempted to add another perhaps at the ECU to an engine bolt. The ECU must be well grounded because it has to pull down the coils and injectors, probably in the order of 15 Amps but just for a  millisecond or two, it can't do that if it doesn't have a good connection. Loss of grounding might explain the loss of power without alarms.

With all the work you have done is it possible something was overlooked?

Earlier Guzzis used to have a ground wire running all the way back to the battery negative but I would hesitate to do that, it used to cause fires in the spine framed bikes if the main ground came loose.

I hope you applied some grease to the battery terminals and ground points.

Kiwi_Roy wrote:

With all the work you have done is it possible something was overlooked?

This is what I wondered, but I’m almost certain this problem first happened before I did any work on it. I distinctly remember power cutting out abruptly during hard acceleration in first gear. I then later explained this away by thinking maybe I had hit the rev limiter, but that doesn’t really make sense; I wouldn’t take it to redline by accident like that.

Also, before I did anything at all to it, at idle it would sometimes light up an error code and stall. After I did the rollerization and flashed the new ECU map, that stopped happening.

So I’m pretty sure this problem in some form has always been around.

Also, if I overlooked something I can’t imagine what it would be. There are not that many things to hook up, and I’ve gone over everything many times.

I hate to think it’s the fuel pump because it’s a very expensive part, but I wonder if that is a thing that is known to fail in this way?

It's certainly not common, I've never replaced a fuel pump.

Have you had the pump out and checked the filter and internal lines for cracking? I'm a bit lost on this one as without the bike to examine I can't suggest much.

An intermittent phase sensor fault maybe but I'd expect that to come up as a memorised ECU fault....?

Yes I had the pump out to change the fuel filter. I didn’t notice any cracked lines in there. And there are no stored codes.

Good to know the fuel pump is not a typical failure.

Are you 100% sure you've eliminated the side-stand switch as the culprit? I had this happen with my '07 1100 and that was the problem. At speed when opening the throttle to pass it would cut out. Scared the crap out me, but bypassing (disconnecting) the side-stand switch solved the problem once and for all.

Try it by all means but please replace the switch if it is the problem. I knew someone who bypassed it and forgot. It cost him his life.

The use of sidestand cut outs has allowed stands to be made less likely to flip up if left down, (Honda had a brilliant system with a rubber *Foot* would flip the stand up on a left handed if it was left down, back in the day. Those days are, unfortunately gone.).

Yes, we all think it's un-neccessary 'Nanny State' stuff. Until it kills someone we know.

Pete

I would have expected to see the very bright side stand switch indicator on the instrument panel if that’s what was happening. I guess it’s possible it is intermittent and the light is not getting triggered somehow or I didn’t notice it. I’ll try disconnecting the switch for my next test run.

Thanks for the insight, and also thanks for the warning Pete — the safety lockout has saved me enough times that I know I’m not immune to forgetting the stand before taking off.

Pete,

Can you elaborate? I'm not sure I understand the risk. My GRiSO Rosso is the first bike I've ever owned that had this feature. I never got myself into any trouble after the bypass, but maybe I'm missing something.

Pete

The danger is forgetting to raise the side stand before you ride off. First time you lean to the left, very bad things happen.

Right, I see that. I guess I never really thought about it. Raising the side stand has always been a function of muscle memory,  but I can see how it could be catastrophic if one were to have a momentary lapse of concentration. Of course, I never have.  Yeah, right!

You see? If we're paying attention,  we can learn something new every day.

GuZeee wrote:

The danger is forgetting to raise the side stand before you ride off. First time you lean to the left, very bad things happen.

I remember back in the day many moons ago when myself and two buddies were out on our Yamaha 125’s. One of the lads took off and we saw that he had forgotten to raise the side stand(no switch back then). We tore off after him to try and warn him.
Went around the bend and saw him on the road and the bike on its side stuck in the ditch, back wheel spinning. Luckily he wasn’t too badly hurt, I am not a big fan of a lot of the modern electric mod cons on bikes but the side stand switch is essential.

GuZeee, are you able to get it occur in neutral on its side stand? Or does it only happen under load? If it can be made to happen under controlled conditions, perhaps measurements or data logged may provide some insight.

beetle wrote:

GuZeee, are you able to get it occur in neutral on its side stand? Or does it only happen under load? If it can be made to happen under controlled conditions, perhaps measurements or data logged may provide some insight.

It has only happened under load while riding. However a couple of times it happened at low speed pulling away from a traffic light, and I couldn’t stop it from stalling, even by pulling in the clutch and giving it throttle. If it was the sidestand switch, pulling in the clutch should allow the engine to run (although it might have been too late at that point, I don’t remember the sequence of events exactly).

The tricky thing is that it doesn’t happen until I’ve been riding for a while, I can go 100 miles with no problems whatsoever and then it kicks in with a vengeance. I’m guessing it’s temperature related.

I have heard of the sidestand bouncing at certain engine revs causing the engine to falter but that would be accompanied by a light show as the ECU resets.
I don't think you see any response from the dash when it happens.
Still no response to the question, what happens if you hit the kill switch, does that cause any alarms or the ECU to reset when it's turned back on? Trying to narrow it down.
(my bike is parked up so I can't try it myself)

I haven’t tried the kill switch experiment yet, I’ll try it when I take it out on Saturday.

OK I went for a ride today after changing the coils, and securely fastening the main ground wires using a lock washer and wave washers (and cleaning with de-oxit). I had a spare key in the gas cap so I could test for tank vacuum issues if it happened.

After around 150 miles in a variety of conditions and load, the problem did not happen. Now I’m not naive enough to believe that it’s fixed (too much false hope and shattered expectations in the past), but at least I don’t know that it’s NOT fixed!

Here’s where I stand on the major theories:

Tank suck/vacuum: I’m not a believer any more for two main reasons: 1) opening the cap never gives any hint of a vacuum, and 2) I can easily blow air into the breather vent under the cap and the air cleanly comes out the end of the hose.

Side stand switch: My GRiSO has an extremely bright very large yellow light on the dash that comes on the instant the switch is triggered. I would have definitely noticed it because I made a point of looking at the instruments the last few times it happened. So I don’t think this is a possibility.

Kill switch: I did test hitting he kill switch while riding, and the behavior was somewhat consistent with some instances of the problem. Nothing lit up, no codes, just cut off. However it does get hard to explain some cases where the engine ran poorly and I could keep it alive by giving more throttle, but I guess an intermittent connection dependent on vibration is a possibility. So I haven’t closed the door on this one but I’m not strong in it either.

Ground wire poor connection: if it’s possible to increase resistance enough to kill the injection system and ECU but not enough to reset the clock and trigger a code, then this is a distinct possibility. I like this explanation a lot since I fastened the ground and the problem didn’t happen this time. But that’s not very scientific.

Coils: not a good explanation because only one cylinder would fail if a coil was bad. But going back to my unscientific wishful thinking, I changed both coils and the problem didn’t happen afterwards

So not a definite resolution yet but I thought I’d update with the latest. Will ride again tomorrow and see if it happens. Each ride where it doesn’t happen gives more confidence that it’s fixed, but right now I am only 5% believing.

I think the coils should throw an error when something is wrong with them that is then stored in the ECU. Did you find some cheaper alternatives than the OEM ones that are like 140€ a piece?

Hopefully it's fixed for you now. If not, could it perhaps be a short somewhere that only plays up at a certain vibration/handlebar angle/high moisture level/alignment of stars?

GuZeee wrote:

....
Tank suck/vacuum: I’m not a believer any more for two main reasons: 1) opening the cap never gives any hint of a vacuum, and 2) I can easily blow air into the breather vent under the cap and the air cleanly comes out the end of the hose.

I think the hole under the cap is drainage for water? For rain / washing? I believe the tank "breathes" through a small brass fitting, number "12" in the picture. This fitting has a holle thats about 1mm and is prone to cloging. If you do not have this "gas recovery system" on your bike, the hoses "6" are just open to the atmosfere. Try blowing air through this, with open tank cap should be easy.
I don't know the direction of the valve "7", I wonder what is expected to happen if it gets stuck closed... (My 1100 is not that friendly to the enviroment, it doesn't have this system, both "12" fittings are vented to the atmosfere.)

[You must be registered and logged in to see this image.]

GuZeee wrote:

Side stand switch
........
Kill switch
.........
However it does get hard to explain some cases where the engine ran poorly and I could keep it alive by giving more throttle, but I guess an intermittent connection dependent on vibration is a possibility. So I haven’t closed the door on this one but I’m not strong in it either.

This door is closed... Side stand / kill switch result to instant engine off. Intermittent connection makes no difference, instant off.
Same aplies to inclination sensor (dropped bike), instant off, maybe this also needs to remove the key for some minutes before been able to restart - it does to my bike (not surre if it is affected by the antitheft alarm thats installed)

GuZeee wrote:

....
Ground wire poor connection: if it’s possible to increase resistance enough to kill the injection system and ECU but not enough to reset the clock and trigger a code, .....

The injection system and the ECU will probably be the last to die, as they do not need much amperage (especialy the ECU, needs minimal). Also at any time, there can be only one injector or coil called to operate.
On the other hand, the component that is expected to "die" first is the the one that needs the highest amps in the circuit, and that is the fuel pump, it needs more than 3A to build fuel pressure.

GuZeee wrote:

....
Coils: not a good explanation because .......

Mostly because the ECU would have register an alarm. "ECU error 22" or 23 or 24 or 25 for shorted left / oppen left / shorted right / open right. Same applies to injectors, errors 26 to 29.

motor-timothy wrote:

Did you find some cheaper alternatives than the OEM ones that are like 140€ a piece?

Mine has engraved the maker and code. Its "Champion BAE800B/245" I found it at about half price than what it has as a Guzzi part. Using the champion code, also found some alternatives to it, like NGK 48013... My 1100 has twin sparks, so yours is diffent, seems to be...

... BAE850? Check for engraving.
Champion cross refence this to Guzzi's part number.
[You must be registered and logged in to see this link.]

You can take it from there, look for the part manufacturer code, and its alternatives... Looks like Marelli 30716500?

motor-timothy wrote:

.....could it perhaps be a short somewhere that only plays up at a certain vibration/handlebar angle/high moisture level/alignment of stars?

Hopefully not, worst case ball busting senario....

Thanks for the info, avgpetro. So, I'm pretty sure that hole next to the fuel inlet under the cap is what connects to the breather fitting #12 in the diagram. I removed the charcoal canister and the valve on mine so that this fitting vents directly through a hose to atmosphere, ending at the bottom with other breather lines. When I blow compressed air through that hole under the cap, that's where the air comes out of. There is also another one (hose 6 in the diagram) that vents directly to atmosphere but I'm not sure where that goes inside the tank. However, since the hole I tried has clear external air passage and it does access the air where the fuel is, it seems impossible for there to be a vacuum in there.

Quote :

I think the coils should throw an error when something is wrong with them that is then stored in the ECU. Did you find some cheaper alternatives than the OEM ones that are like 140€ a piece?

I found a used set of coils off a crashed California for $45, so I snapped them up.

If the problem is still there the remaining theories are:

Quote :

could it perhaps be a short somewhere that only plays up at a certain vibration/handlebar angle/high moisture level/alignment of stars?

Scary! There were suggestions of connecting various key points to a lamp to check if they are failing, but this is a very difficult process to debug and repair.

Quote :

the component that is expected to "die" first is the the one that needs the highest amps in the circuit, and that is the fuel pump, it needs more than 3A to build fuel pressure.

Expensive! I'll replace the pump if I'm sure that's the problem, but it's $795!

There was also an idea earlier in the thread that it might be an internal disconnection in the battery. That is interesting, I wonder if a battery disconnection would cause this, even though there is still power from the alternator? I guess the battery smooths out the power and the "dirty" signal from alternator could annoy the ECU. But no codes?

Going to ride it again and see what happens.

I don’t think he means that you need to replace the fuelpump, just that when the connection to earth is bad that would be most prone to stop functioning, and not the ECU or injectors since they require less amps.

Ah I see, good point! That lends more credibility to the grounding problem, and the fact that it’s possible it is indeed fixed. Short ride today and all is still well.

It’s so nice to have all this help and ideas from all of you to solve this, I really appreciate the thoughtful input.

Quote :

Thanks for the info, avgpetro. So, I'm pretty sure that hole next to the fuel inlet under the cap is what connects to the breather fitting #12 in the diagram. I removed the charcoal canister and the valve on mine so that this fitting vents directly through a hose to atmosphere, ending at the bottom with other breather lines. When I blow compressed air through that hole under the cap, that's where the air comes out of. There is also another one (hose 6 in the diagram) that vents directly to atmosphere but I'm not sure where that goes inside the tank. However, since the hole I tried has clear external air passage and it does access the air where the fuel is, it seems impossible for there to be a vacuum in there.

Oops, I was totally wrong on this, I had it backwards. It is of course just as you said avgpetro, the big hole just under the gas cap does not vent to the inside of the tank, and the right side breather (#12 in the diagram you posted above) is the one. It used to go through a valve and charcoal canister on my bike, but I now have it directly venting to atmosphere at the bottom of the bike. I tried blowing air through it and it does pass through into the tank, so it doesn't seem blocked (at least not now anyway; I did have to unblock it at one point).

So, the latest on the problem: I rode through most of another tank-full without incident, with lots of excursions to redline and plenty of WOT. It ran great. BUT! (Always a but) at the same time as the low fuel light came on there was a minor stumble. Nothing like the severe power off/stall that I had experienced before, just a momentary noticeable decrease in power, then back to normal. As soon as it happened, the low fuel light came on.

This brings to mind some questions: is the low fuel light indicating "switching to reserve" where something actually changes in the fuel routing, or is it just based on a fuel level sensor? If it's just a level sensor, how could it affect the fuel delivery or engine power? Was it just a coincidence? So many questions...