MV Agusta Forum banner

1 - 20 of 34 Posts

·
Registered
Joined
·
233 Posts
Discussion Starter #1
i mentioned in another thread that i have spent about 50 hours reverse-engineering my 910S electrical system in an effort to diagnose why my engine skips/misfires occasionally (it used to stall before i put in iridium plugs). a couple of other members of this forum mentioned it happening to them when they pull the clutch in, which gave me an idea to investigate the sub-circuit that controls the "OK-to-crank" condition along with the "kill-engine-when-not-neutral-and-sidestand-down" condition. these two conditions are part of one (pretty simple) sub-circuit.

one of the key lines in this circuit is pin #7 of the CPU. the CPU holds this pin at 12V through a pull-up resistor. it is the responsibility of the neutral-switch, relay-box, and sidestand-switch to GROUND this pin, otherwise the CPU will forcefully kill the engine.

so i hooked up my scope to CPU pin 7 and got the waveform below. EEE-GADS there are random giant noise spikes that recur over and over! the larger noise spikes seem to happen every few tenths of a second. i mean WTF i could not believe what i was looking at-- my jaw must have literally dropped in amazement. note that the noise lasts only about 2 microseconds, but this kind of noise floating around in a circuit can WREAK HAVOC on things. it is indicative of bad design-- there is no good reason why this particular circuit should have this kind of noise present-- it absolutely should be filtered with a bypass capacitor (inside the CPU). as is, it is very possible that the CPU thinks (for 1 microsecond at least) that it should kill the engine because pin 7 is not at ground, instead is bouncing all over the place with positive voltage up to 6 volts!

more testing is required to determine if this is really what is causing my engine to misfire, and other owner's engines to stall. it's simply impossible to know anything for certain at this point, but this trail looks pretty promising.

 

·
Registered
Joined
·
41 Posts
Kinda hard to be certain, isn't it?

ridesideways said:
... it is the responsibility of the neutral-switch, relay-box, and sidestand-switch to GROUND this pin, otherwise the CPU will forcefully kill the engine.
I'm impressed with your attempts at analysis, but wonder if there's not more going on than you've been able to see with your o-scope. Is an examination of pin-out waveforms enough to really have an idea of what's going on inside the ECU?
 

·
Registered
Joined
·
779 Posts
This reminded me of the time my Gixxer kept cutting out when ever I hit a bump on the highway. It turned out to be weak kickstand spring that allowed the kickstand to dip down and trip the kickstand switch….
 

·
Registered
Joined
·
1,332 Posts
Very impressive analysis!

The noises you are talking about must be happening on all MV ECUs then. Your CPU must be one of those that have more noise than others which don't stall or misfire...

Once you've finished your analysis, it'd be interesting to push it over to the R&D guys at MV Agusta. I have just the right contact ;-)
I don't know if it's too late, but if it turns out to be a manufacturing defect, MV will definitely be on MagnetiMarelli's ass to get this fixed.

One thought though, is there anyway you can fool pin#7 by placing an emulator or something that'd constantly tell the ECU that its pin#7 is OK and gets whatever ground signal it needs to get ????
As for the lambda sensor for instance, once can suppress it by placing a lambda emulator that tells the ECU it's getting A/F values (that are actually bogus because not really coming from actual A/F ananlysis).
 

·
Registered
Joined
·
233 Posts
Discussion Starter #7
asg21 said:
I'm impressed with your attempts at analysis, but wonder if there's not more going on than you've been able to see with your o-scope. Is an examination of pin-out waveforms enough to really have an idea of what's going on inside the ECU?
you are correct, i am certain of nothing. but "certainty" is for lazy people, not engineers. if i needed to be certain of the outcome of every first step i've taken, i'd still be living with mommy in a crib.

when my bike stops misfiring because i've bypassed the noise (or corrected whatever other electrical gremlin that's causing the problem), then you can decide for yourself if you want to keep owning a bike that stalls and misfires, or if you want to make the same change to your bike.

caveat emptor, it is always possible that the problem may turn out to be something that is impossible to correct without major changes to the CPU. i'll risk my time to find out though.
 

·
Registered
Joined
·
756 Posts
gio8856 said:
+1. Where is JamesC on this important technical matter :stickpoke :)

However, sounds like a credible explanation. My uninformed view is that the skip could easily be an electrical issue. Thx for all the effort and report out Ridesideways :)
 

·
Registered
Joined
·
803 Posts
Very interesting sideways, keep up the investigation.
 

·
Registered
Joined
·
41 Posts
Certain?

ridesideways said:
you are correct, i am certain of nothing. but "certainty" is for lazy people, not engineers. if i needed to be certain of the outcome of every first step i've taken, i'd still be living with mommy in a crib.

when my bike stops misfiring because i've bypassed the noise (or corrected whatever other electrical gremlin that's causing the problem), then you can decide for yourself if you want to keep owning a bike that stalls and misfires, or if you want to make the same change to your bike.

caveat emptor, it is always possible that the problem may turn out to be something that is impossible to correct without major changes to the CPU. i'll risk my time to find out though.
I've re-read my email, but can't find the word "certain" - maybe you confused me with someone else. I was just wondering if a problem that's internal to a chip can be understood by looking at output waveforms, that's all. Good luck, though - I hope you're successful. I just think problems with modern engine management systems aren't really amenable to your approach, but I'll be glad to hear your report when you prove me wrong. I'm just glad your problem's not universal - my 910 doesn't exhibit the behavior you describe.

Btw - I'm an electrical engineer, and I'm certain sometimes.
 

·
Registered
Joined
·
233 Posts
Discussion Starter #11
asg21 said:
I've re-read my email, but can't find the word "certain" - maybe you confused me with someone else.
lol check your title

yes it takes more than putting a scope on pins to understand the function of every pin-- you also need a basic schematic, which is available in chassis service manual. the schematic, although it doesn't show every component, clearly shows the function of CPU pin 7. in fact, most of the pins of every component are clarified partially by the schematic, and partially by reverse-engineering testing with a scope.

i'm certain sometimes too. all i'm saying is that it's just not a necessity to progress.

lol i feel like i'm in a pissing contest for some reason. oh well.
 

·
Registered
Joined
·
41 Posts
Oops

ridesideways said:
lol check your title

yes it takes more than putting a scope on pins to understand the function of every pin-- you also need a basic schematic, which is available in chassis service manual. the schematic, although it doesn't show every component, clearly shows the function of CPU pin 7. in fact, most of the pins of every component are clarified partially by the schematic, and partially by reverse-engineering testing with a scope.

i'm certain sometimes too. all i'm saying is that it's just not a necessity to progress.

lol i feel like i'm in a pissing contest for some reason. oh well.
My apologies - I hadn't looked at my original title, and I'm sure at the time I thought it was a cool phrase. I really hope you are successful, especially if the problem you're trying to fix is so bad on your bike - luckily, it doesn't seem to affect mine. I'm on my second Brutale, and I've yet to have any stalling problems. I still think your approach will be too simple to work, as (IMHO) you're not going to be able to accurately determine what's going on in the chip(s), and a partial schematic won't be of any real use. I hope I'm wrong, though - I'm sure it's a drag to experience the problems you've described. I hope you post your results, whether you're successful or not.
 

·
Registered
Joined
·
917 Posts
+1, another interested party. Good luck. (and by the way- it's great to see that it is possible to diasagree and still be cordial. )
 

·
Registered
Joined
·
756 Posts
+1...unfortunate when decorum breaks down. Too much of that. Keep up the good work Ridesideways :). Without any technical skills, but having similar problems with one vintage bike, I'm convinced that skip is electrical in nature and not a fueling issue
 

·
Registered
Joined
·
364 Posts
ridesideways said:
i mentioned in another thread that i have spent about 50 hours reverse-engineering my 910S electrical system in an effort to diagnose why my engine skips/misfires occasionally (it used to stall before i put in iridium plugs). a couple of other members of this forum mentioned it happening to them when they pull the clutch in, which gave me an idea to investigate the sub-circuit that controls the "OK-to-crank" condition along with the "kill-engine-when-not-neutral-and-sidestand-down" condition. these two conditions are part of one (pretty simple) sub-circuit.

one of the key lines in this circuit is pin #7 of the CPU. the CPU holds this pin at 12V through a pull-up resistor. it is the responsibility of the neutral-switch, relay-box, and sidestand-switch to GROUND this pin, otherwise the CPU will forcefully kill the engine.

so i hooked up my scope to CPU pin 7 and got the waveform below. EEE-GADS there are random giant noise spikes that recur over and over! the larger noise spikes seem to happen every few tenths of a second. i mean WTF i could not believe what i was looking at-- my jaw must have literally dropped in amazement. note that the noise lasts only about 2 microseconds, but this kind of noise floating around in a circuit can WREAK HAVOC on things. it is indicative of bad design-- there is no good reason why this particular circuit should have this kind of noise present-- it absolutely should be filtered with a bypass capacitor (inside the CPU). as is, it is very possible that the CPU thinks (for 1 microsecond at least) that it should kill the engine because pin 7 is not at ground, instead is bouncing all over the place with positive voltage up to 6 volts!

more testing is required to determine if this is really what is causing my engine to misfire, and other owner's engines to stall. it's simply impossible to know anything for certain at this point, but this trail looks pretty promising.

My initial thoughts are that as the interrupt line (pin 7) is essentially just tied to ground the spike that you are seeing may not be specifically a spike on the interrupt but could be a spike on the ground line.

It's probably worth scoping other pins simultaneuously in an attempt to determine if this spike corresponds with some other event (a plug coil discharge, injector trigger etc..). If the spike occurs at a fixed frequency then it is probably related to some event. If the frequency is somewhat random then it could be some sort of 'chatter' from a switch, plug or relay contact.

I think you'd also need to determine what the ECU registers as a valid interrupt signal. While 6 volts is probably enough, I'd be surprised if a duratiojn of less than 2 microseconds is. I would expect, given the inherently noisy nature of vehicle electrical systems that the interrupt line would be sampled rather than taken as conditional and it may take two or more consecutive samples of an interrupt state before it's actually registered as valid.

While an interrupt of a single spark may upset the fuelling to induce an audible miss, it won't cause the engine to stall. Think about how the average aftermarket quick shifter works, their interrupt times are typically in the region of 200-400 milliseconds. While I appreciate that quickshifters are normally actuated higher up the rev range, many can still be actuated in the lower rev range with no ill effects.

One thing that doesn't make any sense to me is that if an ECU interrupt was the root cause of the stalling, then why would fitting iridium tipped plugs seemingly fix/mask the problem as many owners suggest? If you have no spark, you have no spark regardless of plug type.
 

·
Registered
Joined
·
193 Posts
Hartley Hare said:
One thing that doesn't make any sense to me is that if an ECU interrupt was the root cause of the stalling, then why would fitting iridium tipped plugs seemingly fix/mask the problem as many owners suggest? If you have no spark, you have no spark regardless of plug type.
Sounds interesting. Are you doing all this work your self? I should have stayed in school longer.

Anyway, about stalling. My 07 910R used to stall on me about 3 times in 50 miles on my way to work, usually only at traffic lights. For me it's all highway except for the last 7 miles when it's city driving.

I removed my cat and silencers and left everything else original, plugs, airfilter, eprom. Then my bike would stall about 5 times and pop when going below 20mph, but at highway speeds it ran strong.

Now I have changed to just the MV Corse mid pipe, original silencers and eprom. I rode the bike yesterday about 60 miles and it stalled once, only after letting it idle for a long time, maybe 5 minutes. I don't think plugs are the answer to the stalling. Iridium tipped plugs, like the Bosch Platinum plugs, help with weaker ignition systems but I doubt enough to make a huge difference. My thoughts only, no one needs to agree with me.

I have been following your posts, please keep going when you have the time.This interests me, as I hate it when my bike stalls, I don't want to get killed by the vehicle behind me.

I think the problem is in the way the exhaust is analized by ecu. I think it measures readings beyond it's parameters for whatever reason and it shuts the bike down. But what do I know, I wouldn't even know where to buy the equipment your using, let alone know how to use it.

Keep up the good work.
 

·
Registered
Joined
·
364 Posts
I'm not doing any work/testing, it's 'ridesideways' that's researching the problem. I'm just throwing some of my initial thoughts into the mix.
 

·
Registered
Joined
·
5,193 Posts
Did you read this post about the cure for 910's stalling?

http://www.mvagusta.net/forum/showthread.php?t=12322

Seems quite valid to me, I have messed with MV's for 7 years or so now, 3 of my mates also have F4's and Brutale's, I used to try just about every chip available for the F4 in the search for some hopefully gained hp that I never really found with the 750.
One chip that was supplied for the F4-750 with SPR headers I just could not get to work properly, by that I mean however I set the bike up with the TPS and co it would just stall 1 time in 10 you stopped every ride.
I messed with this chip for a week or so and just gave up eventually and fitted another chip, I never had this stalling with any other chip I tried, I came to the conclusion this chip had some kind of bug in it that would just cause it to stop computing, it probably did it all the time the bike was running but was only noticeable at idle.
 

·
Registered
Joined
·
193 Posts
Hartley Hare said:
I'm not doing any work/testing, it's 'ridesideways' that's researching the problem. I'm just throwing some of my initial thoughts into the mix.
Well my hats off to both of you. My realm of knowledge in such matters is
a big fat "Null" so I read along and learn from you guys. Now I'll go read the thread 'mikef4uk' suggested.
 
1 - 20 of 34 Posts
Top