09-08-2009, 08:49 PM #1
Gunna.... I looked at your log file
I took the time to review the log file you posted shortly after your engine died on you. I found some interesting information elements you may want to consider. They point to different reasons than bad gas for explaining why your engine self-destructed.
The log file shows you held WOT for a continuous 47.39 seconds. During that time, the ignition timing was steady at 20* degrees BTDC. Then, shortly before piston #2 detonated and melted, ignition timing increased to 22.5* BTDC. It is that 2.5* advance delta that killed the engine after nearly 50 seconds of continuous WOT pull.
Why did timing go from a steady 20* during 47.39 seconds to 22.5* in the last 2 seconds? The answers are all in the log file.....
The steady 20* degrees BTDC during the long WOT pull was possible because total timing was retarded 2.6* degrees by a 'Full Throttle Timer' timing compensation function. That function appeared to have been programmed to start derating power after 10 seconds of full WOT pull. The log file shows gradual timing retard compensation after 10 seconds of continuous WOT pull. So far, so good and the engine kept steady power above 8,400rpm for over 47.39 seconds.
Then something really unfortunate happened.... Throttle was briefly let off below 80% TPD. That caused the 'Full Throttle Timer' to reset to zero second....... which also had the effect to reset the ignition timing retard compensation to zero.......
That brief WOT interruption was barely a fraction of a second, then full WOT power was applied again. At that point, everything was already heat soaked. Adding +2.5* advance fried the engine within 2 seconds. Once the self-destruction process started, at time = 5:54'' in the log file, the rpm started to fade which allowed timing advance to further increase.... up to 25* degrees at over 190kpa.... that accelerated the melting of piston 2. Once the process started, it looks like it took around 0.2 seconds before full meltdown... never a chance to react!
I'm sharing what I saw because I believe it is important to properly setup timer based compensation functions when used to derate and protect engines from detonation. In fact, timer reset should ideally be programmed to take effect no less than 5 continuous seconds of 'Throttle Position (TP)' being lower than 50% value. This allows everything to cool down and be safe once again for full ignition timing advance @ WOT.
My $0.02.... hoping this provides a different perspective on what happened.
09-09-2009, 03:30 PM #2
- Join Date
- Oct 2007
- Brisbane, Australia
You don't know how much I appreciate your input to this Desperado, thank you very much for taking the time to analyse the .ld file.
At the exact point of "let off" I was hitting my first turn of a wide river run and backed off just a fraction (chicken shit!). Then got back into it. This was the let off you talk about. It was like I did the whole thing yesterday.
I still have my doubts of the level of fuel quality and believe that if it had new 98 fuel in the tank the margin of error would have been wider. However this detonation would never have occurred and you may never have found a possible issue.
It has made me adjust my ground rules of always having only fresh premium 98 octane fuel (Australian Rating) in the ski at all times. When it gets put to bed after any outing the fuel tank will be drained bone dry and only refueled at a BP station. No more independent stations for this ski ever again. It'll be just another small part of maintaining this type of ski and it's motor.
It has given (forced!) me the chance to learn about and upgrade the internals of the motor which is still a work in progress. Once the motor is finished and put back in the ski with the Fatty intercooler and other mods I'll be intensely studying your (and others) work on the motec system and adjustments.
My biggest concern is to protecting my motor investment after it's rebuild. I wish to find the best practice for gathering feedback information e.i. sensors, sensors and more sensors.
Then I need to figure out what all this information means and adjust the motec to it's maximum performance. Then pull it back 5% for protection. Not an easy task for one as thick as a brick shythouse.
But I must give a huge credit to motec for what they have developed. The motec is very complex and they have setup a system of plug and play for dummies. All this over a long period of dedicated dyno time and racing their own ski's. And then bringing the product to a sceptical market and supporting it with forums such as this. If they make a few dollars from me, then they well and truely deserve it.
I'll analyise the portion of the file your talking about but I'll probably have lots of dumb questions on how to get into the specific area's of the program. Hope you don't mind?
ps sorry for my sucky gramma.
09-09-2009, 08:31 PM #3
I am glad my observations were useful to you.
I'll take this a step further. A simple spark plug color check reveals #2 and #3 cylinders are running leaner than cylinders #1 and #4. The difference is substantial to the point where the lean cylinders (#2 & #3) would be running on the edge if cylinder #1 & #4 were tuned for maximum power. MoTec included individual cylinder fuel enrichement for #2 and #3 in their early tune files -- that was brilliant from them. Later tune files don't have that feature -- I believe the upcoming E1 base map will require individual cylinder trimming.... I guess we'll se what Pete will do.
Spark plug color checks on my setup confirmed #2 and #3 cylinders were running leaner. I had to trim these individual cylinders with more fuel and slightly less timing. My plugs are now the same color for all 4 cylinders.
Next step for me will be to identify plugs with a 1 to 2 step colder heat range.... this is typical when tuning for more boost and I'm surprised this simple mod has not been discussed yet.
09-10-2009, 06:58 AM #4
- Join Date
- Oct 2007
- Brisbane, Australia
Thanks for the heads up on the spark plug colouring.
I hope I am making the right assumption but this would be due to the airflow thru the inlet manifold?????
It does make sense in the fact that #1 and #4 are on the outer of the inlet manifold therefore receiving less air flow thru the tube type inlet manifold. Fuel/Air ratio making the plug run richer???
In fact there was another piston in the "Big Bang" motor that had score and heat damage to the rings and their landings. I bet it was #3 and was about ready to let go as well. It's a shame that I didn't identify which piston was which as I stripped the motor down. I will know better next time.
I find it is an interesting manifold design that Yamaha has choosen for this motor. It may have been designed out of economy rather than efficency. However it "does the job.
It would be very enlightening if someone could send one of these manifolds to the guy on the forum who does a lot airflow measurements. It may only prove what has already been seen by yourself, that the 2 centre ports get more airflow/pressure than the outer 2.
I beleive Dan JD1 is doing (or about to) modify an inlet manifold. This maybe for this reason (tune in if you will Dan??).
Choosing the 2 sizes down heat range plugs is now also on the "to do list".
Question; At what boost levels would this need to come into effect or is there other characteristics that need to be considered as well?
However all this could be due to the cooling design of the motor as Motec Pete stated over here: http://www.greenhulk.net/forums/show...=94054&page=34 on post #333.
Either way it can be compensated for by the individual tuning of each injector to suit the motor as you suggested. Not that I have a clue how to do it.
As always, many thanks to you Desperado.
09-10-2009, 03:25 PM #5
Great work on your part Desperado.
The guys are very lucky to have someone of your caliber helping them with their stuff.
I actually did individual lambda on that engine whilst on the dyno, I think I mentioned this previously. We looked at the intake when we first saw it and thought mmmmmmmmmmm yuk.
We were surprised to find that at the higher loads and rpms the afr was remarkably even.
Now this of course was on our engine with a C3 at the time.
That is not to say its even for everyone and thats why you have the ability to change it if you have the capability of measurement, be that by plug colour, EGT or lambda probe.
09-10-2009, 03:28 PM #6
this will be very, very important to endurance racers, everywhere that use this system.
09-10-2009, 08:25 PM #7
The spark plug options are as follows:
OEM spark plugs: NGK LFR6A
The design specifics are 14mm thread, 1.04" reach, 5/8" hex socket size, projected tip and copper core.
Upgraded plugs with colder heat range, designated by the number in the plug part number (higher number = colder heat range for NGK), are:
Normal Iridium at around $12 per plug:
- LFR7AIX with 0.032 electrode gap (1 heat range colder)
- ILFR7H with 0.024" electrode gap (1 heat range colder)
- R7437-8 with 0.027" electrode gap and projected ground strap (2 heat range colder)
- R7437-9, same as above except 3 heat range colder
- R7438-8 with 0.027" electrode gap and recessed ground strap (2 heat range colder)
- R7438-9, same as above except 3 heat range colder.
In my view, the R7437-8 would be a great choice for weekend warriors running hight boost. For all out endurance racing, the R7438-9 would be a good option. I wish these could be available through the online store...... Jerry...?
09-10-2009, 09:29 PM #8
Their prices are tough to beat.
09-13-2009, 06:55 AM #9
- Join Date
- Oct 2007
- Brisbane, Australia
4 x Sensors on Individual Exhuast Ports: EGT or Lamda
Thanks for the spark plug info Desperado and Jerry. Looks like the R7438-9's are the way to go.
I know this is a bit off topic for Motec support thread stuff but this maybe done with Motec as well.
OK.......first question/project I have is:
I wish to measure (in real time) the correct fuel air ratio on the individual cylinders on the motor. I note there are 4 x blanks on top of the exhuast manifold which seem to be a place to fit 4 x seprerate sensor bungs. Could these be drilled and taped for such a project?
If so which is the best way to measure the correct A/F? Lamda or Exhaust Gas Temperature. A question my engine builder asked was what type of exhaust temperate sensor or knock sensor did I have on the motor? My answer was "Nothing".
I would simply mount the 4 x meters in the glove box under the handlebars.
I could do it with the Motec or spark plug colour but the worry for me is what if something was to go wrong before I got back to base. It could end up being another "bad day" on the test track. I would like to see what's happening in real time.
Or am I barking up the wrong tree for this type of sensoring measurement?
And yes, I am over-protective.
Any input would be welcomed.
09-13-2009, 04:23 PM #10
The bungs up there on the exhaust go to the water jacket, so too hard to access the exhaust from there.
Cylinder fuelling mismatch tends to be a permanent issue that is usually related to airflow movement in the intake, well lots of things really but mostly that is my experience.
Unless an injector blocks then what you get on your first test ride will be what you get for life.
So the theory of keeping something on board the whole time to monitor is frankly, overly cautious. I donít know of one racing team that does it, with lambda anyway, EGT yes because it is simple and cheaper but its still an overkill for your application I would sugest.
My advice would be have permanent lambda on board and if you are worried you can run races in closed loop.
For the ultimate protection you can get the closed loop knock set up and forget about it.
You can have terrible mixture distribution and not cause any harm to the engine, one batch of fuel not suited to the tune and a perfectly even cylinder mixture engine will go BANG every time.
The whole racing game is about running the max ignition timing possible for the fuel allowed.
Some engines when tuned to peak power are miles away from detonation, so perfection on general tune and across cylinders is nowhere near as important.
But if you are only allowed to run say a 91 octane fuel, but are allowed the best supercharger wheel at 16psi then to get max power you will have your ignition timing very close to detonation the whole time.
In cases like this you need to make sure everything is perfect, no lean engine or cylinders.
So what I am trying to say is donít spend a whole load of money and time worrying about the wrong things.
Now all that said, I have a plate that we got laser cut up to fit between the exhaust manifold and the cylinder head, it has small sample holes that duct exhaust gas away to separate fittings external on the engine.
We then plumbed those fittings to tubes with lambda sensors in them and measured the 4 engine AFRís as well as the collector AFR.
We used our new LTC devices. (Lambda to CAN)
These are like miniature lambda meters that talk to other devices on a CAN bus. (like a network)
You can hook up as many as you want together and send all that info to the M400.
The whole lot gets logged at once and you can tune your engine to perfection.
If you want to go through the exercise I will loan you the plate and all the electronics required to get the job done.
You will need to make the 4 tubes and plumb them to the plate and MoTeC will lend you the rest.
Users Browsing this Thread
There are currently 1 users browsing this thread. (0 members and 1 guests)
By Hydrotoys in forum Yamaha How To & FAQsReplies: 3Last Post: 07-30-2012, 04:46 PM
By boomanch1967 in forum MSX/MatrixReplies: 0Last Post: 08-10-2011, 08:43 PM
By GP1800 in forum Yamaha ECU SupportReplies: 9Last Post: 07-22-2011, 10:26 AM
By Teerayoot in forum Yamaha PWC Performance (4-stroke)Replies: 0Last Post: 08-28-2010, 08:23 AM
By SHOBiz in forum Yamaha ECU SupportReplies: 3Last Post: 08-07-2010, 11:03 AM