450 mysterious rev limiter

[MikeDutsa]

yo,

The Problem...
Our engine starts really easily, revs up pretty well, and maintains an AFR of about 12:1, untill it gets to 5800RPM. It seems to hit a spark cut rev limiter.

A little info on our set up....
We are using a 450 KTM engine. it is using the ignition system out of an '07 KTM 450 SX-F, and a custom ECU.

What we have tried....
New Injector
Removed ECU's ability to cut ignition
Replaced spark plug cap and wire
Changed fuel map and accelerator settings, a lot
Reprogrammed ECU

The deal is that it seems to be an ignition problem, and from what we have researched, the CDI either works or it doesn't, no in between. The problem arose suddenly. a few days ago it worked awesome, yesterday it was fuel issues, and today, the spark seems to cut out. We recently replaced a broken woodruff key on the crank shaft, which was preventing it from running at all.

If any single cylinder teams have seen this or have any ideas as to what it might be that would be awesome.

Thanks,

Mike D

[MikeDutsa]

yo,

The Problem...
Our engine starts really easily, revs up pretty well, and maintains an AFR of about 12:1, untill it gets to 5800RPM. It seems to hit a spark cut rev limiter.

A little info on our set up....
We are using a 450 KTM engine. it is using the ignition system out of an '07 KTM 450 SX-F, and a custom ECU.

What we have tried....
New Injector
Removed ECU's ability to cut ignition
Replaced spark plug cap and wire
Changed fuel map and accelerator settings, a lot
Reprogrammed ECU

The deal is that it seems to be an ignition problem, and from what we have researched, the CDI either works or it doesn't, no in between. The problem arose suddenly. a few days ago it worked awesome, yesterday it was fuel issues, and today, the spark seems to cut out. We recently replaced a broken woodruff key on the crank shaft, which was preventing it from running at all.

If any single cylinder teams have seen this or have any ideas as to what it might be that would be awesome.

Thanks,

Mike D

[TMichaels]

Hi Mike,
hook up an oscilloscope to the ignition coils and fuel valves and check that ignition is really cut. Also monitor the system voltage while doing so. Sounds more like gremlin in your electrical system if you made sure that it is not a problem of the setup of the ECU.

Have a look at this thread for some detailed testing description, since you problem sounds pretty much the same:
h t t p ://fsae.com/eve/forums/a/tpc/f/125607348/m/327109264

Regards,

Tobias

[mech5107]

Check your crank/cam sensors.

And, as someone said, check the ground, It's always the ground...

[JeffreyH]

Check the datasheet for your cam/crank sensors also... we had something similar happen with crank sensor being used beyond its spec. From memory the RS catalogue it was bought from said it was rated to 15kHz, but it was a 4kHz sensor when I looked up the actual datasheet. I now never trust any details from RS/Farnell/Mouser/etc without having checked the manufacturer's datasheet.

[MikeDutsa]

yeah, i've looked into all these things, the bitch is that it was working a few days ago, it reved up fine. Something has changed.... we just don't know what it is to change it back. I think we are going to have to dig through our ECU code and see if there is something there...

[Tom W]

Probably worth checking that your ignition wiring is not getting close to your crank/cam sensor wiring. We had a problem in testing for '08 where a CDI cable ran across the cable for the crank sensor. The car ran faultlessly for the first days testing and the first driver on the second day then after purely a driver change (~5 min) the ECU would cut out at a particular (from memory low) rpm.

In that particular case one of the cables from the CDI to a spark plug had dropped onto the connector in the wiring from the crank sensor to the ECU, the interference only screwed us over at a particular rpm where the ECU (MoTeC M4 in those days) would suddenly get a non-sensical signal and cut everything to protect the engine.

[TorqueWrench]

Second what Tom said. We had an instant where interference from other sources was causing us to lose crank signal above a certain point. Make sure the crank sensor wires are fully shielded and away from most high voltage stuff (coils, starter, etc.).

[MikeDutsa]

We took off the entire intake system and threw a carb on there from a honda 450, it started like a champ, and did the same crap, reved nice to 5k then popped like hell. so we took a coil off a working '92 250cc 2-stroke Honda, and tried again with the carb, same crap. so now we are going to pull the engine apart and check the piston rings to see if they are stuck..... unless anyone has any good ideas. Oh-Yeah, we checked every thing with an oscilloscope and it all checked out.

[dazz]

You've tried a new sparkplug right? Sounds kinda like the timing is retarded to the shithouse. Can the pickup sensor slip, or be bolted on upside down, or something silly like that? Just make sure you've checked over the basics.

Good luck finding the problem and make sure you let us know what the problem was!

[MikeDutsa]

So, we have still yet to find the problem, but what we did find was a brand new KTM 560 race engine.... we'll have to see how that goes. It came with a new CDI and a new stock wiring harness. Will post results later....

[Steve O]

May be a little late to make suggestions now, but I didn't see anything about checking battery/charging voltage.

A couple years ago, working with the YFZ 450, I discovered that less than perfect charges on the battery would drop our REV limiter to 7500 RPM from I believe 10,500. Turns out the charging system on it at idle and low RPMs is perfectly capable of giving you the 13-14V you are looking for; however, if the YFZ ECU doesn't see that voltage over 13 from the battery at around 7500 is does a spark cut. This made it quite tricky to catch at first because we checked voltage at idle and it seemed fine. We noted that even with a charger on it, it did not work. From about 6800-8300 RPM the system drew a considerable amount of power to run(about 65 Watts). Trickle charger doesn't have enough current to supply the system at the higher RPMs and fast start chargers were too low voltage (like I said it appeared to have a relay at 13v and even a 10th of a volt below and it wouldn't rev up). Through a new battery on it and never had a problem until the charge got low. My problem was eventually completely mitigated by rewinding the charging system to output 130W from the stock 70W (and of course by no longer using the stock ECU).

Just to confirm that this issue was not a fluke with the one engine or CDI, this issue happened identically on our backup engine and CDI.

Good luck.

Steve

[PBnG]

What is your ignition coil dwell?

[MikeDutsa]

Thanks Steve, sounds like a plausible thing to have happened. The catch is that our stator has two separate coils on it, one that goes to the rectifier and supplies 120w, and one that provides power directly to the CDI. We now have it wired so that the only thing that the fuel and ignition systems share is the crank position signal. And the ignition system is exactly as the factory meant it to be. We got the new engine running today after transferring the electric start kit and the big stator kit from the old engine, however we were not able to rev it up because our CDI(different year and third CDI) no longer supplies power to the TPS. So now the ECU is not getting the TPS signal to regulate fuel, no big deal, just need our EE to set us up with a 5V wire from the ECU to power it. Another update tomorrow......


oh and PBnG, not sure what the coil dwel is, but it is the stock coil off the bike so it should work fine.... we tried swapping it for a coil off a 250 2 stroke and nothing changed.

If it has the same issue tomorrow I think I am going to cut the car in half. new engine, new ignition, better work.

[PBnG]

<BLOCKQUOTE class="ip-ubbcode-quote"><div class="ip-ubbcode-quote-title">quote:</div><div class="ip-ubbcode-quote-content">not sure what the coil dwel is, but it is the stock coil off the bike so it should work fine.... we tried swapping it for a coil off a 250 2 stroke and nothing changed. </div></BLOCKQUOTE>

I was speaking about the actual signal being sent to the coil, not any certain property of said coil. Overlapping ignition dwell signals cause an artificially low revlimit.

[MikeDutsa]

Got it. Turns out it was a pretty stupid mistake. easy over sight. When calculating the Crank speed in the ECU software, our EE got confused and though there were two rising edges per revolution instead of one, so that essentially, the problem the whole time, was that our tachometer was off by a factor of two. Good news is that the 560 runs like a bat out of hell. Thanks for all the input and ideas. Good luck to all the teams going to FSAE West... TWO WEEKS, see you there.

[L B0MB]

So you were revving it to 11,600rpm and though it was only 5800rpm? Haha did you wonder what it was going to sound like when you doubled the rpm? WAAAAAAAAAAAAA BOOOM!!! http://fsae.com/groupee_common/emoticons/icon_eek.gif

[MikeDutsa]

never messed with a single cylinder that had a tachometer on it. i'm used to the sound of a four cylinder at 13000. just takes some getting used to.... and at one point, a while back, we were taking it to 16000RPM. ooops, good thing we got a new engine. you live and you learn, won't have to make the same mistake next year.