Am working on a problematic 690 KTM 2008. Noted has all ECM equipment removed and replaced stator with older model and using a carb rather than FI.


Suspect a fueling problem with  lift pump (mikuni). Noted a capped bung on header and have been told, which makes sense, why not put any 02 in the hole and look and see if it is running rich or lean, perfect Stoic being .5. Hence if fueling is the issue before stall would we not see it go lean,?????

Also have read and studied burn times noting burn times and disturbances which indicate, lean, rich and a whole host of other issues. Notice on my CDI capture and not an anomaly, as have 3 bikes and the firing event looks the same in all, I.E; Strange when thinking of standard 1 to 2 MS burn times.

Any opinions as I note that the CDI on my KLR is fed by 120 volt AC non diodes straight to the CDI.

Larry

---

Hey Larry, that burn time is normal range for CDI. 80 to 150us is what I have seen depending on brand, usually the lower value on 2 strokes and the higher value on 4 strokes. If you are going to use an 02 sensor make sure it is a heated wideband. Keep in mind this can lead you astray as all you are checking is residual oxygen and there are other reasons you can have high 02 readings other than lean air/fuel mixture. Ie: ignition problems, cam timing etc. Better to use a exhaust analyzer that reads O2 and CO. The values you are looking for is O2 less than 1 to 2% and CO greater than 1 % but less than 7%. This would tell you you do not have and ignition problem and fueling altho maybe not ideal is in an acceptable range. Is there a certain rpm it always stalls at? like coming down to idle?

James, thanks bud for responding.

I sent you the file in your email but will post it here. We thought initially it was heat related but not exactly. He could ride it for maybe 15 minutes, then a rough stumble, always a little backfire then dead. Would not start again till cool. As of late it is a little more defined, I.E. will stall at about 220degrees, Okay fan not on, trying to create the issue, or then again it will die at 180, is this heat related were not sure?

The capture I will post is it running from cold start, run it up a little on throttle, maybe 3,000 then right at the end, gets rough, then speeds up a little, typical of a motor seizing or running lean just before it is out of fuel, then it dies. The troubling part if it is totally fuel, and we still have a hot spark, which we do defined by Pico and also physical out of the bike with plug tester it is there. That said little down the carb, starter fluid, or raw gas and it will not bark. Again symptomatic of a tired motor from my experience with chain saws and lawn mowers. Will start cold but once hot will not start again till cool. Is this what is happening here? not sure, but we do have spark and had it during the stall, neat thing about pico, we are not guessing, spark is there, one win.

With all the removal of the EFI we now know it is not ignition related so one down.

At this point the starter grenaded, so the testing ended for this time. I'm still suspecting fuel pump as they take the pulse for the mikuni pump from the throat of the carb. In my experience on my KLR which I converted to lift pump the pulse in the carb changes with temp, elevation and most of all when you open the blade under hard use and calling for more fuel you now have the reverse, low vacuum just when you need it most.


Also, right after stall looking down the stack, throat is dry and not squirt from the accelerator pump, indicative of fuel but would like to be more defined. My mikuni on the KLR can make as much as 5 PSI cold thus flooding the bowl as most gravity feed systems are only tolerant of 3psi. On the KLR made a bypass line to the tank, have a small on off valve which dumps anything over 1.5 psi in the line, simple yet works. We did not get that far as now need a starter.

On KLR, and also on chain saws and lawn mowers the pulse is taken from the crankcase which is always there and increased along with demand, in the throat of the carb is just the reverse.  We have not been able to test that yet as again starter fried. This is actually a really nice bike with 19,000 miles, has all the goodies, steering dampener, ect and he got it for 1800.00 because factory KTM mech, even the austrian consultant could not make it run. Was that a separate issue I don't know,might be as had 2 ECM's.

I'm new to compression wave form but have been working with Spence and using Fluke 350. The images are quite troubling. the crank with blade closed looks normal enough, crank with blade open show no intake valve open at all. Did it several times, just to make sure not removing the hose, and still the same capture.  I see no mechanical reason for this as looked at the compression release thinking it plays in the game, but is a pin that rotated during crank and don't think it changes the timing historically. Also you will notice a small hump in both cranks at the bottom of the compression rise and I surmise that is the compression release. Again if I show this to the KTM shops it is like a Goat looking at a watch, no comment and seemingly not interest, I get it.


Larry

---

StitchB wrote on Sep 28^th, 2020 at 5:57am:

The images are quite troubling. the crank with blade closed looks normal enough, crank with blade open show no intake valve open at all.

I would say its fine, generally cranking an engine with throttle open kills vacuum and raises compression, just as your screen shot shows. With the throttle closed you have created a restriction, hence the more pronounce vac. pocket and lower compression.

Can you post the compression file plz.  I'm interested in the compression release area. Im going to have to check my bike, I have compression release also.

Larry, Ditto on the no intake vacuum with throttle open, that is normal. have a look at this, its KTM690 for comparison. First screen with throttle closed and then last one with throttle open. I'll compare them when I get a chance.. James

---

Here is another with engine turning over a close to idle speed and the decompression system now no longer in the equation.

---

HarleyMike, and James thanks so much for the input, makes my observations crystal clear.

Mike, here are both files throttle closed and throttle open.

James, wondered how you mimicked crank speed on a single relative to idle? Is this tied to the back wheel on the dyno, or a specific starter, interesting.

That stated when I did my Yamaha I had PSI read out. Whether I forgot to click on the PV350 in my library not sure, but read out is in MV, and mine right at 603.. Don't like to assume but maybe extrapolate that to be 60PSI. Yamaha read out in PSI, and Spence thought it way to high. All on me no doubt but just wondering what I did wrong?

That said is there a math channel to convert MV over to PSI.   Again most of my usage is on diesels and this is my first usage of the PV350 on compression wave form, so still learning.



As long as we are discussing it, is there any trick way to set up a marker on fly wheel using optic or hall effect to simulate crank sensor. Is being discussed in another thread for looking at pre emmisions cars, just a thought?

---

StitchB wrote on Sep 29^th, 2020 at 5:27am:

That said is there a math channel to convert MV over to PSI.

I think there is a custom probe already made on the drop site, for your PV350

https://autonerdz.egnyte.com/app/index.do#storage/files/1/Shared/Custom%20Probes/Autonerdz


Yup, about half way down the page.

Here is the TW 200, noted no hump on the intake as bike probably does not have Compression release.

Thanks again for the feed back.

Larry

---

I built a remote starter for kick starter bikes and incorporated a VRS setup and 60 tooth wheel for better accuracy than the 1 trigger pulse per revolution. See pictures. You can put on the correct size socket to turn over the left side crankshaft bolt. Without it, it is near impossible to get much in the way of scope readings when you can only get a couple of revolutions with the kicks tarter. Works good in this case as well even with the electric start.

---

James H wrote on Sep 29^th, 2020 at 9:57am:

I built a remote starter for kick starter bikes and incorporated a VRS setup and 60 tooth wheel for better accuracy than the 1 trigger pulse per revolution. See pictures. You can put on the correct size socket to turn over the left side crankshaft bolt. Without it, it is near impossible to get much in the way of scope readings when you can only get a couple of revolutions with the kicks tarter. Works good in this case as well even with the electric start.

[smiley=rockon.gif] [smiley=thumbsup2.gif]

Pretty sweet. Also see a CPS, which I would assume you  put at TDC, the sync the sprocket at that point and you now have a CPS.

Thanks again guys for the input.

Larry

James H wrote on Sep 29^th, 2020 at 9:57am:

I built a remote starter for kick starter bikes and incorporated a VRS setup and 60 tooth wheel for better accuracy than the 1 trigger pulse per revolution. See pictures. You can put on the correct size socket to turn over the left side crankshaft bolt. Without it, it is near impossible to get much in the way of scope readings when you can only get a couple of revolutions with the kicks tarter. Works good in this case as well even with the electric start.

That is COOL!!!

Spence wrote on Sep 29^th, 2020 at 8:03pm:

That is COOL!!!

+1 [smiley=ok.gif]

James, so sweet, could also be used as a post hole digger with longer handles ;D