| Author | Message | ||
     Jim_sb |
Howdy Gents, 2006 Ulysses, 16k miles since new, one owner, ECM Spy V. 1.12.1 and cable, stock bike. Always run on 91 octane from Shell, Texaco, Chevron, etc. Rider comfort kit recently added for additional cylinder cooling. No change to pinging AFAIK. My bike has pinged during roll on throttle in the 3200 - 4500 rpm range since new. Not during steady cruising in that range. Asked my local Buell dealer to perform a static timing check and they refused - said the bike was fine. Said they could not hear the pinging. I'm glad their business is hurting. They deserve it. So I rode it as is until yesterday. Finally decided to get serious about the problem and attempt repairs myself. 1. Performed static timing check. For the LIFE of me I could not get the static timing mark EVER to show up in the window anywhere near the 5 volt indication from the Cam Position Censor (CPS) as shown on ECM Spy diagnostics page. I could get it to show up nearly spot on with a ZERO indication from the CPS sensor (presumably meaning #2 cylinder is firing?). Questions: Are we supposed to be setting this for the front or rear cylinder? Stupid follow up - how do I know? I rotate the rear tire with the bike in 5th and it feels as if there are 2 closely spaced compression strokes followed by no compression for awhile. Duh. So which compression stroke am I after here? Obviously I am doing something wrong... 2. TPS Reset - the ECM Spy manual says 4.6 degrees whereas my Buell FSM says 5.2 - 5.6 degrees for the TPS reset. I used 5.3 degrees (when in doubt, trust the FSM?). Is that correct? 3. After giving up on the static timing check, and being frustrated with the pinging I drilled out the timing cover, put Sharpie marks on the adjustable timing plate and manually retarded the timing 2 tick marks (2 degrees?). Counterclockwise. I then went for a test ride. Pinging was completely GONE. I absolutely could not get the bike to ping - no matter what gear or speed. Snap the throttle wide open and the bike would accelerate. BUT - I feel like some power has been lost in the process... I'm happy the pinging is gone, but she sure doesn't seem to pull as hard. So - what's the trick here? Slowly advance timing 1 degree at a time until pinging returns then back off a tiny bit and call it good? Maybe if I had some type of clue on the static timing (I don't, because as noted above, I could never get the timing mark in the window anywhere when the CPS was on 5 volts). I'm open to suggestions. PS. Froggy said "Race ECM" and I sent an email to Erik Buell Racing but have not heard back from them yet. I don't mind the $250 if it resolves the issues but I feel a bit frustrated by the static timing issue. It shouldn't be this hard. I must be doing something wrong. | ||
| Reepicheep |
I'm going to dust off this old thread instead of starting a new one... I'm also reserving the right to make mistakes here, my first version of this is written this morning from memory of things I have done before, and things I plan to do this weekend. I'll update with corrections as I go forward. I'm playing with the timing on my 2007 trying to chase down a rattle. First to answer Jim's original question, the CPS goes to "high" or +5V six times for every revolution of the cam, and the cam rotates once every two engine revolutions. Only one of them is the right one for top dead center to set the timing, so there is one right one and five wrong ones. You picked the wrong tooth. This is because the cam position sensor assembly consists of a cup with 6 teeth (3 evenly spaced, a short gap, three evenly spaced, a bigger space). All bolted to the cam. That cup rotates past a hall effect sensor (basically a semiconductor that acts like a magnetic field switch, turning on when there is something magnetic close, and turning off when there isn't). Hall effects are less precise than an optical sensor or mechanical switch, but they are less likely to fail or foul. Here is an osciloscope trace for the Cam Position Sensor output for a properly running engine. You can see the three pulses, gap, and three pulses. You can also grind out the math with the time base for the pulses and see exactly how fast the motor is rotating, and the length of each tooth in the stator cup.  Truth be told, because the cams turn 1/2 time for every crank rotation, there are 12 (6 per revolution) "+5V" transitions from the CPS to choose from, and only one of them is Top Dead Center (TDC) for the front cylinder intake stroke. Which is where you time these things. So to set the static timing, you have to get close to the "right" tooth, one of 12 possible tooth positions on the crank position sensor cup. The right tooth is top dead center on the intake stroke for the front cylinder. So how do you find that? The easiest way is to start with a bike that is running Ok, but perhaps not quite timed exactly right. Mark the position of the timing plate (that contains the hall effect sensor) on both the plate and the case. That gives you a "back to factory defaults" setting. Then rotate the engine until you see the slot machined into the crank through the crank inspection hole (which you previously removed the allen head bolt from). (TBD... figure out an easy way on the XB to figure out TDC front cylinder on intake stroke... Something that doesn't require rotating the engine down. Blow air up through the exhaust? Only the right one will even show the slot anyway if you are close to factory defaults? I'd like to figure out something that means you don't have to take out the spark plug) Now that you have a good reference for top dead center on the intake stroke on the front cylinder, you can "change the timing" by either advancing the plate clockwise (advancing the timing and making a spark happen sooner) or retarding the timing by moving the plate counterclockwise). Ideally, the slot in the crank should be dead center in the inspection hole, and you should rotate the plate so it *just* goes +5v at that position. It's fussy and difficult to really feel like you are being precise. (people that know more than I do... help me with these two paragraphs) If you advance the timing too much, the spark triggers early, and the piston is still rising and trying to compress an already expanding explosion. The result can be an engine that pings badly, and runs too hot. If your timing is too retarded, the explosion happens while the piston is already moving away and you get less "push" from each explosion and you make less power. If you retard the timing by about 4 degrees (each hash on the CPS sensor plate is one degree), the timing will be so retarded that the bike does not run well, to the point it may not start or could stall if you blip the throttle. (help me with a value for Y here) If you advance the timing by Y degrees, you will get bad pinging (which is very bad for your motor) even at light throttle loads. The degree to which you will ping is related to the quality of your gas, the amount of carbon in your combustion chamber, temperature, etc. That really gives you an "x" degree window to play with, 2 degrees retarded through y degrees advanced. A 2007 Uly with the timing retarded 2 degrees ran cool without pinging and still made decent power, though not quite as much power as if the timing was set correctly. Would it tolerate cheaper gas? Probably. Would it prevent carbon in combustion chamber? No idea. Would it prevent pinging in harsh circumstances? Yes. (Message edited by reepicheep on September 26, 2014) (Message edited by reepicheep on September 26, 2014) | ||
| Reepicheep |
Lets dissect that scope trace just for fun... First, you can see from the top that the scope is set to "1V/Div". Which means each large square vertically represents one volt. The square waves are about 5 volts from top to bottom, so the CPS is changing by 5 V every time a tooth passes. The horizontal axis on the scope shows time passing. The "time base" is set to 20mS / Div, which is 20 milliseconds per division. That means .020 seconds pass for each larger horizontal square. Since we can see from the scope when each cam rotation is complete, we can add that up to determine RPM! Count the number of horizontal squares from the start of two successive "big gaps" to establish the time for one revolution. In the case here, about 4 squares. 4 times .020 seconds is .08 seconds per revolution. If it takes .08 seconds for one revolution, it will make 12.5 revolutions in a second (1 rev/ .08 seconds = 12.5 revs per second). In one minute (60 seconds), that would be 750 RPM (60 times 12.5). But remember the cam is spinning at 1/2 the crank speed, so we need to double that RPM So the RPM of that motor at that moment is (750*2 =) 1500 RPM. What have we learned? First, we have learned I'm a pathological geek. Secondly, we have learned I set my idle too high. Thirdly, by looking to the left of the scope at the length of tubing kept under my seat, we can learn that I might not be above stealing gas from someone if I was stranded  . (Message edited by reepicheep on September 26, 2014) (Message edited by reepicheep on September 26, 2014) | ||
| Reepicheep |
OK, dug into the 2007 Uly and did some "what if" analysis to try and put together the pieces. My first question was about the biggest PITA factor of setting the timing, which was finding TDC on the front cylinder intake stroke. It's annoying to get in there and get that spark plug out and figure out which position is what. So I tested both, TDC of intake stroke, and TDC of exhaust stroke. On my Uly, both were extremely close to each other, as expected. Perhaps half a degree different to get the "right" timing. I assume this is because of that stamped cam position toothed cup, and that the stamping isn't exactly perfect. So if you do it on intake stroke instead of exhaust stroke, what happens? Not a lot, its just a little different. So if you want to set it up to be exactly right, you do need to get the right position. However, it won't be *that* different. So you might be able to take another approach, do both positions (one crank rotation observed by watching the timing mark through the inspection hole), and pick the one that retards the timing the most (meaning it may be exactly right, it may be half a degree retarded. Given these are air cooled motors, having the timing either exactly right or possibly a half degree or so retarded seems like a reasonable approach to me. I tried riding my Uly with the timing retarded about 2 full degrees just to see what it felt like, and I noticed just a little less power, but not much difference over all. If you try this, and see more than a degree of difference, then maybe you need a less sloppy timing cup. |
