Idling but problems with spark at high RPM

[aliga]

Hi,

we're having troubles with the configuration in Megatune in the spark settings. We have an engine with 1 cilinder 4 stroke with a decoder wheel of 24-2. There is a file attached showing the configuration of the VR sensor. We don't know if the next configuration is ok because we're having troubles when the engine rise at 3500-4000 rpm (at this moment, we're using only the spark, not the injection part):

- Trigger wheel teeth: 24
- Missing teeth: 2
- Skip teeth: 24
- Delay teeth: 6
- engine stroke: 2 stroke ( to calculate rpm ok)
- trigger offset: it works with 10º but we're not sure if it is set ok.
- skip pulses: 1

other configurations are in the attached images. The *.msq file is also attached

Thanks!

base ignition settings.jpg

(50.08 KiB) Downloaded 16 times

wheel decoder config.jpg

(30.74 KiB) Downloaded 11 times

[1979-mini]

Aliga,

Is your advance table too aggressive (to advanced) at 3500-4000 rpm? What happens after 4000 rpm? Usually, the advance table tapers off to a level number around 3500.

Good luck.

[grippo]

Is the timing, as measured with a timing light, correct between idle and 3000 rpm, and does it agree with the MT timing display ? In looking at the msq, the pulse tolerances are between 5 and 15%. That seems too low. I would put them at the default values or al least in the > 35 % range.

[aliga]

We will try tomorrow with other advance table (soft) and change the pulse tolerance. The timming seems Ok with a timing light. Tomorrow I'll post the new results. Other assumptions are ok? (engine as 2 stroke, skip pulse, trigger ofset) If we put the engine as 4 stroke, it results in a double rpm.

Thanks!

[grippo]

Your assumptions seem ok, but what is the trouble you are having at 3500 rpm. Does it miss. If so we need a short datalog of from idle to when the missing starts..

[aliga]

Dear friends,

we have tried differents configurations. Now, it feels like the engine is running better at low rpm but have missfires at ~3500 rpm and it can't accelerate. It turns off if we try to accelerate it. We are now using a very simple advance table: the whole table have a 15º of advance to eliminate the effect of MAP sensor. It results in a 19-18 degrees of advance assuming the extra cold advance. We tune offset with trigger wizard and the trigger light, it is about -3 degrees. The engine is a 35cm3 (yes, too small becouse it is used for a low comsuption fuel vehicle race). We need this engine running in a few days. Attached there is the new .msq file, which have a lot of corrected issues. Next days I can post a datalog file but if you have any suggestions now, please send it!

An other question: we did not connect the TPS becouse we are not using it. It is strongly necessary? Or we are in the correct way and uS do not need it? Is it the error? (We assumed that is was not necessary!)

Thank you very much!

megasquirt200804231651.msq

(25.01 KiB) Downloaded 44 times

[grippo]

First you need to verify that your timing is correct. At idle look at the guage in MT and the timing on the timing light. They should agree, if not, adjust trigger offset until they do. Then move the rpm up to about 2500 - 3000 (before the miss) and compare timing again. They should be the same. If they are, then you have set it up ok - don't change any more except for the advance tables.

As far as the miss, you may be getting noise in the tach input signal. That would be shown by looking at the guage in MT and see if the rpm jumps to 0 for an instant when it misses. (The best way to tell is a datalog.) If there is no drop of rpm to 0 (not 0 on the car but on the datalog/ MT gauge), then something may be wrong with the wiring, or coil, or the coil driver(VB921) may be dying.

[Bruce Bowling]

An other question: we did not connect the TPS becouse we are not using it. It is strongly necessary? Or we are in the correct way and uS do not need it? Is it the error? (We assumed that is was not necessary!)

If you are not using the TPS, then I would tie the TPS signal to sensor ground so that it does not trigger enrichments.

For the miss problem I would first make sure you were not one of the few lucky ones who got the harness error where the shield and the center conductor are reversed - the center conductor (VR+) goes to ampseal pin 32, and the shield (VR-) goes to pin 33. Next, I would place a 1K to 5K resistor (start small and work up) in series with the VR+ input in case there is sensor overdrive to the VR circuit.

See what that gets you...

- Bruce

[aliga]

More things done:

- coax and center wire were reversed. Now they're in the correct position.
- Timing between 2000-3000 is ok
- Offset setup is ok with the timming light
- A new coil and VR sensor were tried, no changes

Still we are having rpm drop to 0 at ~3500rpm. Now, we must try with the 1K-5K resistor and let's see what happens.

When we tried to connect the uS the first time (three weeks ago) we fried the VB921 drivers due to the coil (bad connections, newbie experience...). Next, we substitute the fried drivers with a new one on IGN_1 (there is no driver in channel IGN_2 now). The PCB has a little burned zone, but when we substitute the driver we verified that there is no bad conections. Maybe it is not soldered ok and is this the problem? The burned zone in the PCB is in driver IGN_1. Can we use the IGN_2 instead IGN_1 only for 1 cylinder?

Thanks.

[grippo]

If you use Ign2 with your setup, you would have to use only the driver section and cut the trace at some point from the driver back to the processor because the processor is not putting out any spark on this pin. Then you would have to connect the ign1 output from the processor to the ign2 driver.