Syncloss only happens before and after the engine runs, given that its 1 cillinder 4 stroke it seems ok to me.
What type of ignition are you using at the moment, if its a CDI you can't run it whit the microsquirt unit.
Why is battery voltage dropping when you try to start this bike ?
I mean its a kickstart what causes the battery to drop till 10 Volt, bad or empty battery ?
What is the plan on microsquirt unit ignition and fuel or only ignition, wide band sensor ?
Yes it is CDI. CDI charge goes directly from magneto. I put dwell to constant 1ms so MS gives 1ms pulses for CDI trigger.
Battery is only small 1Ah. Magneto should be running but i wondering why voltage is still low.
I will use MS for both fuel and ignition.
Now you made me curious how do you trigger the CDI unit, how did you connect this to the microsquirt ?
How is the fuelpump driven, mechanical,vacuum,electrical ?
How much fuelpressure does youre system use ?
CDI needs pulse which goes first positive and then negative. Trigger happends +/- zero crossing point. Microsquirt INGOUT is logic level output. I use it to drive transistor which gives 12v pulse when IGNOUT gives pulse. I connected this pulse go though capasitor so i got needed waveform.
Fuel pump is electrical I can not remember bike which it is taken. Throttle body is from newer type Husqvarna. 3bar fuel pressure.
I would suggest you'll need at least an 8Ah battery, maybe even a 12Ah. I would also suggest you take a really good look at your 12v charging system; a kick-start carb'd bike with an AC (high voltage charging coil) CDI is NOT going to have much spare juice to drive a pump and injector.
(A 1Ah battery?! Seriously??)
Temporarily shut down - back soon!
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module TTR Ignition Systems
dontz125 wrote:I would suggest you'll need at least an 8Ah battery, maybe even a 12Ah. I would also suggest you take a really good look at your 12v charging system; a kick-start carb'd bike with an AC (high voltage charging coil) CDI is NOT going to have much spare juice to drive a pump and injector.
(A 1Ah battery?! Seriously??)
Does not have space for so big battery. I need only run fuel pump and keep MS alive short period before got bike running and charging. I changed AC to DC so charging should work.
Do some reading on the LiFePO4 batteries used by a lot of race bikes - they're a LOT smaller and lighter than a standard lead-acid battery. They're also a lot more stable and reliable compared to what they were just a few years ago!
Temporarily shut down - back soon!
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module TTR Ignition Systems
dontz125 wrote:Do some reading on the LiFePO4 batteries used by a lot of race bikes - they're a LOT smaller and lighter than a standard lead-acid battery. They're also a lot more stable and reliable compared to what they were just a few years ago!
Is there available plug in lipos? Usually lipo batteries need own charging systems (too complicated).
Modern motorcycle-purposed LiFePO4 batteries can be charged off the bike's standard charging system. The only gotcha deals with bench chargers, as any desulphination circuits must be deactivated or disabled. Since most bike trickle chargers don't HAVE a desulphination circuit ...
Temporarily shut down - back soon!
QuadraMAP Sensor Module -- PWM-to-Stepper Controller -- Dual Coil Driver
Coming soon: OctoMAP Sensor Module TTR Ignition Systems
About 5Ah would be nice but they are very expensive. I dont need bigger because does not have starter.
Some sources say that these are not working in winter conditions but if installed near any heat source i think it is not problem.
I thought that i got great idea how i handle starting situation so that MS can not give spark before TDC. I wired CDI trigger which gives pulse at TDC to VR2. I found missing teeth and welded it back. Configured MS to Dual wheel and put cranking advance to -10. Hoping that MS will detect second wheel pulse and start calculate spark advance from that.
Composite logger looks good. Got both signals properly. Tried kick decompression lever pressed. Fuel pump runs after every kick. Timing lamp does not flash at every kick sometimes two times at every kick. Rotated engine using drill. Again timing varies too much. Sometimes it gives spark 20dec before TDC sometimes 10 dec after TDC. Removed prediction and noise filters but same result.
Is here any developers who can explain how spark timing is done? I supposed every timings between pulses are calculated but pulse triggers are static.
Here in eastern Finland snowing literally laterally, but when i got collected enought courage to go carage i will try configure engine to even fire two cylinder. Fake 36 wheel to be 18 and that TDC pulse to be cam pulse. If not any change then started to thinking how i can install starter motor.
Good site about bikes CDI. I can already trig my CDI using MS.
I read MS source code and realised that kickstart will not going to work properly in my case. In worst case MS needs 1 rounds to get sync to toothed wheel and second round to get first spark. Kickstarter mecanically rounds engine 1 and half rounds. Flywheel is so light that it does not help much.
I must move to original plan and use relay to swich between original CDI trigger from magneto and MS trigger. Or start to hack MS code.
I am talking to myself-
I tried think through what happening in SW and why timing advance varies so much.
Engine rpm varies too much for SW to calculate proper coil fire timing. If i put 1ms dwell and spark timing to TDC SW needs estimate when engine is in 1ms before TDC to put coil charge ON to get coil fired at TDC. Because compession cycle slows down rpm so estimation is wrong and SW must trigger coil (timer is ellapsed). This have to do with "normal" coils to prevent coil to burn.
Because i dont need to worry about dwell times i have plan to prevent coil trig til i get trig2 pulse from magneto which comes at TDC. Fuel i will squirt every 36 tooth count (or every 18 half of amount). These only in cranking phase.
Does this sounds like a plan?
Never seen so much flags, global variables, relations and goto's in SW. I feels like if i change towels place in bathroom does my light still work in livingroom.
You are way over thinking this. There are a multitude of people who have successfully MS'd big bore,kickstart singles. Search for some of these post's and you might find a simpler solution to your install.
old guy wrote:You are way over thinking this. There are a multitude of people who have successfully MS'd big bore,kickstart singles. Search for some of these post's and you might find a simpler solution to your install.
Please could you give link? I did search but did not find complete project.
Yes i can make it run now with MS but it takes average 10 kicks every time. And risk to get kickback. I am started to scare those kickbacks after i got big one. Kicksstarter shaft bent and mechanism was toasted and many months of pain in leg.
My CDI trigger works fine when i use fixed dwell pulse. I will try does i get fixed spark timing if i use "fixed duty" dwell timing. When dwell increaces positive and negative pulses will be separated in my current setup. Any good suggest which kind of circuit i will need to get good down slope at different dwell times? Two different cases came to my mind. Using coil to get voltage spikes at every slope or adding resistor parallel to capasitor. Not tried yet neither. (Actually my input is inverted in drawing)
Does anyone know what "time after spark" and "Charge at Trigger" dwell types are? Help says that used in Saab and MSD type ignition.
Because MS have separate dwell for cranking and running state i am wondering if i put dwell type to "fixed duty" does it give fixed duty dwell in cranking phase also or does it give fixed time dwell which is set to cranking dwell time?
