Code Request for Honda RC-51 and the likes.

[OB6D]

I wonder where these come from... Ohw wait... that's where the operater makes adjustments i suppose...

[OB6D]

I guess it already shows on the left side of the above picture. But i think i'm having lean spikes. And i think they're caused by either those VAG cops or the way i've set them.
I'm using these http://www.bremi.de/user/eesy.de/bremi.de/dwn/20116.pdf the Bremi 20116. And i've got the E version, which from what i've read somewhere is an older model then F and G?

I've been fumbling with the fueling for quite a while, and up till 3000 RPM it's fine, but then i seem to hit some sort of invisible wall, i can't get past that 3k mark.

Did the output test on both COPS, and strangely ennuf it seems they do better with less dwell. I ran them both at 12k RPM with 1ms of dwell and they still work (in the open air). So i suppose i've been overdwelling them??? Would that kill them? Partialy? Totaly?

And this is a nice link it has a picture of a lot of those COP things grouped:
http://www.en.bremi.de/user/eesy.de/bre ... glisch.pdf

They all look the same, what's different about them and which one's the best buy?

[OB6D]

So, what i've done now...

I had this feeling i was having spark issues with the VAG COPS. So i wired up the original CDI ignition that came with the bike. You steer this CDI amplifyer box with the Us on Logic level just like the COPS, so that was easy done. But to my surprise the engine runs totaly different on that. It stomps. The part of the Ve table i had tweaked with the VAG COPs seems now drasticly rich. Also the idle speed i'd set at 1300 is now in the 2000's all of a sudden. I pulled out the strobo and aimed it at the timing hole, but i can find no marks no more...

I've read somewhere some time, that the ECU of the bike holds the spark time table, but i'm now thinking it's maybe in the CDI box. But that's a guess... as is with many things in this project...

I dono if it's common to put a spark table in such a CDI box, but i'll look into it... when i find the courage again...

[OB6D]

Figured out the spark arrived too late. Put cranking advance to zero. Put advance at fixed at zero. Then when cranking i looked at the marks and gave it an offset. It needed -20º to arrive at (TDC, flat on the head, zero degrees) the set time. Then i put the plugs back in 'n tried to start it. Spark was now 20º too early. I removed the offset. Then it started (ran) again. Then i stopped it and put ignition back to use the table. It seems okay now, for now, it idles properly again...

So make a change, undo the change and then something has changed.... ???

This is never gonna work.

[jsmcortina]

Could you try cranking the engine and strobing it, then stopping and repeat say about 4-5 times. Is the timing consistent or does it vary from start to start?

If you are getting a variation (like 20 deg different on one start) then it would suggest the code is picking up the toothed wheel incorrectly. IF that was the case, then it should be fixable.

James

[OB6D]

Could you try cranking the engine and strobing it, then stopping and repeat say about 4-5 times. Is the timing consistent or does it vary from start to start?

If you are getting a variation (like 20 deg different on one start) then it would suggest the code is picking up the toothed wheel incorrectly. IF that was the case, then it should be fixable.

James

If the toothed wheel, of either the crank, or the cam, would be picked up incorrectly, i think i should get a loss of sync error then. I think it wouldn't sync at all. I normaly get a few sync losses at cranking, then it fires up and it will never loose sync again, unless i provoke one. I've had numerous sync losses, but i figured out, they always occur on a missfire.

But as i "progressed" i learned the diff between TCI and CDI and so figured out, i was running a hybrid system. MicroSquirt (TCI), driving CDI coils from the original ignition. And recently i've been told, it isn't really CDI, the box is just an amplifyer. I suspected this to be at least part of the problem... I could not even touch the throttle then without going boom, but it would start and idle always and just fine.

Anyways, i switched to logic level and the VAG COPs. This seemed an improvement. I never looked at the timing again though. I can't even, how do i get the strobo working with those... It sounded alright to me.
Then i went tuning the fueling. I got up to 3000 RPM in the end. And until there it would be fine. But i've never been able to push beyond 3000 RPM then it starts spitting, coughing, it goes boom, and it looses sync all over again. And whatever i changed, i could not get past 3000 RPM.

Knowing these things are sensitive to overdwell, that i got em as used parts from a salvage yard and that they're the old E type ones... And knowing that i spent ages of fumbling with the Ve table, i figured these COPs must be failing me past 3000 rpm then. But i realised it might also be the timing, because on spark output test they both hum soundly at up to like 12000 RPM, 100 sparks a second...

In the meantime i'd figured out that i could get sparks from my "CDI" amplification box the same way, the VAG COPs work. So then i wanted to go for that, since it's original parts and i would be able to use my strobo again. So with my latest .msq and no changes to it, i just went for it. And it fired up, eventualy, only to run completely different. Just not drowning, that rich, my idle rpm went from 1300 to over 2000 and the engine made stomping noises. But it ran...

My current strobo, doesn't have a retard function. But i'd made paint marks at the head of the alternator bolt behind the bigger center hole and i could see ennuf to tell it was way off timing. So what i did is, i set advance to 0, then my Ft (front is on top) and Rt (rear is on top) marks should appear at the designated timing hole. The didn't, not until i set -20 as an offset. So i figured ok, that's it then. Then it wouldn't start anymore, until i undid it and the timing seemed okay, all the same... This was yesterday...

Then today after reading through the manual once more, i got the idea of having the input capture or output wrong, or both. I quote:

If your module relies on MegaSquirt® to set the dwell, the 'Spark Output' should be set so the coil does NOT charge when the engine is off.

You can test this by looking for:

a voltage drop across the coil (which means current is flowing, which it shouldn't be if the engine isn't running), or
by hooking a test light in place of the coil (it shouldn't light if the engine isn't running).

So i test this, my testlight doesn't light, i must've guessed it right. So i figured, i'd change the setting anyways and then my testlight should thus come on. It didn't... Me puzzled.
I'm now guessing this only applys to kettering style ignition. With breaker points, i've been there, done that... this seems familiar...

Anyways, i decided to have a wild guess and just changed falling edge to rising. Also changed crank spark, from 5º before to -1º after and the fixed advance from 15º to 5º. Then it almost fired right up, almost...
And it runs at 1300 RPM again and it doesn't seem so rich no more either. And, that stomping sound went. So, to me it's clear i do have some timing issues.

I quote, from http://www.megamanual.com/ms2/tune.htm
You *could* use either edge IF both event are fixed with respect to the crank position (TDC). This is true for optical sensors, and may be for Hall sensors as well.
With a VR sensor, both of these event are definitely NOT fixed, only one is (it corresponds to the 'zero-crossing'), so you have to be sure to get it right.

I don't get it, but i do get that it has to be right, twice i suppose, both crank and cam is VR. I do not know for sure i got it right. I went for what seemed to give the best result.
I do understand rising/falling edge, that's just where the sinus output, crosses the Y-axis, either the upgoing slope or the downgoing one is relevant. I get that. But i don't get why it matters, both events should have a fixed space in time... And i sure don't know how to tell which one is the right one in my setup.

Also, i'd like to know if this input capture depends on the sensors used, or on what they're wired to. Do these input capture and output settings stay the same no matter what coils you use? Or do they have to be changed to cope, with COPS, high out straight to some coils, or again low level to some amp box? Because the last one in particular, seems to influence timing...

And i've ordered a new stroboscope lamp, so when it gets here i'll be able to tell, just how much the ignition is off and if timing wanders.

Could you try cranking the engine and strobing it, then stopping and repeat say about 4-5 times. Is the timing consistent or does it vary from start to start?

If you are getting a variation (like 20 deg different on one start) then it would suggest the code is picking up the toothed wheel incorrectly. IF that was the case, then it should be fixable.

James

First thing when i get my new lamp. I'll do that.

[OB6D]

On the picture on top of this page, i reported these PW to zero droppings. I can't say i'm still having that with my "new" ignition. Also on the VAG COP logs i had noticed lean spikes, another reason to believe they'd not be 100%. Can't say i'm having those on my recent logs either...

The connectors for those VAG cops i found in a waste wiring container...
Now the pinout was clear, power ground, ECU ground... But, the wires on the plugs, were a thick and a thin one and they suggested something else... So i ehhh... X

But the pinout of the short ones, is different from the long ones...

[OB6D]

The crank Y and run Y bit, mostly come together, which means that sometimes i'm cranking with my ignition advance, not cranking advance. I'm also thinking that there's some timer function on this switch of state and that it takes some time to get back into cranking mode? Under cranking settings i'd already raised cranking rpm a few times, i've kept it on 500. Then why does the Run Y bit come on, when it's still reading RPM far below that? And why if it does think it's running do i get no PW?

There never is any RPM in the trend while i'm cranking, but occasionaly it shows 1 RPM. Even on flood clear, with WOT when it cranks faster, it's 1 RPM at most.

Sometimes when a loss of sync occurs, it triggers the crank bit, this goes high and then i'm running 2200 RPM or so on my crank advance.

Also i've noticed that the lost sync count and the count of lost sync reason spikes is not always the same number.

My ignition timing may be off by a few degrees, but it is seems to be consistent to me. Also triggering from either edge, going high, going low, it doesn't seem to matter much, it will run and when it does the engine shows the same behaviour. Changing the timing of the spark though, does make significant difference, with starting and running.

[jsmcortina]

Diagnosing startup issues like this is very tricky.

They key thing here though - the code is not syncing to your wheel reliably.

I have been doing some work recently on improving starting conditions. The SD logger in MS3 has really helped me log tooth by tooth and examine many internal variables within the code. From this I have made some improvements to start spark and fuel a little sooner.

However, this all presumes the wheel is syncronised. As you are getting a loss of sync, your engine is not going to start.

Please could you record composite logs during startup and send them to me. jsm@jsm-net.demon.co.uk

James

[OB6D]

Files created and sent as requested.

[OB6D]

Got the new timing light. First i wanted to report that the offset didn't do anything at all. But since it's pretty confusing i had another shot at figuring it out. I just could not figure out how to bring the Ft (front is on top) mark to the timing hole. [well in retrospect, with a 28 offset and the lamp at 18 it would line up, i was just thinking the wrong way].

But i'd already reported it was late with the spark, which i still believe (it's just not 20º late but "only" 10º), but this adjustable lamp only works one way and thus not in this case.

I had to put advance at 10, to get the spark at TDC. And the Ft mark lining up.

So i then put an offset of -10 and advance at 0, which should also be TDC and thus the exact same, this was correct.

Then i put advance at 15 and i end up with 25, which is wrong.



Tested cranking, i'll try a running one later.

Also, i have this feeling the timing... what's it called oscillates? That it's one time correct, then one time not. Cuz many times, while testing and looking for the mark, i caught myself going, yes, no, yes, no, yes, no, yes, no...

[jsmcortina]

Sounds like the wheel definition in the code is incorrect.

James

[OB6D]

Sounds like the wheel definition in the code is incorrect.

James

Yes.

I tried to run it with the offset, but it didn't want to. Then when i put offset back to 0 and crank adv to -2 and fixed adv to 5 (as i had it), it wanted to fire up, but the battery was empty.

Anyways, it's in your hands again now then. Just trying to provide all the info i can...

[OB6D]

I think a timetable starts at 0º with cylinder 1 on top, ready for the intake stroke. I know my timetable didn't have that. Also i suggested that the rear would be number 1 and the front number 2. Ignition output 1 sparks the front, so it looks like you didn't go along with that. So i made a new timetable with #1 being the front and starting at zero...

[OB6D]

[OB6D]

[OB6D]

[OB6D]

From C to the inlet stroke it must be about 90º times two somewhat is 180 + 120, yeah i got it right this time...

I should try a 360º offset. Can't believe how "well" it worked being so far off... strange...

Because your "new" diagram is the exact same as the other one dude, just shifted 360º ...
Someone's been looking into the stroboscope lamp for tooooooooooooooooo long.

[OB6D]

Sounds like the wheel definition in the code is incorrect.

James

Don't want to push, but do you have an ETA James? Want to use this bike for the next track season... if you can't help me any further i've done all i know how... then i'm not gonna put the Us on the bike, then i'm gonna put the bike on the Us.

[jsmcortina]

Sounds like the wheel definition in the code is incorrect.
James

Don't want to push, but do you have an ETA James?

Little thing called Christmas has been in the way.

then i'm gonna put the bike on the Us.

Take a video if you do (!?)

James