MicroSquirt.com read only forum (Jan.15,2018)

Looked through the manual as well as i could, must not be that bright, but i cant seem to get the settings right or i completely misunderstand what it is they mean.
Here is what i have.
MS2 (blue) duagtercard
V3 board
V2.980 code
75 degree odd fire v twin COP and i injector per cyl. single crank pick up 24-2 teeth.

Here is what i have tried so far:
first effort was MS2 extra v2.xxx ( latest standard release)
got it reading rpm just fine and set it to odd fire, but could not get the second spark output to work, would not accept two coils on a 2 cyl. in wasted spark. Ok, no big deal thought it was worth a try.

So now I have made the board mods required to run dual spark on the daughter card, lifted pin 4 of the 74HCxx chip and tied it to pin 1, ( only 1 tach input ) and used pad 4 for my second IGBT driver.
loaded the 2.98 code and set up MT to run it as a microsquirt (2).

so, if i disable the dual spark and input the info for crank wheel, ( 24 teeth, 2 missing ) it reads RPM just fine. as soon as I try to enable dual spark, things go bad. RPM reads 4x what it should. under niether of these set ups does the second coil fire when it should consistantly. I do not have the injectors or coils hooked up at this time, rather i am using a 4 channell lab scope to monitor the stock signals, and am using a 12v wire through a 10k resistor to see what the individual MS outputs are doing.

So..... what in the heck am i doing wrong? what should my settings be?

ok, made some progress but still confused....but im easily confused. found this

For example, if tach1 occurs at a trigger offset of +40° BTDC, tach2 occurs -320° ATDC (and there is 360 degrees between firing, so a 2 cylinder engine). For dual spark mode we always want a negative trigger offset, so you would use -320° instead of +40°.

The -320° offset seems unusual, but there is a reason for it. The original code started with a very robust triggering system where people could put their sensors anywhere and use any kind of advance with any kind of offset. This was a nightmare to code and impossible to maintain with the dual spark system, which was based on a strictly next cylinder firing setup. In fact it sets up to fire 2 cylinders before the actual spark occurs. This allows plenty of time for dwell at high rpm. To maintain the previous terminology for users as well as keeping the code consistent, the advance offset was left as it was, but whereas for the original (non-dual spark) system it was best to use a positive trig offset (but negative offsets would also work), while dual spark requires a 0° to negative (ATDC) advance offset

so i started playing around with "-" numbers, still no good.
Here is what is actually working right now:
Trigger wheel teeth = 24
missing teeth = 2
skip teeth = 12
delay teeth = 19 ( gotta kick myself on this one, waited way too long to actually scope the VR vs. stock coil firing to figure out approximately where TDC was in relation to the missing tooth )

Single crank wheel input (dual spark)
offset(advance) for cyl. #2 (degrees) = +250

trigger offset =0

now cause I dont fully understand the math yet, the +250 offset for cyl. 2 may be off by +/- 5 degrees or so, on my scope it looked very close to the same firing point as the stock ecu, but it was getting late and i desided to fine tune it later. So with those settings I can let MS controll both plugs and it seems to run very close if not identical to stock. only really able to compair low rpms because it set the whole timing table to 14 degrees ( acording to the stock map this is where the factory has it).

But I am still trying to figure out how +250 works, according to the manual positive numbers dont work as good as negative ones, but i couldnt get anything with a negative input. I also need to understand the math a bit more, i cant make 250 work with a 75 degree ( -15 from normal 90 )in my head....

in any case, those of you more familliar with how this works, are these numbers ok? i mean it runns at 1500-6000 rpm with 14 degrees advance, but will it run at 10000rpm with 36 degrees?
thanks, i am going to go back and study the manual a bit more to see if i can wrap my head around it

If you get it running right at low rpm it will run right at 14000 rpm, at least I have never seen a problem unless it was the coil or injector drivers that got too hot. If you match up your PC advance gauge and timing light advance at say 1000 and then at 3000 or wherever you have full advance then you should be good to go.

The idea of the odd angle is that you just treat your 2nd cylinder as if it were not odd-fire, so pretend it comes up at the same time as the 2nd cylinder on the same bike with an even fire engine. Ok now relative to that, how much fixed advance do you want to add to the 2nd cylinder. It will have normal table advance, same as 1st cylinder. But in the odd fire case you had a fixed advance on top of that. You could also add a negative retard, but generally it is easier to add advance, and if necessary, swap the 2 cylinders to make this possible. The part that is not so simple is really understanding how your engine works, because you have to know that to be able to set the inputs properly.

Thanks Al, I actually think i figured it out today at work, I was making it way too complicated (something i seem to accell at)

250 works because its very close to 255, witch is what it should be, after #1 ( we will say for now, more on that later ) fires and completes its 180 derees to BDC, cyl #2 will be at TDC 75 derees later, i.e. 180+75=255,(common crank journal 75 degree vtwin) way to simple. the fact that #2 is actually on its exhaust stroke is meaningless as it is in wasted spark. I was trying to do all manner of back calculations and rotational computations because i was sure it had to be more complicated
as far as swaping the cylinders, after realizing what a dolt i had been for over complicating the issue, swapping the trigger inputs around to the rear cyl became an obvious alternative ( smaller numbers ). But, as it it firing ok now i think i will leave it. and I was aware that older code versions had more trouble with lower RPM tach calculations than higher, i should have realized this would also aplly to the newer code.
I spent most of my spare time today studying the alpha-beta-gamma trigger predictor, what a neat bit of code that is. I think my first goal will be to tune those trigger settings to achieve the lowest average error factors.
then i will move on to playing with the blended alpha-n fueling. the 2.98 code is not what i was expecting, may need to play with that a bit to get the desired results.
My thinking here is that this engine will benefit more from a low rpm SD blended to a high rpm alpha-n. this is due to the fact that this engine has two MAP sensors, 1 per TB, and auxillary ECU controlled butterflies that should allow it to build a usable MAP load signal at low RPM mid throttle settings.
So, also on my todo list is to try to find the old thread on the MAPsequencer, and get the shcems for the version that everyone agreed on and see how it will work on a 2 cyl.
The end goal is to get as much HP out of the motor as i can,and make it smoothly, but also to prove a point. Surfing around on several bike specific forums it was pointed out by many that this bike was prone to snatchy and twitchy fueling, and spending lots of money on aftermarket parts and professional tuning was mandatory.
When I spoke up about MS I was summarily run through the ringer by quite a few well intentioned folks. But MS has never let me down, and i've MS'ed at least 5 different vehicles and had great results with all of them. This is my first motorcycle though, and a high revving twin to boot. Looking foreward to it and hope as the project progresses i can supply more info and less questions to the group

What you want to do is put the 2 maps thru Bruce's synchroMap circuit and send 1 map to the microsquirt. That should considerably reduce fluctuations.

I wouldn't waste too much time with the alpha-beta-gamma prediction - it doesn't seem to provide much improvement. What really provides accurate timing is to use the dual spark option with a 24 tooth wheel.

For hybrid alphaN, there is an option that allows you to change the sense of alphaN and SD. Depending on this switch (an input option) you can pick AN below Rpm1, SD above Rpm2, OR, the opposite: SD below Rpm1, AN above Rpm2.

Thanks, I will play with that as soon as i can get my next issue sorted out.

So, here is whee im at, I've got the tach input settings squared away, one minor issue with cranking timing that i will get to later.

But right now i am having a VERY strange issue with injector control. once i got the spark signals matching up on my lab scope ( stock vs MS ) I was able to let MS take over control of the spark and everything worked perfectly.

So then I went on to the injectors, I started by scoping the stock injector signal, and with a 10k pull-up resistor ( hooked to 12v+) I scoped the MS injector signal. It didnt take long at all to get those two signals to match exactly either, with injector timing offset and Requ_fuel I was able to make the MS fire its (dummy) injector at exactly the same time and for exactly the same duration as the stock ECU, 2.4ish-ms.

Only when I try to let MS fire the actual injector ( small bosch ~12ohm) it wont fire it. I checked the wire to the injector with a test light to make sure it would carry the current ( ~1 amp ) and that was ok.
So I am a bit confused as this MS came off a running car about 16 months ago. I scoped the pins on U4 (V3 board) and pin 6 has 12v+, pin 3 is grounded, and pins 2 and 4 are showing 5v+ when MS is powered up (engine not running) .
So that all seems ok, then I started the motor ( now only hitting on 1 Cyl, and scoped pins 2 and 4 again, straight 5v, no signal from the MS2 card at all. ????

So then I scoped pins 2 and 5 of U2 (MS2 daughter card) and I do see short 5v+ spikes on those pins, so I am certain that the MS2 CPU is ok, but its pointing towards a board problem somewhere between pins 39 and 40 of the DIP socket and 2&4 of U4, or a problem in the daugther card itself between pins 2&5 of U2 and pins 39&40 of the dip socket.

I am verry confused as to how this could have happened, as i said, This board came off a running car. I will do some more testing to try to isolate the problem, but any help is apriciated!

Ok, I think I see the problem, was looking at the output of the MS2 card, inj1 and inj2 are basically hard wires to pins 39&40, so i thought those were the injector driver channels, they are actually the fly back channels. So I need to go look ans see if I missed something in the MS2 (card) to Microsquirt code conversion. So just to clarify, does the Microsquirt fire its injector channels the same way as as the MS2? Do I need to do some more modifying to the daughter card to make it work or is something more sinister going on?

I just checked for continuity from pins 3 and 6 of U2 (MS2) and pins 2 and 4 of U4 (V3) and thats ok.

Or, am I completely barking up the wrong tree? I was led to believe that Microsquirt code could be implemented on MS2, is that not true?
I'll go search again to see if i missed something.......... all I remember seeing was the mod to pad 4 of U2 to use it as the second spark driver.
Thanks, Joe

1. Don't lift pin #1 off the NAND gate.
2. Lift pin #4 off the board, and jumper it to pin 1 with a wire so that the PWM1 pin is output to both pins 1 and 4.
3. Use the former pin #4 pad for the 2nd ignition channel as above.

This is what I did, set up for 1 tach input. but the option for two inputs is:

1. Lift pin #1 on the NAND gate and connect to pin #2. This in effect makes a simple inverter,
2. Do the same thing with pin 4 - lift and bridge over to pin 5. Now, there is nothing on pad #1 and pad #4,
3. Pad #1 is used for the second tach input (Input 2),
4. Pad #4 is for the second ignition output channel (Output 2).

So, understanding a little about the NAND gate , most likely my injectors are not firing because pins 1&2 and 4&5 are not bridged, ( I just bridged 1 to 4 as the instructions stated ).

So...... I am assuming there is a problem with the instructions? if I understand the NAND gate correctly, both pins 1 and 2 have to be activated to produce an output to pin 3 ( and thus to pin 2 of U4, V3 )
and the same applies to pins 4,5 and 6 of U2. Am I correct?

OK, so that did it, pins 1 and 2 and 3 an4 4 of the NAND gate still need to be shorted together, not pin 4 to 1 as the instructions stated.

Now I get to start tuning fuel....... alpha n, blended......AE, now the real wok begins. Thanks!