So, how many teeth do I really need?

[SQLGUY]

Back home from vacation today. I desoldered the VR trigger wires and tried switching pickups and changing polarity. It didn't seem to make much difference which pickup I used, but Microsquirt did have an easier time synching after mechanically switching polarity and changing trigger capture from falling edge to rising edge. It's still not perfect, but it probably synchs one of out two times when cranking rather than one out of 15 times. Both pickups worked better with inverted mechanical polarity (a setup which couldn't be done with dual tachs, since the pickups share a common "ground").

I ended up switching to the bottom pickup, which changed delay teeth from 6 to 2. Again, I didn't see any difference in behavior when switching pickups, but it just made more sense to me to use the 1/4 pickup rather than the 2/3 one.

If you think of anything else I might be able to do to improve initial synch reliability, please let me know.

For now I'm going to go ahead and dyno tune with this setup.

Cheers,
Paul

[24c]

Hi Al.

I'm finally up and running on this project again..the FZR1000 crank with the large slot. I ran the TachRef program and got the plot below. Can I use these for a crank tach signal and combine it with a cam signal?

My next step is to get some BNC adaptors and scope the crank/cam and ignition outputs, but if this doesn't look OK.

I am going to switch cranks in my rig for a 4 crank teeth +1 cam tooth set up, get that working and then try and mod the teeth values for the "8" toothed crank later(with a cam tooth pick up too). I have got stable cranking revs etc, no loss of tach, etc on a datalog, but when I drive an LED to pretend to be a coil the output is intermittent and not regular except at higher cranking speeds.

Any thoughts ?

[grippo]

If you put a piece of paper over the plot and mark off the rising edges, you will see that all the marks are about the same distance apart, so as far as the processor is concerned (which only detects the rising edges) there is no missing tooth. If you set it up for falling edges, things look better, not perfect, but it might work. Then just set the cam sync so its detection edge falls approx. in the middle of the missing tooth.

As far as the missing output pulses, if a datalog shows no missing tach inputs, that is no +/- triggers and no rpms = 0, then either something is wrong with the configuration or wiring or code. I would need the exact msq, code version and rpm range to see what might be the problem.

[24c]

Hi again Al
I have been playing some more with this crank set up, trying all sorts of permutations and using your TachRef program to analyse the cam/crank signals. It appears that I can get the MicroSquirt to work on the rig.

Eg using the "Dual Tach Inputs" it produces a very overscaled rpm, trigger +- & tach is stable, rpms start cranking easily and I have regular flashes on the LEDs off the injector outputs and a much faster LED flash rate on the IGN1 output. There is nothing on IGN2, but it says this on the documentation. I might be able to get this to work with a hardware mod on IGN1 so that I alternate this output into my ignition drivers, and maybe I might be able to do something scalar with the rpms, but worst case they would be bigger numbers than reality, and I'd compensate elsewhere.

My question is your TachRef program doesn't see the crank and cam tachs together, I had to switch the cam sensor vr signal into the crank sensor of the MicroSquirt to make it work, and do them separately. I notice on your active window you can change colours to the cam and crank, but I only ever got a VR1 input into the TachRef program. Is this expected? It would have been nice to have seen both together, for synching in my case ( I used the oscilloscope in the end).

[24c]

Hi Al and others,
Just a few ramblings, I experimented with placing steel washers over the slot, using double sided Scotch tape, and using yourTach Ref program I was able to see what was happening.

I played with the "Dual Tach Inputs" , but then reverted to the "Dual Inputs, Timing from 1 Cam Tooth", which had partially worked intermittently, using an 8 tooth wheel, 0 missing, 4 skip teeth & 0 delay teeth.

By adjusting the Alpha,Beta,Gamma to 50/50/50, I could get it to crank, but as soon as it flagged ASE/WSU ON, RPMs =0 and trigger +- was going negative. I modded the "Next Pulse Tolerance" settings for "After Start" to 90% (crank was 90%), and it kept holding Tach, no RPMs=0 and trigger +- =0, but when it got to normal running, it died again RPMs=0 etc.
Again modded next pulse tolerance for running to 90%, and everything seems OK. I will post a datalog later, after I check the outputs with the LEDs.

Looks promising though, as it cranks on the rig, the RPMs fire up straight away, then it's CRANKING, and goes through the ASE/WUE routine, and keeps running.

UPDATE, after testing the outputs with LEDs...the injection outputs are getting confused after cranking mode, because of the large tolerance values I think, and I only have IGN1 output, nothing on IGN2.

UPDATE 2 It's not the Pulse Tolerance, as I thought, it's the Alpha,Beta,Gamma settings...just tried 60/60/50 and LED on IGN1 looks OK upto 2K rpm. I'll have to read the manual more...see what is going on

UPDATE 3 It seems the Alpha, Beta, Gamma plays a bigger role with this crank wheel, because all of a sudden other Input Options on selection are holding tach, cranking OK etc, not perfect looking at the LEDs on the output, and not sure about the timing of events yet. Also I can disconnect the Cam Sensor ( Dual inputs Timing from 1 Cam Tooth) and there appears to be no change in LED light pattern. The Tach ref program shows it's working OK.

[24c]

Hi

I have decided to give the unmodded Yamaha YZF/FZR crank a rest, as I have just added a solid bit of metal to bridge the large slot, and used the "SingleCrank Wheel" option, banged in data, 8 teeth, 1 missing, 4 skip etc, and left all the defaults alone. Cranked the rig up after disconnecting the cam portion, and up and running straight away no problems. Getting alternate ignition outputs, simultaneous injection, until running, then it switches to alternate banks... easy.

Can even switch polarity, or rotation with no apparent issues with this set up...yet

[grippo]

If you look on page 2 of this thread you will see 2 plots on the same graph, one blue and 1 yellow. It may be that both need to come in one the VR1 and VR2 inputs, but I don't think so. As best I remember it always looks at both input channels, and if there are no edges on a channel, then you won't see anything on the plot.

You don't want to use Dual Tach inputs - that is not your setup and large changes in pulse tol and alpha-beta-gamma is just a crutch. From your last post it seems you should be able to put cam synch back in, set Dual spark option to crank wheel with rising/ falling cam synch and get that to work with no other changes, since the crank wheel works for real and the cam synch works on the Tachref program. But you have to have the cam synch edge arrive somwhere in the middle of the missing tooth. This doesn't have to be exact, you just want it to always come in after the last tooth and before the 1st tooth after the missing tooth. Cam twist/ chain stretch at hi rpm shouldn't be a problem with only 8 teeth.

[24c]

If you look on page 2 of this thread you will see 2 plots on the same graph, one blue and 1 yellow. It may be that both need to come in one the VR1 and VR2 inputs, but I don't think so. As best I remember it always looks at both input channels, and if there are no edges on a channel, then you won't see anything on the plot.

OK I'll read up, but I just had VR1 and VR2 wired upto the MicroSquirt as per normal, and running the 2.4 version of TachRef, I couldn't get the two inputs on the one dialog box...I had to wire the VR2 input into VR1 to "scope" it. I must be doing something wrong

Thanks for the advice, re wheel selection, but it seems I don't need the cam sync, if I mod the crank...here's the crank pulses before

YZF_crank_MSpulses_CW.jpg (26.45 KiB) Viewed 1493 times

[24c]

and after adding metal to the crank slot

YZF_metal_added_8-1_NORMAL_polarity.jpg (26.27 KiB) Viewed 1492 times

so you can see I have a missing tooth now, that'll let me reference a sync to, unless you know otherwise.

PS By altering the Pulse Tolerances, I can now get the ignition and injection outputs to flash the LEDs and start working at just 60rpm on the rig, whereas before it was nearer 100+. I don't know if this is expected, but it seems pretty slow turning to me for an 8-1 wheel to start working.

PPS I will be using the cam tooth sync on the GTS1000, which has a 4 "tooth" wheel (no missing teeth) later, so your comments are appreciated. Thanks very much.

[24c]

I will be using the cam tooth sync on the GTS1000, which has a 4 "tooth" wheel (no missing teeth) later, so your comments are appreciated. Thanks very much.

I swapped the cranks over, loaded up the TachRef program, and for the first test I did get a crank and cam tach plots, but thereafter only the crank one. I adjusted the VR polarity to look like the images above, loaded up a new file back into the MicroSquirt. Created a new project with default settings in the MicroSquirt, added the "Missing wheel" data, 4 teeth, 0 missing, 2 skip, ( 0 delay, but will change this when Microsquirt is on the bike, and as I need to dial in the crank and cam to TDC) and selected the "Rising Cam Sync with Tach or Wheel" option and bingo, once I got to around 120 rpm, I had my injection and ignition ouput LEDs flashing like a Xmas tree.

Thanks again for the advice, next step for me is to get the Dyn-Evo ignition driver module (pairing the inputs to create wasted spark) to work with the Yamaha R6 COPs using the rig as a dummy engine. Unfortunately, I am waiting for the wiring diagram, as I have mislaid mine, but hopefully this won't be too long.