Trouble Triggering Microsquirt

[R100RT]

Are you sure the opto circuit is working ?

You could test it this way.

Connect opto- to ground.
Connect opto+ to 1K resistor.
Connect other end of 1K resistor to testlead (- black) of an multi meter.
Put multimeter in "mA" and put the testlead (+ red) to 12Vdc (battery)
You should measure around 8mA.

Ther's also an other way but that requiers a led and 2 1k resistors in paralell.
You can use this instead of multi meter and 1K resistor.
Led needs the +Vdc on a certain wire so maybe you have to reconnect it the other way.
It won't damage if it is backward in this circuit.

About the coil.
If you have a coil whit 1 primairy set of wires (+12Vdc and grounding/switched) and 2 plug outputs,
you only need 1 diode and resistor.

A test of the opto circuit sounds like a good idea & I was thinking it might be more productive to pursue the "coil negative" method here and wondered how your connection at the coils was made? (due to them being twin 6v and connected in series). I suspect you are running the bike (or cranking with plugs removed) to test the signal recognition? I'd be yet tempted to leave everything hooked up, with the exception of the bean can which I'd jerry rig a simple hand drill drive for once unbolted from the cam (black plastic tape, wooden dowels onto the offset hub, on a small drill bit come to mind) and in that way try all manner of combinations of connections and settings while having the diagnostics in TS and tooth logger enabled to coax out a signal. I haven't done anything with the simple coil negative triggering so I can't really comment on how she works.
I was going to compare some settings but my old Dell lap top seems to have experience the "black screen" of display this morning and won't come to life as she boots , plugged a second screen in & your settings seem in order.
Hope to see some photos of things as you move past this initial hickup.
Lorne

[Tennesseejed]

Are you sure the opto circuit is working ?

You could test it this way.

Connect opto- to ground.
Connect opto+ to 1K resistor.
Connect other end of 1K resistor to testlead (- black) of an multi meter.
Put multimeter in "mA" and put the testlead (+ red) to 12Vdc (battery)
You should measure around 8mA.

Ther's also an other way but that requiers a led and 2 1k resistors in paralell.
You can use this instead of multi meter and 1K resistor.
Led needs the +Vdc on a certain wire so maybe you have to reconnect it the other way.
It won't damage if it is backward in this circuit.

About the coil.
If you have a coil whit 1 primairy set of wires (+12Vdc and grounding/switched) and 2 plug outputs,
you only need 1 diode and resistor.

I'll do your test to confirm, but I'm pretty sure the OPTO circuit is working because it will trigger with either a square of sine wave from a function generator.

Edit Test resulted in 7mA.

Regarding the coil, I tried again to trigger with OPTO+ to Coil- and OPTO- to Coil+. Also tried the diode pull up resistor circuit. I used one diode (1N4007) w/band toward Coil- followed by a 1K resistor to switched 12V, OPTO+ connects between the diode and the resistor. Am I not doing this right? The schematic in that thread may show switched 12V coming from Coil+. Should that be my switched 12V source?

A test of the opto circuit sounds like a good idea & I was thinking it might be more productive to pursue the "coil negative" method here and wondered how your connection at the coils was made? (due to them being twin 6v and connected in series). I suspect you are running the bike (or cranking with plugs removed) to test the signal recognition? I'd be yet tempted to leave everything hooked up, with the exception of the bean can which I'd jerry rig a simple hand drill drive for once unbolted from the cam (black plastic tape, wooden dowels onto the offset hub, on a small drill bit come to mind) and in that way try all manner of combinations of connections and settings while having the diagnostics in TS and tooth logger enabled to coax out a signal. I haven't done anything with the simple coil negative triggering so I can't really comment on how she works. I was going to compare some settings but my old Dell lap top seems to have experience the "black screen" of display this morning and won't come to life as she boots , plugged a second screen in & your settings seem in order. Hope to see some photos of things as you move past this initial hickup. Lorne

I have abandoned trying to trigger from the Hall. I know it can be done and I think part of the solution involves blocking the 12VDC the Hall square wave is riding on. Other issues may relate to the length of the signal and whether it is symmetric/asymmetric. I really thought the "Trigger Return" setting was designed to address this input condition.

Spent some time trying to trigger from Coil- this morning as explained above. I can clearly see the voltage spikes coming off Coil- on my scope. There is just no reason that this shouldn't be working. I have even soldered all my leads to prevent signal loss/corruption.

Over the last two days I have logged two single triggers from the bike. One was while trying Hall and the other today trying Coil-. I cannot reproduce them am suspect they are just phantom events.

Sorry about your computer. And a teaser on my build (I'll send you pics as soon as its running), I'm using a dual throttle body (not 1100/1150 individual bodies) mounted on the battery bracket. Injectors are mounted at the cylinder intake.

[LAV1000]

Did read youre post, opto in circuit seems to right.

Did read youre set up for trigger from coil, seems to be missing something.
But maybe I did read it wrong.

The way it should be:
Connect opto- to ground.
Resistor from opto+ to 12V feed ignition coil.
Diode from opto+ to the coilside wich is switched to ground (oem ignittion) or like old style points.
Diode band to coilside, away from ms unit.

To get the 12V wire you need to disconnect both wires from coil and measure them.
Coil is low resistance, when measuring on coil to ground, both seem to be 12V.

[Tennesseejed]

LAV1000, I tried again with the diode circuit using Coil+ as my switched 12V. No triggers. With one diode that circuit is a half wave rectifier so at best it would square one half of the AC sine wave. The video below shows the actual Coil- output and I don't think the lack of a DC square wave is my problem. Appreciate the thought though.

Took some video today which may be helpful in understanding the situation. When I was done I realized that I didn't turn my scope to AC so the AC signals coming off Coil- could be compared. They are, however, also very similar.

https://vimeo.com/120988748

And Lorne, something for you at the very, very end.

I'm pretty stumped at what the problem is. Any thoughts or comments appreciated.

[LAV1000]

Did see the video.
When using function generator puls modulation isn't 50%
When using starter motor puls modulation is 50%.

Opto in, diode, resistor.
I believe it works like this.
Resitor is always feeding the opto in (pull-up)
And everey negative wave of the coil shorts the resistor via the diode.

Opto in is most of the time on and pulled down when when the coil does ignite.
Didn't know youre using the bosch module.

[R100RT]

LAV1000, I tried again with the diode circuit using Coil+ as my switched 12V. No triggers. With one diode that circuit is a half wave rectifier so at best it would square one half of the AC sine wave. The video below shows the actual Coil- output and I don't think the lack of a DC square wave is my problem. Appreciate the thought though.

Took some video today which may be helpful in understanding the situation. When I was done I realized that I didn't turn my scope to AC so the AC signals coming off Coil- could be compared. They are, however, also very similar.

https://vimeo.com/120988748

And Lorne, something for you at the very, very end.

I'm pretty stumped at what the problem is. Any thoughts or comments appreciated.

Very diligent efforts there Jed, I think you are probably on the right track with regards too small a signal to capture on the scope. I know on the final reiteration of ignition circuits that I have enabled,& going the opposite direction - the bike tach would not accept any manner of signals from MS to operate properly (and hence the gutted Bosch relay acting as a coil and signal "spiker" with the Bosch module on their own to run the tach). If you can score a touch down in utilizing the stock ignition to trigger MS, that will actually be a helpful stepping stone for the oilhead engines where guys want to do a plug'n play type project and remove the Motronic controller.
I'm "teased" for sure, need to see more.
And I have other lap tops so not that big a deal, I just like that particular old Dell with its serial port. I have one of the EFI Analytic high speed GPS devices to test out this year, will allow amongst many other things HP and torque measurement which I hope will allow icing of the cake on the road tuning approach to things.
Hang in there.

[Tennesseejed]

LAV1000, I built the diode circuit exactly as you described. Maybe the way I described doing it was confusing. Have also verified which side of the coil is switched 12V.

Something about this BMW Hall sensor & Bosch ignition unit just aren't working with MS.

I can't get the OPTO circuit to fire from Coil -. Can't get VR circuit to fire was Coil-. Can get both to trigger with a function generator.

Today I thought I had the answer. The VR circuit will only trigger if the voltage on one side goes negative with respect to the other, e.g., the voltage on VR+ has to go below the voltage on VR- to trigger the circuit. Both VR+ and VR- have a 2.5V reference voltage. I thought maybe, just perhaps, the coil trigger on VR+ wasn't ever getting negative with respect to VR- so it wasn't triggering. I have a variable DC voltage supply and tried several different ways to achieve triggering. On the scope I could see the Coil- signal attached to VR+ going both positive and negative to relation to VR-. Only got one phantom trigger. Then I would put the 5VDC TTL on VR+ and it would trigger. Go figure.

I looked at the video again and it sure does look like the TTL square wave is not 50% duty cycle. But when I put it on the scope unloaded it is certainly at 50%. Good eye though.

Lorne, I'm really stumped at this point. I figured triggering from the bean can would be the easy part of the installation. Boy was I wrong. Do you have any thoughts about putting a VR/Hall on the flywheel? Maybe you considered going that route during your build. I've got to drop my transmission to lube my output shaft after I get my forks on. The machine shop is doing a little triple clamp work to fit some WP 48mm forks from a 2007 KTM 250SX. It's a crazy simple job.

[LAV1000]

LAV1000, I built the diode circuit exactly as you described. Maybe the way I described doing it was confusing. Have also verified which side of the coil is switched 12V.


Try whit 470 or 500 Ohm ?
2x 1kOhm parralel = 500 Ohm

If opto in current becomes 20mA it is still safe.
Otherwise check datasheet on 4N25.

[R100RT]

LAV1000, I built the diode circuit exactly as you described. Maybe the way I described doing it was confusing. Have also verified which side of the coil is switched 12V.

Something about this BMW Hall sensor & Bosch ignition unit just aren't working with MS.

I can't get the OPTO circuit to fire from Coil -. Can't get VR circuit to fire was Coil-. Can get both to trigger with a function generator.

Today I thought I had the answer. The VR circuit will only trigger if the voltage on one side goes negative with respect to the other, e.g., the voltage on VR+ has to go below the voltage on VR- to trigger the circuit. Both VR+ and VR- have a 2.5V reference voltage. I thought maybe, just perhaps, the coil trigger on VR+ wasn't ever getting negative with respect to VR- so it wasn't triggering. I have a variable DC voltage supply and tried several different ways to achieve triggering. On the scope I could see the Coil- signal attached to VR+ going both positive and negative to relation to VR-. Only got one phantom trigger. Then I would put the 5VDC TTL on VR+ and it would trigger. Go figure.

I looked at the video again and it sure does look like the TTL square wave is not 50% duty cycle. But when I put it on the scope unloaded it is certainly at 50%. Good eye though.

Lorne, I'm really stumped at this point. I figured triggering from the bean can would be the easy part of the installation. Boy was I wrong. Do you have any thoughts about putting a VR/Hall on the flywheel? Maybe you considered going that route during your build. I've got to drop my transmission to lube my output shaft after I get my forks on. The machine shop is doing a little triple clamp work to fit some WP 48mm forks from a 2007 KTM 250SX. It's a crazy simple job.

Jed, if you are unable to suss out things with LAV1000's help for the stock Bean Can approach, I would be temped to go for a tone wheel on the end of the crank rotor. I don't know how Les Garten mounted his little 12-1 wheel that "old guy" made for him, but I'm sure we could go to school on that technology and you'd be up and running with a superior signal in no time at all. Equally, this would put things better in line for when you approach that "direct injection" part of the project

[LAV1000]

+ 1 on R100RT
If you want to use the full potential of this system get a toothed wheel on the crank somewhere !!
And use the hall cam sensor to run it sequential.
Stil trying to help you out on hall or ignition trigger option.

[old guy]

I basically took the Silent Hektic rotor and made it a 12-1 wheel. I assume the Silent Hektic mounted the same as the stock unit, but I have no idea where that is.
I don't know how hard it would be to put a wheel on the crank on that engine, but If it was mine I would definitely consider it.
If you decide to take this route, I do custom machine work geared toward motorcycle stuff at a reasonable rate.

[Tennesseejed]

LAV1000, I tried 1K, 500Ohm and 270Ohm all without success. Can clearly see the signal on my scope when using the diode pull up circuit to try and trigger from Coil-. My ignition system still fires and the 5VDC square wave from the function generator will still trigger both the OPTO and VR inputs.

As I mentioned earlier, I also tried floating some DC on various leads in conjunction with the input signal trying to ensure the input signal crosses, for example, the bias voltage of the VR leads.

I tried a bunch of different setting in MS/Ignition Settings such as rising edge, falling edge, trigger return, basic trigger, changing trigger offset angle, etc.

It just doesn't work. Which is impossible because electricity follows the laws of physics. I know this can work, but its been a week and there isn't any combination of wiring or signal inputs I can think of that I haven't tried.

I'm going to head down the 'add a new trigger sensor' rabbit hole. I wanted to do fuel only this Winter and transition to fuel/ignition next Winter when I had the engine out for an overhaul. Guess I'm moving things up a bit.

Lorne, I pulled the nose cover off the engine and it looks like you can barely get a ham sandwich in there with the bean can and alternator. This is why I was thinking a flywheel sensor might be easier. I've seen magnets fitted into the flywheel using a VR sensor. Maybe its possible to mount a 36-1 on the inside (toward engine) of the flywheel as well (?). I need to get my forks on before I can remove the transmission. Should be later this week.

Didn't you do an external alternator and use a crank mounted 36-1? I don't think that's an option for me. Do you have any thoughts about a trigger wheel on the front end? Old Guys mentions machining a 4-1 into a 12-1. Did Les ever get his bike running? Are there picture on this site or somewhere else I could look at (I'll search too)?

Edit Just found the Silent Hektik unit: http://www.silent-hektik.com/BMW_Z1.htm Looks small, compact and attaches to the alternator (crank). Site is in German, but I gather that is a Hall sensor. What would prevent a guy from grinding off all but one of the teeth and calling it a day?

[R100RT]

LAV1000, I tried 1K, 500Ohm and 270Ohm all without success. Can clearly see the signal on my scope when using the diode pull up circuit to try and trigger from Coil-. My ignition system still fires and the 5VDC square wave from the function generator will still trigger both the OPTO and VR inputs.

As I mentioned earlier, I also tried floating some DC on various leads in conjunction with the input signal trying to ensure the input signal crosses, for example, the bias voltage of the VR leads.

I tried a bunch of different setting in MS/Ignition Settings such as rising edge, falling edge, trigger return, basic trigger, changing trigger offset angle, etc.

It just doesn't work. Which is impossible because electricity follows the laws of physics. I know this can work, but its been a week and there isn't any combination of wiring or signal inputs I can think of that I haven't tried.

I'm going to head down the 'add a new trigger sensor' rabbit hole. I wanted to do fuel only this Winter and transition to fuel/ignition next Winter when I had the engine out for an overhaul. Guess I'm moving things up a bit.

Lorne, I pulled the nose cover off the engine and it looks like you can barely get a ham sandwich in there with the bean can and alternator. This is why I was thinking a flywheel sensor might be easier. I've seen magnets fitted into the flywheel using a VR sensor. Maybe its possible to mount a 36-1 on the inside (toward engine) of the flywheel as well (?). I need to get my forks on before I can remove the transmission. Should be later this week.

Didn't you do an external alternator and use a crank mounted 36-1? I don't think that's an option for me. Do you have any thoughts about a trigger wheel on the front end? Old Guys mentions machining a 4-1 into a 12-1. Did Les ever get his bike running? Are there picture on this site or somewhere else I could look at (I'll search too)?

Edit Just found the Silent Hektik unit: http://www.silent-hektik.com/BMW_Z1.htm Looks small, compact and attaches to the alternator (crank). Site is in German, but I gather that is a Hall sensor. What would prevent a guy from grinding off all but one of the teeth and calling it a day?

LiMa_3b.jpg

Hi Jed, added a view of the S.H. rotor in the above view (partly for "old guy" to see how his handy work mounts to the alternator rotor held onto the tapered end of the crankshaft). I suspect Les had the five teeth cut off and the new pattern of 12/ one missing teeth machined into the remaining greater diameter and the Hall sensor they supplied slightly moved inward to line up. I am just discussing this with Les over on the ADV private message zone and he may be able to supply a photo of that finished item which I'll post here. I tried to count the teeth on one of our ring gears and its either 110 or 111 (got called to dinner and tend to loose count about 2/3 way around the circle ) - both of which I believe is not helpful in MS trigger land. There isn't much room to deviate either forward or rearward from the ring gear if placing the sensor in the timing hole was the plan. And I believe the strong advise goes against trying to run the system with a single tooth - just don't get much resolution for our controllers liking. I guess your at the point of throwing the towel in on the stock bean can or coil negative triggering - not sure anyone else can pipe up with some knowledge on how to coax the existing signals to the level that MS requires to recognize (you might try offering another post on "Tuning" uphill from here on the posts).
From what I can see and understand, the Silent Hektik rotor and spacing works with a stock front cover, so I'd be putting my money on going with that location, and mounting, but modify like Les for a traditional pattern like 12-1. It takes more diameter to get 36-1 and I don't believe it would be that big a difference. Then, as LAV1000 mentioned, you'd be in the market to capture the bean can signal off the cam (just blank one window out for a single blip per cam rotation) to arrive at the gates of "Sequential Ignition" ,for when you want to play with that. I know some have had lesser experiences with VR sensors but they are the most robust, cheap and easy to locate anywhere (including the junk yard) and have work in stellar fashion in my experience.
You've still got a few months of snow up there don't you so there's time to add this minor change to the project
Lorne

[R100RT]

Here are some of the images from Les Garten's triggering set up - the S.H. rotor modified by "old guy" into a 12-1 wheel. This would require a longer rotor retaining allen bolt, but easily doable. And I believe that all fits under the front cover without mods. I suspect the outer tone ring is sweated/ interferrence fit wise onto the original mount adaptor.
Rather than purchase a rotor and modify it, perhaps O.G. can build one from scratch for you?

!cid_6B9A56A4-6D1B-43C0-B460-1BCCB656ED87.jpg

!cid_F1E73C42-1C2F-462C-9878-CC640A26BB04.jpg

!cid_982EE081-E2C2-499A-A3AD-E2EC98022EA2.jpg

I did remove everything off the front of my crankcase, and add the unusual 900 watt belt drive alternator - the drive shive is on a modified rotor and also contains the 36-1 chopper wheel I got from DIYA. Perhaps a bit overkill, but I can perform welding now (just kidding).
Cheers,
Lorne

[Tennesseejed]

Lorne, thanks for taking the time to put up pics of Les' 12-1 wheel. Looks very promising.

Old Guy said that he was able to get good signals on his bench. I guess the $64 Million dollar question is whether Les was able to get his wheel to trigger Mega/Microsquirt? Any chance you know the answer to that? If so, I'll PM Old Guy and get a quote for services.

Now that you mention it, I recall people having little or no success with single tooth triggers. So that is probably a bad idea on my part.

Been a real mild Winter in Alaska this year. NOAA forecaster says high likelihood of late snow dumps in the Spring, but I'm thinking another 4-6 weeks before people are riding. I want to be one of those people so I got to keep this project full steam ahead.

[LAV1000]

@R100RT.
Seems like nice 12-1 wheel.
Why has that "S.H" wheel such a strange tooth patern ?
What type of sensor is used, just curious.

[jsmcortina]

Why has that "S.H" wheel such a strange tooth patern ?

From the look if it, It's a 4+1 wheel, same idea as some 4cyl Daihatsu.

James

[old guy]

Just to clear things up. The unit was tested on a test stand with a variable speed electric motor. I can turn the rotor over 10 grand and this setup had a perfect signal. But you must consider that it was driven by an electric motor so no variances in rotor speed as seen on a lot of motorcycle engines. Also there was no other electrical component in the circuit to cause electrical noise.
It was actually tested on a Microsquirt VII and the unit uses a hall sensor. As a note the Micosquirt was modified with the hall off setter circuit that I think was posted by Bruce Bowling on the Microsquirt forum.
I have read that this circuit is not optimum for the hall sensor, so I am thinking of replacing it with the later VR conditioner circuit.
The outer ring on the Silent Hektic is a shrink fit on the inner rotor which is made of brass.
It's been a long time, but I think I charged Les $75.00 for the work.
The thing is that it still mounts to the camshaft, but I think Les was looking for the easiest solution and retaining the stock appearance.

[R100RT]

Lorne, thanks for taking the time to put up pics of Les' 12-1 wheel. Looks very promising.

Old Guy said that he was able to get good signals on his bench. I guess the $64 Million dollar question is whether Les was able to get his wheel to trigger Mega/Microsquirt? Any chance you know the answer to that? If so, I'll PM Old Guy and get a quote for services.

Now that you mention it, I recall people having little or no success with single tooth triggers. So that is probably a bad idea on my part.

Been a real mild Winter in Alaska this year. NOAA forecaster says high likelihood of late snow dumps in the Spring, but I'm thinking another 4-6 weeks before people are riding. I want to be one of those people so I got to keep this project full steam ahead.

Jed, my understanding is that Les hasn't completed the job on the bike yet, but the outcome looks undeniable towards success I'd say.
I'm going to combine this post and answer a few of the questions that "old guy" raised as well.
I don't have a useful photo of the stock view of the front of our engines (my images don't count as the modifications are beyond casual), but picture the boxer case being long top to bottom, and contains the crankshaft in the middle and the cam shaft directly below. The alternator rotor pictured above with the S.H. 4-1 wheel, is off the "crank shaft" and that is where Les has mounted the 12-1 modified wheel. The cam shaft below contains the aforementioned "bean can" which is an ignition device with a twin window rotor triggering a hall sensor, and then Bosch Ignition module for the coils. So for both Les, and you Jed, if you add the tone wheel to the crankshaft and leave the bean can on, you would be orientated towards having sequential ignition as the next step easily I believe.

Keep in mind for the manufacture of a tone wheel, that you might be able to notch things up even further, to a 24 tooth wheel perhaps. Our engines are loopy on cranking but seem reasonably calm for my 36-1 wheel that is crank mounted. I have my 900w alternator just next to the VR sensor and wheel and have no noise/ interferrence issues. I'd recomend sound wiring practises, with wire twisting and foil drain wrap over that grounded at one end for insurance. If you were to check the space available behind your cover (some kind of spray foam in a bag that takes the shape of the space while cover is on - and study that) there may be enough to go larger diameter then, and more teeth. If not, the 12-1 pattern would be entirely suitable I'm sure.
I would tend to draw it up and manufacture from one disc - the S.H. unit had the brass inner to work with and you wouldn't have that. Probably easy enough to take measurements from the depression on the end of the rotor and draw it up from there. I don't have a usable rotor any more so can't help you on that - but should be a small job. Some of this is repeat, but I believe this will be an easy add on for your project.
Cheers,
Lorne

[Tennesseejed]

One option, the easy option, would be to purchase the Silent Hektik unit and have Old Guy machine a 12-1 ring. Or a 24-1 ring is that is within his superpowers.

Another option would to be to buy a 4" 36-1 wheel (http://www.diyautotune.com/catalog/361- ... p-593.html) and either (1) have it cut and pressed onto the S.H. inner ring or (2) attach it onto the end of the alternative center shaft bolt (which obviously would need to be replaced). With a 4" 36-1 wheel I would probably use a Hall sensor mounted on the edge of the wheel rather than the S.H. Hall which is mounted to the side. For some reason I'm skeptical that the S.H. Hall would trigger properly using a 36-1 ring. The 4" wheel would extend to the edge of the alternator casing but there is plenty of room to the sides to locate a Hall sensor.

Good idea about using foam to check clearance. It looks to be pretty tight between the alternator and nose cone.

Took a pic of the alternator:



And one from the side to show how the brushes extend beyond the alternator casing:



I'll sleep on it, but I think the right thing to do if I want to be on the roads early is just buy the S.H. unit.