Crank Trigger Pattern

[fast4motion]

Short Version: Can the Link Atom trigger from a crank/flywheel with a tooth pattern like the attached picture and description? Engine is an oddfire V4, hence the uneven tooth spacing.

 

Long Version: It's a motorbike which was converted from carbs to injection with ITBs many years ago, but has never run well since. The factory crank trigger is used by an ignition controller, which runs wasted spark. Currently a coil signal is used to trigger an unknown brand fuel-only ECU, which runs batch fire injection.

We'd like to fit a Link with sequential injection, and use it to control spark as well, but we're worried about how to trigger it. If the Link can operate from the current crank pattern, I'd look at fitting a cam position sensor to synch it, hopefully picking up on a factory hole at the end of one of the camshafts. I've also heard of using a MAP sensor connected to just one ITB to provide a pulse to synch the ECU when cranking. Is this a possibility if I can't use the cam?

If we can't use the factory crank pattern, the only thing I can think of is to remove the flywheel and have the teeth machined off. Then either drill a series of holes around the flywheel, or somehow attach a trigger wheel. Problem being it's all internal to the crankcase, and incorporates the rotor for the alternator, so would be very difficult compared to a car flywheel/crank pulley. I'd love to hear any other options.

[Adamw]

Thats a bit of an odd one.  If I've got it right you have a 70deg Vee, with 2 shared crank pins that are 180 apart?  So firing angles would be 0, 180, 430, 610?  Weird that we have 3 different firing angles.  

Right now we would only be able to run it with a more generic trigger wheel pattern.  Preferably a missing tooth pattern which would allow the MAP sensor sync, provided it has individual throttles.  

I can talk to the firmware team on Monday to see how much work it would be to support that wheel.  Im assuming that is the factory wheel?  Do you know if it is hall effect or VR sensor?  That long tooth I think is going to be a problem if it is a VR sensor.  

[fast4motion]

2 hours ago, Adamw said:

Thats a bit of an odd one.  If I've got it right you have a 70deg Vee, with 2 shared crank pins that are 180 apart?  So firing angles would be 0, 180, 430, 610?  Weird that we have 3 different firing angles.  

Right now we would only be able to run it with a more generic trigger wheel pattern.  Preferably a missing tooth pattern which would allow the MAP sensor sync, provided it has individual throttles.  

I can talk to the firmware team on Monday to see how much work it would be to support that wheel.  Im assuming that is the factory wheel?  Do you know if it is hall effect or VR sensor?  That long tooth I think is going to be a problem if it is a VR sensor.  

Yes, it's a 70deg V4 and the firing angles you've stated are correct.

It's the factory wheel, which is usually only used by the ignition controller for (wasted) spark timing, in conjunction with a factory vacuum sensor for controlling advance.

I believe it's a 2-wire sensor, so that would suggest it's VR?

In my previous research, I found someone talking about utilising both the rising and falling edges of the teeth to determine which is the long tooth, and using that for sync instead of the usual missing tooth.

I've attached a few pics I found of the "flywheel". The bottom right of the engine pic shows the inside of the crankcase cover, so for orientation, the stator sits inside the flywheel cavity when fitted. Don't be confused by the notches (which look like trigger teeth) above the flywheel in that pic, as it's just an oil baffle plate.

 

[fast4motion]

On 5/26/2023 at 9:50 PM, Adamw said:

Thats a bit of an odd one.  If I've got it right you have a 70deg Vee, with 2 shared crank pins that are 180 apart?  So firing angles would be 0, 180, 430, 610?  Weird that we have 3 different firing angles.  

Right now we would only be able to run it with a more generic trigger wheel pattern.  Preferably a missing tooth pattern which would allow the MAP sensor sync, provided it has individual throttles.  

I can talk to the firmware team on Monday to see how much work it would be to support that wheel.  Im assuming that is the factory wheel?  Do you know if it is hall effect or VR sensor?  That long tooth I think is going to be a problem if it is a VR sensor.  

Hi Adam, did you have a chance to talk to the firmware team?

I was thinking that if the long tooth is a problem, I could have it machined down to the same length as the other teeth and then rebalance the flywheel. But I'm assuming that the main problem is the uneven tooth pattern? 

[Adamw]

Sorry for the delay, I have only just done so now.  

This one is not an easy trigger pattern for us to support, it would require quite a chunk of code development due to the irregular tooth spacing and I dont think it would be justified for the current demand.  

One option if you have enough meat maybe something like below, I have done this on a jetski before, I used M4 hex screws in a 12-1 pattern around the flywheel.  

[fast4motion]

Thanks for the reply. Great idea regarding the hex screws, as that's something I could do myself at work. And the factory sensor could be retained in its original mounting position if we use bolts with heads the same height as the old teeth.

I'm assuming a capscrew would be best because of the uniform outer diameter, and the hex hole in the centre would be small enough to either not be detected by the crank sensor, or any fluctuation caused by it could be ignored via trigger level settings?  I'm worried that if I used normal bolts, they could cause erratic readings depending on differences in head orientation when they're tightened (ie. if the sensor is first presented with the flat or point of a bolt head as it passes). Or am I overthinking it?

I'd be using the MAP signal as sync connected to #1 ITB. Is it best to have the missing tooth 90 degrees from #1 TDC? Thanks.

[Adamw]

I would stick with hex head bolts.  With a reluctor sensor the ecu uses the zero crossing as the trigger point which is actually the center of the tooth mass so the orientation of the hex flats will have very little influence on the zero crossing point.  With a cap screw or Philips etc the hole in the middle will cause a double bump in the waveform.  

For the missing tooth, have it pass the sensor somewhere around 90deg before or after TDC will be good. 

[fast4motion]

Thanks Adamw. The bike owner had given up on the EFI, so tell your boss you're owed the commission on the sale of a G4x Atom, thanks to your input and the technical advice on the Link forums. I'm in the process of removing the flywheel in order to modify it exactly as you suggested. 

We're going to run COP using smart coils, as it's the easiest option given the space constraints on the bike. We're considering either R35 or 1NZ type coils, as I've got friends with them which I've wired up with Link ECUs, so can borrow them to check physical fitment. Do you have any advice or anything we should consider when selecting coils?

[Adamw]

No I cant think of anything relevant really.  Definitely you want COP, I think wasted spark would be a bit of a challenge with these firing angles.  With the R35 ones they have a removable "stalk" and there are several aftermarket/short ones which maybe useful if you need something especially short. The VAG/Audi ones are a decent option as well, available in a couple of different lengths.

 

[fast4motion]

I've modified the flywheel as you suggested, and it's currently at the machinists having the bolt heads skimmed down and then balanced. We also bought some R35 coils, which will be fitted with shorter stalks, so thanks again for the advice.

I'm in the process of rewiring the bike and wiring in the Link. I want to eliminate some of the complicated factory wiring and relays, and wire the sidestand switch and neutral switch directly into one of the Link Digital Inputs, so the Link will stop the engine if the bike is put in gear while the sidestand is down. The logic will be: Gnd at DI input = run; Open circuit at DI input = stop. Before I get too far through wiring it up, could you please confirm that this logic will work? Thanks.

[Adamw]

Yes, that will be fine.  

[fast4motion]

Finally got the flywheel back from being balanced. I installed it and temporarily fitted the crankcase, then rotated the engine by hand to check the teeth don't hit the crank sensor. Unfortunately, because the sensor is internal to the crankcase cover, I can't see or measure the gap to the teeth. I'm hoping the level readings on a triggerscope will show if it's adequate or if the gap needs to be closed up, in which case I plan on slotting the mounting bracket hole for the sensor a little at a time.

Could you please have a look at the attached triggerscope and let me know your thoughts. When I was cranking I could see the MAP signal (ANV1) drop every 720deg on the triggerscope, but don't know how to view it when I open the saved file so haven't touched the MAP offset. I'm also not sure if the the CAS is wired correctly, or if I need to reverse the wires, so I'm hoping you can tell from the triggerscope signal?

Thanks

TriggerScopeLog6.llgx Vmax.pclx

[Adamw]

Yeah it looks like the crank signal is usable.  The ecu missed the gap on the first revolution as the crank done a sudden slow down right near the missing tooth but I think that will probably improve once it is running more often and with a more optimised tune etc.  

I have numbered the teeth on the crank wheel in the pic below.  Sync tooth 1 is the first tooth after the missing tooth.  We see maximum depression on the MAP sensor at sync tooth 10 or 11.  The MAP sensor voltage is normally about 3.94V (~94kPa) and drops to about 1.83V (~48kPa)at sync tooth 10.  

 

So I would initially set the MAP sync up with sync tooth 10 and a MAP offset of -25kPa as you may not always get the ~46kPa depression if you have to adjust the throttle stop for idle or something later.