Jump to content

Trigger Wheel Guide


Recommended Posts

So i thought it was about time we got some hard data down on components of an EFI system. The biggest problem seems to be trigger wheels.

I am just going to present what i did with as much detail as possible to try and give others an easier time of it. 

This is all based around my experience doing an install of the older (no longer available) Microsquirt V2 which had inbuilt high current drivers NOT the currently

available Microsquirt V3, but all we are looking at here is the actual wheel and connections so shouldn't be too dissimilar and please, anyone else with experience of 

working 36-1 wheels on V3's feel free to add your findings to this as well.

Microsquirt V2 page2_blog_entry13_3_zpsjj2tzp1x.jpg

Microsquirt V3 download_zpsdhhwrfn2.jpg

so you can source the 51mm trigger wheel needed from here. http://trigger-wheels.com/store/contents/en-uk/p7.html

The suzuki wheel is 50mm so you will need to reduce the diameter slightly but that comes later.

So before we take the oem wheel and 36-1 wheel and fit them together we need to make sure we can keep alighnment of where we want TDC to be on the finnished wheel.

There are 2 ways to do this:

 The simplest way is to scribe the TDC mark on the OEM wheel in towards the centre of the wheel as far as possible.

Then on the 36-1 wheel, with the missing tooth at 12 O'clock, count back anticlock to tooth 9, scribe a lign from the leading edge of this tooth to the centre of the wheel.

Then when the two machined sections are pushed together, simply keep the marks aligned as near as possble. This will then be your new TDC mark.

 

Or, the way that i did it was to make a jig that the oem wheel fitted on with a marker @ TDC. Then after i pressed the 2 pieces together, as near as possible, i then put it back on the jig and marked on the TDC on the new wheel.pictures below.

20160619_130844_zpsi5jty2ug.jpg

20160619_130859_zpsz2r0fdqc.jpg

20160619_131002_zpsgddcohb4.jpg

P1000851800x600.jpg

So, once you've worked this out get the oem wheel machined down as close to the center section that goes in the crank as you feel practical, then machine out the center of the 36-1 wheel to make it an interference fit onto the oem center, then push together to form your new wheel while using your prefered alignment method from above. Mine was a good tight fit so i didnt wled it or anything. No problems so far with it but up to you.

 

So, if you offer up the wheel to the engine now it wont quite fit, as its slightly to large. So at this point i started turning the teeth down and rechecking etc until i just had clearance to the VR sensor. This also does the job of finally truing up the wheel.

 I have 0.15mm clearance between the wheel and the sensor

It should look like this

20160619_130416_zpsd8gu0orn.jpg

20160619_130618_zps3c6dtekm.jpg

Ok wiring,heres a picture of the OEM Vr sensor i am using from my origional 1988 GSXR750 slingshot. So this VR sensor has been on the 750 with a low comp 813cc kit in it, a 1052cc motor with 1mm decom plate and a stcok B12 engine, all have different cranking speeds and patterns but it has worked fine on them all.

20160619_130407_zpsxjzzaqb5.jpg

So, different Vr sensors may have different colours etc, im not sure so i thought i would do this so you can see what goes where and make adjustments to suit your wiring to match.

I have yellow nearest the engine and blue on the outside. This relates to the my Microsquirt V2 as follows.

Yellow = VRIN1+ (centre of coax)

Blue = VRIN- (screen of coax)

 

I beleive the V3 dosnt use coax but a shielded 2 wire set up but the wire names are the same and have the same inputs.

I also used a mini timer connector (like male and female injector plugs) for this connection as i felt they offered a good weatherproof connection.

20160619_131143_zpseeo57jby.jpg

Setting in Tunerstudio are as below and using ms2extra code 3.3.1

Ignition%20settings%20B12_zpsgf2qpajl.pn

Now, if you power up your microsquirt and crank the engine, even with nothing else connected to the ecu you should be able to open the Diagnostic and  High Speed Loggers in Tunerstudio and see the wheel being logged and it should look similar to below.

B12%20composite%20logger_zps0buvavgt.png

B12%20tooth%20logger_zpshqha9smp.png

It looks very messy in the tooth logger but in the composite log, you can clearly see the wide gap of the missing tooth every 35 bars.

But heres an example of the Tooth Logger while the engine is idling. Very clean stable signal compared to above.

B12%20tooth%20logger%20idling_zpsgwh7vl6

 In doing this i have found some documentation that the VRIN circuitry was changed between the V2 that i have and the V3 that most people are currently working with. My V2 basically grounds the sheild of the caox (VRIN- is permantly grounded essentially) and feeds the AC sign wave into VRIN1+ where as the V3 has a propper 2 wire scheme hence the change from a caox to a shielded 2 wire cable in the loom. But this means i think any issues with the VR sensors or wiring leaking to ground causes more problems. See TXT below

Second, on the previous MicroSquirt® controller's VR sensor interface, the circuit was ground-referenced, meaning that the return path from the VR sensor is at 0 volts potential (or ground level). With the new interface circuit on MicroSquirt® V3, the VR- input has a weak pull-up voltage applied at 2.5 volts thru an internal resistor connected to the differential amp on the MAX992x (see the figure above). Also, the MAX992x comparator (+) input has 2.5 volts as well - this voltage will automatically adjust depending on the current adaptive mode but is centered around 2.5V. So, it is important to connect both VR sensor wires to the VR+ and VR- input. Also, make sure that there is no internal connection of the VR sensor to ground - use a multimeter in resistance measurement mode to make sure. The majority of VR sensor signals are floating and do not connect to ground.

I wonder if the oem VR sensors are not well insulated to ground or something which is causing the problems that some are having with the V3's?

Anyway, thats about it from me, hope this little lot helps someone out.

 

 

Edited by slingy1157
LINKS NOT WORKING
Link to comment
5 hours ago, luke2152 said:

Is there any reason not to use the stock trigger wheel with the big tooth ground off to make a 4-1 wheel.  Of course it has less resolution than a 36 or 24 wheel but if 4 teeth is good enough for OEM surely it must be reasonably workable.

It can work as that's the set up I used on mine although I used a Motec and it has a cam sensor to sync the timing. I think without the ign won't know where it is and what pair of plugs need to be fired. With the Cam sync the ecu knows where #1 is and will fire the coil on the next crank tooth it sees. The Motec is pretty well adjustable for any triggering set-up considering its age - its operating system is DOS based so its not a new kid! Suzuki obviously use the wide tooth as a precursor to actual trigger point for a particular coil knowing that a short tooth 180 degrees later will be the other coil. With the cam sync I can have full sequential fuel and sparks.

Link to comment
7 hours ago, luke2152 said:

Is there any reason not to use the stock trigger wheel with the big tooth ground off to make a 4-1 wheel.  Of course it has less resolution than a 36 or 24 wheel but if 4 teeth is good enough for OEM surely it must be reasonably workable.

I think it is because, the teeth create the sign wave as they pass the sensor but the gaps between create the pattern. So if you remove the large tooth you end up with 3 small teeth spaced 90 deg apart and 3 lage gaps, 2 the same size, roughly 90 degrees and one twice the size at 180 degrees.  I'm sure it would run ok like this but the issue will be sarting to get it running!

Because the cranking speed varies durring each revolution so much, (Just look at the varying line widths in my composite log above) each gap will take a different amount of time to pass the sensor resulting in 3 different sign wave events with no clear "This is the gap"moment. I think the oem large tooth leading edge would be used for cranking timing to eliminate this issue.

So you could do it, or you could even grind the large tooth down to a matching small tooth to make a 4 tooth wheel, keeping the TDC mark on that tooth but either way it would need a cam sensor to create a crank reference point for starting which you can do with microsquirt as well, though it wont do sequential still. But it seems a PITA to fit a cam sensor compared to making a 24-1 or 36-1 wheel. Could be worth a crack by someone though. 

Anyone tried a GSX1400 Trigger wheel? Do they fit an oil cooled crank?

Edited by slingy1157
Trying to type on a phone with sausages
Link to comment

Cam sync is only needed for sequential ignition.  For wasted spark it doesn't need to know where the cam is.  My mazda runs 4-1 megasquirt and has no problems starting or otherwise and times the spark as accurately as I can see with my timing light.  Granted a bike engine cranks differently and accelerates through the revs much faster than a car one but I would have thought it would still work.  I reckon I'll give it a go anyway.

Link to comment
5 hours ago, luke2152 said:

Cam sync is only needed for sequential ignition.  For wasted spark it doesn't need to know where the cam is.  My mazda runs 4-1 megasquirt and has no problems starting or otherwise and times the spark as accurately as I can see with my timing light.  Granted a bike engine cranks differently and accelerates through the revs much faster than a car one but I would have thought it would still work.  I reckon I'll give it a go anyway.

Cam sync is a necessity for any sequential running but can still be used for a batch fire wasted spark setup for better reference. I still aggree that its a complication not needed for wasted spark.

Car engines generally crank alot more evenly on the starter and so generate more even tooth signals durring cranking but by all means please try it, and post your findings back here. There is alot more experimentation and variations to come from efi 

Link to comment
23 hours ago, luke2152 said:

How exactly did you do the cam sync?  Glue a magnet to a cam and drill hole for sensor in the cam cover?  I'm looking at doing exactly this (4-1 wheel + cam sync) when I get my efi up and running

Just read your post closer - you can't have a 4-1 wheel! The smallest number of teeth must be equally divisible by cylinder number! Hence I used a stock wheel and reduced the big tooth to a small tooth to get a 4-0 wheel which had to have the cam sync even running wasted spark.

The cam pick-up was easy - drill the cam for a M6 screw, drill hole in cam cover over the screw and add a plate holding a Hall effect sensor. As long as the screw is within 1mm of sensor it works fine.

I may get round to re-doing my crank wheel - maybe not 36-1 but 24-1 or even 12-1 if only to improve starting response which suffers with only 4 teeth!

Link to comment

From diyautotune website (the makers of microsquirt which I will be using)

"2) Trigger Wheel on the crank: Since the crank runs at full engine speed, which is twice the cam/distributor speed, you only need half as many trigger positions on a crank mounted wheel (evenly spaced of course). This is also usually a toothed wheel with a VR sensor and is mounted on the crank pulley up front, or the flywheel on the back of the engine. Note it can also be a ‘flying magnet’ Hall Sensor setup. The results are the same. 2 trigger points for a 4 cylinder, 3 for a 6cyl, 4 for an 8cyl, you get the picture. "

No idea if it will work well but it works in theory

Link to comment

Just don't see how a 2 tooth wheel would work on a 4 cyl! You'd get 2 spark trigger events every crank rotation and with two, dual ended coils every cylinder would get a spark but couldn't the spark occur in the wrong cylinder at the wrong time? It may work with a single dual ended coil fired twice as often but still not convinced!

Link to comment
On 6/24/2016 at 7:48 AM, luke2152 said:

From diyautotune website (the makers of microsquirt which I will be using)

"2) Trigger Wheel on the crank: Since the crank runs at full engine speed, which is twice the cam/distributor speed, you only need half as many trigger positions on a crank mounted wheel (evenly spaced of course). This is also usually a toothed wheel with a VR sensor and is mounted on the crank pulley up front, or the flywheel on the back of the engine. Note it can also be a ‘flying magnet’ Hall Sensor setup. The results are the same. 2 trigger points for a 4 cylinder, 3 for a 6cyl, 4 for an 8cyl, you get the picture. "

No idea if it will work well but it works in theory

Yes. Found this txt so i could see it in context. And it relates to a single coil distributor based system. As such you dont need a reference for the each spark event such as a missing tooth, just how many spark events required in the 720 deg of an engine cycle. The single coil then sparks this amount of times and sends the spark to the distributor (which obviously is referenced normally off the cam) to send to the correct cylinder.  

So in our application a 2 tooth wheel wont work.

The minimum in theory would be a 4-1. But i think there will still be starting issues

But to crisp up starting timing the backing plate could be slotted to allow the vr sensor to be rotated to put about 8 deg permanent  advance into the system as cranking advance . Each coil could then be triggered on seeing the tooth rather than trying to calculate the advance off a very large fluctuating timing gap in between teeth. Hopefully the missing tooth would be obvious enough for the ecu to work out the 2 tach events. As soon as it catches and fires it should run ok in theory i suppose. I have a spare oem wheel. i might check it out in the interest of science lol

Edited by slingy1157
typo's
Link to comment

Well.............................Just tried it, hacksawed off the long tooth, attacked the slot on the back of the wheel with a grinder to put some static advance on it for cranking (made this too big, reckon i had about 15+ deg so had to take out 5 in TS. Slotting the backing plate would deffo be alot better and more controlable but just wanted to try it so did this dodge).

Through some settings at it in Tunerstudio and took some data logs and photo's for you guys. GOD I am good to you lot lol

 

Link to comment

Ok, so here is my findings on the 4-1 modified OEM wheel.

Like i said earlier, just hack sawed the long tooth off and then used a grinder to widen the crank pin slot to allow me to quickly turn the wheel in relation to the crank in order to put some fixed timing in for cranking timing so that the ECU isn't trying to calculate the cranking timing off the large gaps in between the teeth which are turning at varying speeds so making the calculations inaccurate.

I fitted the wheel, changed the settings in TS and tried cranking the engine over with the injector fuses pulled so it would not run. Then i tried doing a high speed composite log of the wheel, and got nothing! Then i tried doing a tooth log and got a good trace :) . So then i closed the logger down and opened the dashboard screen and tried cranking again to see the tacho register a steady 270-280 rpm. So put the fuses back in and tried to start it . It coughed and fired a bit but would not start. But it back fired on the starter a fare bit (because i had ground the slot out too big on the back on the wheel) so i put -5 deg in on the cranking timing in TS and this was a lot better and after a couple of coaxes it started up and died. Then it started a bit easier and kept running. Then it was starting really easy, virtually straight away. So i logged these and it running which i will show below.

When running it ran fine, really well actually, and as it is revved up the tacho was getting to the max advance value then holding perfectly steady at all revs above so telling me the ecu has no problems working out the timing while running from this setup. The settings i used were a bit of a guess, the amount of cranking timing was a complete guess and i dont really know what it was. Somewhere between TDC and probably 15*BTDC. So there is alot of scope for fine tuning this with a timing light. If the VR sensor backplate holes were slotted so the cranking advance could be dialed by moving the VR sensor in relation to the wheel then this would be ideal i feel.

So heres the pictures of the wheel and logs

 20160625_091018_zps5krittig.jpg

20160625_091118_zpsyhz8wity.jpg

20160626_091116_zpsizx1pinz.jpg

Heres the tooth log of the 4-1 wheel cranking.

B12%204-1%20OEM%20Trigger%20Wheel%20Toot

Heres a log of it starting. This was one where it just started straight up as you can see by how quickly the graph drops to running speed.

B12%204-1%20OEM%20Trigger%20Wheel%20Toot

Then this is just a clearer log of it idling

B12%204-1%20OEM%20Trigger%20Wheel%20Toot

 

So then i thought i would be really good and actually film the difference in starting between the 2 crank wheels with abit of a description. So here you go. But bugger me if it wasnt alot harder to start for the camera than it had been the previous day. But it was alot cooler, and you can here it back firing against the starter still a couple of times. So like i say, these settings were very rough so i think alot of improvement could had here, and be honest i dont think its that bad even like it is for the ease of making the trigger wheel lol. Have you seen videos of Guy Martin trying to start his EFI turbo oil boiler? Holy F*#K, hes got a good battery on that thing as hes cranking for EVER. And thats suppose to be a pro built machine. BUT  the 36-1 wheel just works brilliant in comparison. ENJOY :) .

 

  • Like 1
Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...