Trigger related page for MembersPage/DTechnologies/Hayabusa project
I need to configure the firmware for my engine.
Basic info
- 4 cylinder, dual overhead cam, 1298cc displacement
- Redline: 10,900rpm
- Firing order: 1-2-4-3 (inner 2 cylinders and outer 2 cylinders move in unison)
- I have a picture showing the order and (very) rough timing taken from the service manual. I've posted it below.
- crank and cam triggers involved - both are variable reluctance sensors. The 2nd LM1815 bought separately (only 1 LM1815 is populated by default) is already mounted on board (?).
Original Manual Timing Diagram
Unfortunately, this is as much info as the manual can provide. It's great at showing the relationship between the sensors, coils, and injectors...but no numbers to speak of.
- Question: If the above diagram shows the cam pulse as being during the exhaust phase of cylinder 1 (180 to 360 degrees of 720), and this data can be trusted, is that enough of a parameter to program the firmware for cam sync?
- no: the current STABLE1_0 branch firmware requires N-m type multitooth wheel or 1 pulse per event ("coil type") trigger signal.
- I just thought of something. If I'm sparking 2 cylinders at once anyhow because I'm using an EDIS style wasted-spark coilpack... why would I even need the cam sensor? Have I been spinning my wheels in regards to cam sync? If I could leave it disconnected, that would simplify things.
- as you say, if 8-1 or 16-1 or similar toothwheel provides cranksync, you don't need camsync, you can just use wasted spark like most of us do
- Are you sure about wasted spark?, Hayabusa have cam sensor and genboard can synchronize it, Hayabusa have Coil on Plug, Are going to change it to a simple wasted spark coilpack???
- I have already installed an MSD Blaster EDIS coil. The 2 outside cylinders will fire together and the 2 inside cylinders will fire together.
- CAM-sync and timed injection(-ending) is not needed to fire injectors once every cam-rotation (which is desirable at 12000 RPM). In fact the common setup is just this: crank-sync only and fire injectors once every cam-rotation. It's possible with either batch or sequential injector arrangement, just a matter of setting up h[0] and config.alternate.
- with 8-1 trigger any reasonable RPM ( > 40000 ) would be possible
- In hayabusa engine you can use the LM1815 for crank sensor, but not for cam sensor , cam sensor is too close to Cyl 1 Coil-on-Plug and it give a lot of problems, with a simple osciloscope you can see what is the problem. Fortunately there are a simple (and very used in races) solution, there are a hall sensor in RS-components (RS-code 235-5706) It is a honeywell hall sensor like ABS one, original Yamaha R1 have also a hall sensor in cam.
- I haven't installed the second LM1815 yet...maybe that will turn out to be a good thing.
- GSXR have a small window (it is a threaded cap in right side of engine) it have a mark to locate TDC, you can remove all sparks and then push start button and adjust TDC with a timing light. - Jose L Cortes -
Crank trigger
- there are 8 equidistant teeth on the crank
- positions: ... I don't know the timing yet
Cam trigger
- single tooth (natch)
- timing: I don't know the timing yet
How do I figure out the timing? - with the least amount of fuss
We usually remove a sparkplug, and move the car in gear 5 until the given engine is at TDC. Than look at the trigger wheels.
There are often marks on the crankwheel (that can be at TDC or other position in custom-hacked setups)
This is something I've been fearing. I've had the engine apart but I can't determine anything that way because the sensors themselves are mounted inside the covers (valve cover & stator cover). At best I'd only be able to guess how many degree's forward or back they are of the tooth/teeth.
How to I set up the firmware appropriately?
The bad news is that current firmware does not support this crankwheel directly, unless engine is a 16cyl.
Options:
- bolt on another wheel, eg. 12-1 (or 36-1 or 60-2). Feasible ?
- 2003 Hayabusa uses a 24-1 wheel. The wheel is also the magnet wheel for the generator...so it's a bit expensive. The search for a used one has been going on for months.
- Aftermarket options are available. Fast by Gast makes a 2 magnet trigger wheel which is driven by starter gears. Requires entire starter housing to be replaced with custom assembly cnc'd out of billet aluminum...even more expensive unfortunately.
- grind off 1 tooth from crank to get a 8-1 multitooth wheel
- Teeth are very wide. This would definately cause balance issues.
- firmware modification: feasible, but significant table-testing needed
- Considering my inexperience, this might not be the best path.
- Adding a magnets to the wheel and using an aftermarket sensor?
A good idea...maybe.
I think I've come up with the least expensive solution yet...I'll have to do some inquiring and see what kind of quote I can get. The original stator looks like this:
Because the 'teeth' are so large, I think I could have each one cut into 2 or 3 at a machine shop. Then have 1 of these teeth ground out and the whole assembly balanced. This would create a 16-1 or 24-1 toothed gear. As long as it's not prohibitively expensive, I think it should be the best option. I'll have to pull the stator out and bring it to some machinist friends and see what they can do...and compare the cost to a new 2003 wheel.
Mustangstef : I currently use this setup, but with DTA engine management. I use 24-2 teeth, cost me about 50 euro at the machine shop. Planning to replace DTA with VEMS
- you could maybe fill up (with magnetic material) one toth to get 8-1
- maybe cut 7 notches to get 16-1
- Just priced out a new 2003 Rotor. $645.68 CAD. Machine shop here I come!
- I found a machine shop that specializes in custom car and bike parts. Nice and small but with some computerized CNC equipment. $95 total to turn the rotor into a 24-1. Please examine a 60-2 and notice that missing tooth is "not as deeply missing as the normal teeth" - neglecting this will cause InputTrigger/RunOut.
- I checked out InputTrigger/RunOut. I think I understand what the problem is but I'm not 100% clear on the solution. I was thinking about how to have the rotor modified (I don't have much time, it's at the shop right now but they're on hold for me) and I've come up with several possible -1 options.
- Leave no tooth gap at all between 2 of the teeth leaving a smooth plateau between 2 of the gaps.
- Grind most of one of the teeth off as the -1.
- Have the rotor machined so that there is a plateau between 2 of the gaps that is about 1/2 way between the lowest and highest points on the tooth wheel.
- I checked out InputTrigger/RunOut. I think I understand what the problem is but I'm not 100% clear on the solution. I was thinking about how to have the rotor modified (I don't have much time, it's at the shop right now but they're on hold for me) and I've come up with several possible -1 options.
- I found a machine shop that specializes in custom car and bike parts. Nice and small but with some computerized CNC equipment. $95 total to turn the rotor into a 24-1. Please examine a 60-2 and notice that missing tooth is "not as deeply missing as the normal teeth" - neglecting this will cause InputTrigger/RunOut.
Thoughts about having the rotor machined
- I just thought of something this morning. I realize I don't know where to have the '-1' tooth cut out of the 24. I suppose the firmware can be calibrated for just about anything I think but this is a one shot deal...
- the 'best' position is so the tooth after the missing tooth is at 65..90 degrees before TDC (of any cyl).
- One thing I noted was that the rotor I have, the notches are rounded but on many tooth wheels the notches/teeth are square. I wonder which is better and if I should make the notches in my own rotor just match the OEM ones or have them all squared off.
- if you are curious about the waveform, you can record it with a PC-scope or a soundcard. I doubt that homemade cosmetics is worthwhile