Autotune creating weird values.

[classic-indy]

So we finally got enough snow in the area to ride the snowmobile and start tuning. This morning I warmed the sled up to above the minimum CLT temp for the autotune function to start correcting the table. I was getting very good results for 10,000+ data entry points. My corrected table from idle to around 6300 rpm seems great. The VE table in the regions below 6300 rpm are a nice blend of % ranging from high 60%-90%.
The problem area arises around the exhaust valves opening. My 2 stroke motor has exhaust valves which open at the higher rpms to allow higher port timing. This gives the engine better low end torque with the lower exhaust port timing and then raises port timing for WOT runs.

The VE table is in excess of 150% at points. The engine started running very rough at the higher rpms even dieing out.

My engine is a 440CC 2 stroke twin rated around 95hp. I'm running 50lb/hr low Z injectors (running 60% duty cycle) and my fuel pump is a snowmobile specific pump which is off a sled with almost double the hp. My alternator is 280W and the battery voltage is stable, and the auxilaries .

Possible problems?

Injectors dirty? I'm running PWM so I don't think I've damaged the injectors. How fragile are Low-Z injectors? Could there be something I'm over looking?

[dontz125]

Do you have a MAP sensor installed? Do you have 'Multiply MAP' turned on? At max torque, the resonant stuffing pulses in a well-tuned pipe can produce overcharging equivalent to 7-8psig of inlet boost, so 150% VE isn't unreasonable.

Are your PVs on/off guillotines, or modulating spools? The dead-band for on-off transition can really make tuning in that rev-range difficult if you aren't careful, especially if the timing is incorrect. If the PVs open at 5500 (increasing) and close at 5200 (decreasing), it is entirely possible to have two very different tuning cases at 5300 rpm, depending on whether or not the PV is open or closed. I remember riding on the highway on a Suzuki RG250 right in the turn-over band. The bike would creep up to the 'open' set point, the PVs (slotted barrels) would go to the 'high speed' position, but - because the PVs were mis-timed - the engine would LOSE power, slowing the bike. Engine speed would drop to the 'close' set-point, the valves would flip to the low-speed position, power would increase - and the engine would speed up to the 'open' set point ...

[classic-indy]

The PV's are guillotines which are spring accutated. There is a hole from the exhaust port up to a bellow which when the exhaust pressure is great enough the guillotines open. I'm not sure if they open & close at the same rpm. I do have a MAP sensor, but I don't think I've got multiply MAP active. I can't find it in my tunerstudio settings to be honest.

From my riding notes I think they don't, and neither does my clutch. I started getting a stumble @ clutch engagement, and PV opening.

Should I tune with duty cycles for clutch engagement? What other options are available?

[dontz125]

Yeah, pressure-driven PVs are great for WFO pulls, but they absolutely suck dead goats for varying throttle modulation. There's a reason all road bikes used either electronic control or (early days) centrifugal actuators. The entire point behind the exercise of fitting a port-height adjustment device is to change the pipe length to engine speed relationship. Changing the port height for a reason OTHER than engine speed obviates the entire purpose of a PV. If possible, you may want to consider something like a boost-control solenoid (operated on/off, not PWM) to force the PVs open or shut according to revs, not pipe pressure. Plug the internal passage, drill and tap for an external pipe?

I don't know enough about sleds, but I'm pretty sure you damn well want to tune to lockup that clutch!

[classic-indy]

attached to my Stock computer are solenoid controls which allows the PV to open close @ a specific rpm. too them off because it makes the engine bay dirty. Guess I should put those back on! I could tune the clutch issue for accel. not decel. since I will almost never be running that slow.

[dontz125]

Oh, yeah - put them back on! You can control them with the uS using one of the programmable output pins. The WLED and ALED drivers have the same capability as the fuel pump relay driver, and can handle the solenoids easily. I'd say wire both solenoids to one pin (why waste an output?), but add 1N4005-ish flyback diodes across the solenoids themselves - the flyback pulse will be far more severe than the actual opening load.

[classic-indy]

The PV's will still be spring loaded, but the solenoids control an air bypass which stops the build up of pressure. Once the solenoids close the air pressure builds in turn opening the valves. Arctic cat i believe has direct solenoid control of the PVs. might look into getting that to work. Need to figure out a solution soon. as the season isn't very long.....

thanks, for the help I'll keep you posted.

[classic-indy]

Just did a couple hours of work on the project.

I've made it so that 1 solenoid will control both PV's. I'm going to wire it up to WLED. The solenoid is rated for 12V. My plan is to get power from the ignition switch fuse it with 4 amps and connect the second wire to WLED.

I will activate the solenoid for the delineation between two bins in the VE x RPM table. That way I will eliminate any overlap. Along the X axis (rpm) what does the value defined for each bin equate too, is that the lower limit/middle/higher limit value for that bin?

So I've gone into OUTPUT PORT SETTINGS, and I've configured PM5 - WarmupLED as follows.

Enabled with a power on value of 1, and trigger value of 0. I then set the conditions as output channel to rpm < 6920(threshold) and 0.0(hysteresis). If my settings are correct the solenoid will be powered under 6920 rpm.

Ideally I want my PV's to open around 6700rpm. So I'm thinking of changing the rpm bin such that it changes bins when the PV's are activated. thoughts?

Also hysteresis, I've looked up the definition but I'm unsure of what a value of 1 means. is it +/- rpm meaning if I set the hysteresis to 20 with a set value of 6920rpm. the solenoid will not be activated or deactivated between 6900 and 6940rpm?

[dontz125]

I'd suggest you don't take the power from the ignition switch, but rather from the fuel pump relay. That way, the solenoid isn't energized until the uS is ready to control it. Probably not necessary in this case, but it's good practice and a pet peeve of mine.

The VE table is more of a 'wire-frame' than 'bins'; the VE is specified at that POINT with interpolation between points. Assuming the engine load / TPS / MAP remains the same, two entries of 6000 (24) and 6400 (28) would return a VE value of 24 at 6000rpm, and 26 at 6200 rpm. Hope that made sense ...

Yes, your Power On and Trigger values mean the solenoid is activated BELOW the threshold.

The hysteresis is normally + OR -, not both. As you say, it would cause confusion for both the operator and the processor! A setpoint of 6920 and a hysteresis of -20 means the solenoid turns OFF at 6920 increasing, and doesn't turn ON again until the revs drop to 6900. As to the actual setpoint, this is something you'll need to do some tune-and-test work to determine.

[classic-indy]

This helps a lot. The description of how the VE table works makes a lot of sense. Hopefully by the end of the week I'll be able to get all the parts I need and have it back up and running. At which time I will be uploading another youtube video.

The stock opening time is 6700rpm. I will start from here and decide if a change is indeed needed.