auricomXL

hello all,
my buddies and i have been using MS to controll our s5 NA 13b so far. we took the tuning as far as we could go without loading the motor (which isn't too far in the grand scheme of things). now we are working on getting our mock up (spare) motor fitted into the car. this will take a bit of time considering it is a Nissan 240sx. in the mean time, the electrical engineering side of our group has begun a senior design project to controll the 13b, which ideally we will use instead of MS when all is done.

we appreciate MS, but feel that a superior solution can be created when only focusing on one type of motor. universal compatibility is one of MSs biggest advantages, but since my friends are getting school credit for this project anyway, we figure we should try making a controller specific for rotaries, and still keep it under a couple hundred dollars.

basically the mech. eng. side of the group will be telling the EE side exactly what parameters need to be working in the ECU, then they will program this into the microcontroller and make it happen. i will post a follow up indicating in brief what we are planning on doing (without going into excessive detail), and i'd love to get suggestions from others on what to add. at the end of the project, we will have the school pay for a printed circuit board, and if everything works well, we may be willing to sell this product off. it will certainly be more geared to the 13b, and will hopefully be less expensive than MS.

auricomXL

Critical Inputs:
*****************************
24CAS for engine angular velocity

2CAS engine angular position counter, resetting once a pulse

(With 24CAS and 2CAS both the velocity and position of the engine will be known)

IAT + MAP for air density (rho = P/RT). MAP will also be used to find x coord for many lookup tables

Semi-critical inputs
**************************
Oil press - for safe shutdown

TPS - for accel enrich and activation of secondary injectors

Coolant temp - for warning light followed by safe shutdown if light is ignored

Wideband O2, firstly just for tuning of speed density, eventually for closed loop operation

2xEGT thermal couples - for tuning spark advance, eventually some closed loop functionality(?)

Outputs:
**************************************
4 different signals for 4 fuel injectors. At least the secondaries will be peak&hold capable. both the "duty cycle" and "pulse start position" will be controlled by ECU. The pulse will start at or slightly before the calculated "peak runner air velocity". easy user input for injector type and saturated/P&H type (so when you upgrade injectors, you only need to change one value in the ECU). Signal derived from map similar to megasquirts.

PWM signal for water injection valve (probably designed to work around the aquamist "system 2c" setup). Duty cycle only (no timing)

Ignition control for single 2 tower coil (may add trailing spark capability later). Field collapse timing derived from map similar to megasquirts.

2 signals (type TBD) for VDI and aux ports. 5th/6th on same air circuit, VDI on its own. each signal will activate as a function of user defined RPM.

boost control solenoid (PWM) for waste gate actuator. signal derived from a boost map on the ECU (at least 8x8, or 4x16) - yes we are turbocharging our NA s5!


this is what we are going to try and start with. we will also probably have a very crude 1 dimensional warmup table (based on coolant temp). we will also have a "cranking", "warmup" and "running" mode that limits functionality appropriately.

we are also still debating on the best way to do accell enrichment (deltaTPS or deltaMAP). as far as the EEs job is concerned, another big goal to strive for is superior real time tuning, which we think MS is a little lacking in.

please feel free to make suggestions. We might not be able to implement everything within the semester long project, but future development is all about making our potential clients happy with the capability of this unit.

Cheers,
Max

toplessFC3Sman

iTrader: (1)

One addition that would make MS better for rotaries is being able to delay when the primary injectors stage off after you hit the staging setpoint to account for the differences in spatial placement of the primaries and secondaries. Besides this, MS seems to work pretty well for me, but the best of luck on your project!

auricomXL

i think the reason we even started talking about doing this was that megatune kept erasing our settings, so we would have to re-do a few tables every time we would start up. My EE buddy thinks he can desgin a comparable circuit to MS much more efficiently, and he looks at it as a learning experience too. i have to admit, even in helping him plan this project has forced me to teach myself about how engine control works, so at the very least we will have gained some knowledge.

topless,
so you set your primaries to turn completely off at some point after the secondaries turn on?

muythaibxr

I plan on adding a configurable delay for the primaries to scale down after the secondaries come on.

auricomXL: The only reason you'd end up with the behavior you're talking about with the tables is if you had something bad going on with a USB->Serial controller or something similar. I have _never_ seen that behavior in almost 4 years of writing MS/MS2 code.

Ken

auricomXL

ken,
we actually thought it was something like that or possibly the REALLY old and shitty garage laptop we have dedicated to megatune. we could probably fix the problem on this end, but the MEs in the group have been dealing more with engine mounting, driveshaft fabrication, and the forced induction system. the EEs are starting their senior design class anyway, so they needed a good project to work on. we just put 2 and 2 together. i would imagine we will keep using MS as a backup and as a big informational resource.

on the topic of primary injectors: what is the advantage to scalling down primary duty cycle to zero after secondaries come online? as i said, i was planning on including a linear function that keeps primary duty cycle at or below secondary duty cycle at high rpm WOT (even with much larger secondaries than primaries), but it never occured to me to eventually shut off primaries all together. what is the benefit?
thanks,
Max