Discuss all things tuning in this section. News, products, problems and results.
We've had out FSAE car up and running for a couple of months now, and its progressing into a decent, reliable, quick machine. We recently set up flat shifting, and are having some interesting results. I'm hoping you guys might be able to offer some advice?
The car setup:
Husqvarna 450cc single cylinder racing dirt bike engine, with the stock gearbox.
Air shift solenoid system, which is triggered from the ECU.
Shift paddle inputs into the ECU.
I've got the 'Up Shift Request' set to digital input 9, and (i forget which) aux output to actually fire the up shift solenoid. With almost completely untouched settings, we can get a reasonably reliable shift, but also a corrosponding 'crack' form the exhaust that is really bloody violent, and contributing to continued exhaust failures.
Driven Up Shift Torque Reduction Time is set to 200ms across the board,
Driven Up Shift Torque Introduction Time is set to 100ms across the board,
Driven Up Shift Cut is set to 90% across the board,
Driven Up Shift Ignition Trim is set to -10 across the board,
and Driven Up Shift Fuel Trim is set to 0 across the board.
These settings seemed a little to violent for our single cylinder engine, as cutting a spark even kills engine output for a complete engine cycle, as there are no other cylinders to help sustain things. I tried changing the Up Shift Cut to 0 across the board, Up Shift Ignition Trim to -15 across the board, and Up Shift Fuel Trim to 10% across the board, in the hope that retarding the ignition timing would be a bit gentler than cutting ignition events all together... This is somewhat successful, but the shifting is less reliable, and there is still a pretty decent back-fire every shift!
Can anyone clarify that the Up Shift Fuel trim of 10% that I've set corresponds to 10% fuel being pulled out, and not added in? It wont let you put in a negative number, and the help file is pretty ambiguous. To me this seems a bit dangerous, as flat shifting is a high-rpm, high-load environment, and we don;t want to run lean here, but we also need to reduce the engine torque to get the thing to switch gears!
Note: We're only talking about up-shifts here, not down shifts. We don't have an e-throttle (YET!) and aren't allowed to add any sort of throttle blipping device, so we just have to deal with the drivers needing lots of training to nail the down-shifts.
Another strategy I've tried is using a timer on the Up Shift Request digital input. It immediately triggers the reduction in engine torque, but waits 50ms before firing the up shift solenoid. This did't seem to make a massive difference, but we also had an unrelated failure shortly after the change was made, so we didnt get to properly test it.
Any advice / input is greatly appreciated!
University of Canterbury Motorsport, UCM15.
What would really help would be a copy of your map uploaded and as much dataloged information as is relevant, jpegs as well as actual software files.
Personally I've only used the flat shifting on a 6 cylinder and used timing retard rather than ignition or fuel cut.
Obviously on a single cylinder you're right in saying any misfire is a missed cycle and a huge power loss on an already restricted engine. Feel free to post up some pictures of the car in the uild section
Cheers Chris. We've got another test day planned for tomorrow. What parameters would be the most useful to log do you think? I'll make a post and upload the configuration and logging info once we're back :-).
If you're using a timed cut mode then the cut time will have a dramatic effect on the way the gear cut control works and how smooth/effective it is. From my own experience 200 ms is way too long and will result in a large amount of unburnt fuel/air entering the exhaust so I'm not surprised you're having issues. Try reducing the cut time to 40-50 ms and see how that works for you. This is the sort of cut time I use on the NZV8 TLX cars running a TTi sequential box and I've had good results and reliability.
The fuel trim is additive and I'd personally leave that set to 0.
If you're using a barrel sensor for gear position you also have the option of setting the cut up as a closed loop system so the cut is only active as long as necessary for the shift to be completed.
On a single banger engine I would be looking at not cutting the ignition but rather just retarding it and retarding it a lot. Not sure if you can do this in the Link or not.
As Andre says, the timers look too long. I would set the introduction time to 0 and only increase it if you get wheelspin.