Paddle Shift

Hi Andre, enjoyed the webinar on paddle shift. The Motec looks so much more intuitive than my Pectel.

On the actuator pre-load you talk about and given that the solenoid signal is half bridge, does this mean that it is a PWM signal? Is a PWM sent to the pneumatic solenoid to allow a partial pressure to the actuator (in a similar way to a boost solenoid) or is full system pressure (circa 800kpa in your example) applied to the actuator during this pre-load phase? I would have thought that if full system pressure is applied, unless you are at WOT, there is a risk that the actuator could rotate the barrel too far before the cut is applied unless the cut itself is initiated by barrel position (not by the shift command), though this would require very precise measurement of encoder/pot voltage? Or is it simply a time delay allowing pressure to build in the actuator before the cut is initiated?

I have a WRC paddle which uses a hall effect switch giving variable voltage output (which presumably means I need to set up shift voltage thresholds) is this something you've come across? I will be using a Modena sequential with Shiftec hardware and SQ6M ECU on an asphalt Impreza rally car.

Any advantage of an encoder over a pot for barrel position? Modena put the pot inside the box with the oil which is a bit of a concern.

I'm also I/O limited so am looking to multiplex up/down/detent(neutral) onto one digital input. Anything I should watch out for especially re battery voltage?

I want to retain the lever as a backup, should I worry about its inertia interfering with the shift when using the paddles?

Cheers, Chris.

Hi Chris, glad you enjoyed the webinar. I'll answer the questions in order below:

1. The preload is full pressure however until the torque reversal occurs there's no danger of the barrel moving as the dogs are locked into mesh. This comes down to your throttle position that defines a power on vs power off up shift too. To be clear, the shift is requested by the paddle and the preload is an adjustable time that you can define.

2. I haven't personally used a paddle shift system with a variable voltage but I should have mentioned in the webinar that this is an option and the M1 has this as an option. You'd need to define the thresholds for an upshift/down shift as you've guessed.

3. A mechanical pot is not typically used in motorsport applications for gear position since they are prone to failure due to vibration. They are also limited in their rotation. The Penny & Giles contactless hall sensors seem to be industry standard in motorsport.

4. Multiplexing the shift request isn't something I've personally done and will likely be specific to your particular ECU. For example this can't be done in the M1 however you could request the shift via CAN.

5. Retaining the manual lever is quite common although I've only ever dealt with paddle shift systems that eliminate the lever entirely. I doubt you'll have issues and at worst you could step up to a larger air solenoid for the actuation.

Nice one, thanks. I'll keep the forum updated how I get on. This new build it a real step up in complexity to anything I've done before but if it all works I'm hoping it will be very competitive in class.

Sorry, one more question Andre; in a situation where you have ALS on and are running a significant amount of retard and a fair about of 'JOT' on overrun, if you want to automate a blip of the throttle for a downshift (whether with an external blipper or DBW), how does (or do most) ECUs handle that?

There is an argument which says that it may not actually be necessary to blip the throttle very far or even at all from its 'closed' ALS position. It probably is only necessary in some cases to re-introduce some ignition timing and remove some/all of the ALS fuel and/or ignition cut to get the required blip.

Of course, with ALS on there will be a TPS % where normal fuelling and ignition are resumed anyway and presumably if the blip demand exceeds this TP then normal fuelling and ignition timing will be resumed and the blip will be possible but below a certain RPM and below a certain TP demand you may need to put back the timing and fuel that the ALS has taken out. For example, my gpN car runs quite a lot of JOT and will happily idle at 4000rpm with ALS off and all fuel and ignition cut removed, thus if the TP demand (for a blip) doesn't exceed the min TP to reach this RPM in a 'free rev' state fuel and/or ignition CUT will still be present to some degree (depending on exact TP demand). This may make the blip more lethargic than it need be or require a greater TP demand than would be the case without ALS. For example, if I am asked to rev the engine to 4500rpm for a noise test, I need to give it quite a lot more throttle even though it would otherwise rev at 4000rpm on its own because the throttle is being used more as a ‘torque demand’ device than a traditional throttle. At the TP required the free rev the engine to 4500rpm I might still have 10 degrees of retard and some fuel limit in the map.

Because of this you may need a different strategy between ALS off and ALS on.

I need to try and figure out whether the paddle shift request settings ‘trump’ the ALS maps and if/how it is possible (if indeed necessary) to ‘re-introduce’ ignition advance to initiate a blip in the Pectel. Do you know how the Motec or others handle this situation, is it something you have encountered or considered or am I (as is often the case) over-thinking this?

Thanks

This is going to depend on the particular ECU strategy to some degree. In the MoTeC M1 for example when the throttle blip is requested during antilag operation the retard/cut etc is momentarily eliminated to allow the blip to take place cleanly. Normally the blip will be somewhere in the region of perhaps 50-60% so this would be beyond the antilag retard/cut threshold anyway. I have only dealt with paddle shift on the M1 and Life racing systems and I've not combined these with ALS myself so I can't be too much more help i'm afraid.

Hello @all,

are there any news or experiences on "how to down paddleshift with antilag" ?