MoTeCM130 GPRP - What is the best way to approach configuration of the auto blip feature for rev-matching upon downshifts? I would like to ensure that I am implementing this safely and keeping my gearbox healthy. This is for an FSAE 4 cylinder platform. Thank you for all your help recently!
Is your engine setup using Drive-by-Wire? GPRP throttle blipping only works with DBW setups. If you don't have DBW, you can still do rev-matched shifts, but the driver needs to still mash the throttle pedal.
First think to focus on is to make sure the wheelspeeds inputs are working, and the gear ratios defined correctly. Log the channels "Gear Ratio" and "Gear Estimate Ratio" and verify that those numbers match (or are very close (say within .05 to .10).
Make sure your detection (gear lever, or paddles) are correctly identifying upshift and downshift. The channel "Gear Shift Request" should reflect the proper state, "Gear Shift Request Diagnostic" is useful for figuring out what might be missing.
Read the help for "Gear Shift State" -- this describes exactly what happens when. The "Gear Shift Throttle Aim Main" is where your blip amount is requested, the duration is specified by the various "Gear Shift Throttle Aim xxx" parameters (see the help for each one).
Okay thank you! I should have mentioned this is a DBW setup with clutch-less shifting via fuel/ignition cuts (motorcycle engine). I guess this may counteract auto-blip on downshifts? Any further input with this in mind? Gear shifts are dialed in but adding this function would be a great advancement.
I also suggest taking a look at fueling on tip in to make sure enrichment and throttle response is really crisp. This can be rough checked quickly simply by revving the engine from various speeds and making sure response is almost immediate, very crisp, and adjust fuel functions as needed until it's working really well.
If you do not have an actuator changing the gears, but just a linkage from the driver (cable or mechanical activation), you still need some way to detect up / down shifts (i.e. strain gage, potentiometer, switches). Tell us more about this.
Thank you for all your help David! We are currently tracking shifts through paddles as a driver switch, and the shifter is actuated pneumatically with an compressed air tank. Attached are some screenshots of what i got going on. I think you got me on the right track?
I would also suggest a non zero value for your gear shift actuator preload values. Pneumatic shift systems don't have enough force behind them to pull a gearbox out of gear and naturally have a delay from actuation to full torque or 20-50ms depending on system configuration and operational pressure. Typically a starting value of 15-30ms works well in the gear shift actuator timing for preload.
You will also want some pre shift time in your downshift gear shift timings. The pre shift is the area where you reduce engine torque output with ignition retard and/or ignition cut and start the throttle blip. As a typical DBW motor will achieve a controlled 0-100% opening in 70-100ms, you are likely looking at 30ms or more to achieve your 50% aim from your minimum throttle aim/idle pedal offset value. The idea in the pre shift is to maintain engine speed, charge the intake with sufficient fuel and get the throttle open for a crisp blip. Any filtering on your intake manifold pressure sensor will hinder the performance of the blip, although you can assist this a little bit with a well tuned inlet manifold pressure estimate table and the inlet manifold pressure decay and blend values.
All I can say is that this is a car specific issue, because not all cars have this issue. Also not all cars rev straight to the redline, but only have 500 or 1000 RPMs too much throttle auto-blip (you can check the topic I linked above).
Has this been done? I found a company in the UK that claims to offer a kit. It seems one negative would be that you would have to let the throttle move within your hand when the solenoid blips the throttle.
This seems like 10000x more trouble/risk/money than it is worth. You would have to have some kind of actuator that pulls the cable for you. Seems VERY odd. Sure, anything is possible with enough money, but if this is something you really need, I feel you either need to just practice or get a new ride-by-wire bike.
I am not sure, but for me it is or must be muscle memory, I do not pay much attention to the mechanics. Maybe a slipper would be a better application for you. I would be very skeptical of any product that actuates anything that takes control of my bikes components. Like Cruise control in my car, man, they never told me NOT to take a nap when I turned it on.
I worked with race cars that did these; it is fairly commonly done when they are running sequential dogboxes. They would use a pneumatic actuator to "kick" the throttle quadrant with varying levels of accuracy. It can absolutely be done, but it can also be an absolute nightmare. Rev-matching is only worthwhile when it is close, otherwise it is as bad or worse than when omitted. There have been some very good tidbits on here about how to manually do it smoothly, and that has helped me a lot. Namely, focus on keeping your right hand's "C" shape constant underbraking and rotate it (allowing the fingers to slide on the brake lever) so I'd exhaust those resources first before committing to an additional gadget
Have a buddy with a C7 Corvette (damn, what a freeking nice car) . It auto-blips the throttle with downshifts. It's really cool, but still takes some getting used to coordinate the shift with what the computer wants to do. Shift too slow and it doesn't go that well. The one time I drove the beast, i got it wrong a bunch.
Speaking of askew body parts - the old R-bikes had footpegs that were equidistant to the cylinder in front of them. As the cylinders were offset front to back by the same distance that the crank arms were offset, one foot was always further forward than the other. After riding 2 of those old R-heads over 2 decades, it took a long time to get used to having my feet in the same plane when the offset was gone.
I swapped to the R6 throttle in order to cure my rev matching affliction. Works just fine. I had the muscle memory but there's just too much twist to the stock throttle drum to make rev matching work easily.
For my particular application, I have a Honda K24 2.4L coupled to a 6-speed Honda K-series transmission. That said, the methods I used should be considered to be universal and easily adapted to most any engine/manual transmission combination with a little innovative thought.
The first item needed is the 'Throttle Blipper', or technically a pneumatically-actuated device capable of mechanically blipping the spring-loaded throttle cam. Here is the actuator that Flatshifter manufactures. It makes use intake manifold vacuum to drive the throttle cam via the throttle stop screw located on almost all throttle bodies.
In order to ensure the actuator spindle shaft was compatible with my Skunk2 70mm TB, Flatshifter needed to know the thread size of my stop screw and length of shaft required. I provided Graham with this photo.
I made the bracket in my garage from a piece of 1/8" cold rolled steel. While not the best photo in the world, you get the idea. The actuator was solidly mounted and easily moved the throttle cam without binding.
Here is a photo of it installed on the transmission. You can see the adjustment screw on the top of the rod that allows you to adjust the exact force needed to provide a trigger to the Throttle Blipper.
In my application, I needed the force sensor shift rod to give the blipper control box a trigger in BOTH directions. That means a modified force sensor shift rod that sends a trigger to the control box on extension and compression. Graham (Flatshifter) was happy to help me out with a slightly modified version of the shift rod sensor and even equipped it with rose joints on both ends (not typically included).
But this modification would now mean I get a blip while upshifting, which is definitely undesirable. To solve that problem, I installed a small pilot relay (with N.O. contacts) controlled by my +12vdc brake light wire. This relay was installed between the shift rod sensor and controller, effectively allowing the trigger to activate the control box only when my brakes were applied. This ensures blips can only occur when braking and downshifting. When upshifting (no brakes), the blipper is disconnected electrically.
Very cool, though I just have to ask, as an engineering student, why not just alter the throttle pedal to make it easier to blip manually? I'm pretty damn uncoordinated, but mange by having wide feet, more or less toe toeing my downshifts.
EXCELLENT post!! Thanks for taking the time! I've thought a little bit about how to achieve the same thing but never worked myself up even to the point of scoping out the necessary parts. I think you've done a very cool thing and i hope it serves you well!
3a8082e126