Haltech Elite 1500 Flat Shift with air shifter on Suzuki GSXR 1000

Good afternoon, does anyone have any information or advise on how to properly set flat shift with air shifter on a GSXR motorcycle? Got it to a point where it will shift but at low rpm, the more load we have it wont disengage, I would assume we don't have enough torque reduction, but I have tried Fuel cut, Inition cut and Ignition retard. First It would feel like hitting a wall when the button was pressed but, probably two much cut. We don't want to damage the transmission shift forks, so was wondering if someone has experience with this settings on a bike. This is a 2001 Suzuki GSXR 1000 turbo. We been using transmission type as manual, and use a button for flat shift and an generic output to control the air shifter. I might think we should use sequential motorbike as transmission type instead, and in trasn. cont. settings assigned the air shifter solenoid to the used output in the wiring tab. Any recommendations or suggestions would be really appreciated.

At a GUESS, I suspect it may be a timing problem between the power reduction and the initialisation of the shift?

The same back-cut principle that helps lock the leeth in engagement under power will normally be used to prevent the gears jumping out of engagement on the over-run/engine braking. If the gears are loaded in the over-run state when attemptin to shift, it might be too much for the air-shifter to overcome.

I don't know what you're using, but it may help if you led the power reduction with the shifter, so it's slightly pre-loading to pull it out of gear as soon as the torque is released/reduced.

We race cars with GSXR 1000 engines and air shifters. It can be done with timed ignition shift cuts (40-60 mS typically for our cars).

If your gear shifting is attempting to used closed loop control (ie, it cuts until the gear position sensor indicates the next gear has been reached), then you will need to change the gear position sensor to a potentiometer, as the stock is just a switch, and I'm not even sure if it's a "make before break" type, so it might see no value in between gears.

Best way to turn gear shifting with with at least 100hz logging of the related channels.

David thanks for the tips, I been analyzing the logs and it seems that there is no time between the pressing the flat switch and the solenoid activation, therefore the torque reduction takes place to late and the gears wont disengage. I would guess its the reason it will under no load but not under load. With the current setup I don't have a way to delay the solenoid activation following the flat shift activation allowing torque reduction to take place. But that is using the flat shift combined with a generic output with transmission type set to manual. I think there is my error, looking at sequential transmission type there is an option under transmission control "Flat shift link option" and it has the option to start the torque reduction before shift start and gives a time table. So I will try this and see how it works. Also we do have an analog gear sensor for gear position.

For an upshift:

So the problem is the torque reduction is happening before the shift lever moves. Depending on the air pressure and air lines, there will be a delay from when you activate the solenoid until the piston starts to move. This puts the dogs on the wrong side, and you can't get out of gear without adding torque

You need the shift lever to be "preloaded" before the torque reduction, and then it will just pop out of gear. Try this manually: While accelerating lean on the shift lever (pedal on a bike), continue accelerating and it won't change gears -- until you lift slightly on the throttle (so shifter load first, then torque reduction).

Then try this the other way. Get up to speed, back off the throttle and try to upshift, you won't be able to until you add a bit of torque (throttle).

David has given some great info. The timing of events is a huge factor in the shifting process, and the amount of cut or retard and how it's reintroduced are big factors in how the engine response feels.

People often cut more than necessary when the timing of the sequence isn't right, and too much cut feels awful. If you have a sensor that allows you to see the timing between shift request and gear pre-load, that removes a lot of guesswork.

The standard bike gear detection is often noisy and unreliable for closed loop detection so timed shifts are often needed.

David thanks for the replies, appreciate the help. I understand the the workings of the transmission and the problem we have with the torque reduction. But the reason I am asking for advise is that I couldn't get the gears to disengage under load, I know why it wouldn't but I couldn't get it working no matter what settings. My take is that with the previous setting (manual type) I didn't have enough time between the shift command and the and the solenoid activation. With the sequential motorbike type I got now the option to set a delay in order to have torque reduction applied before the actual shift starts. I will play with this and see how it goes. Hope this works.

Mike we have an aftermarket analog type gear position sensor, I think my problem relies on my previous reply, we tried all types of cut methods but all was happening at the same time, hopefully with the new setting I have now the option to set a delay and this might give enough time time between the actual shift command and the solenoid actuation to have the torque reduction start and allow the shift. Hope this works.

Sounds good! I would also verify system pressure, check that your actuator has good alignment and throw for the application. If the actuator is moving in a direction not parallel to the item it's acting on, you lose significant force.

Mike the air shift is not a new install, this was and MPS shift kit and was working previously through its module and oem ecu, 120-130 psi system pressure, same it has always been with no problems with OEM. All we did was remove the oem ecu and shift module and install the haltech. So mechanically all should be good. I hope with the new settings we can get it to work.

That's great news. It's always easier when there are less variables at play.