Clutch Switch

Update Jan 2009 ffaspector a member on TLZone who has done some reverse engineering of the TL ECU and cracked a lot of the mapping confirmed half of what I presumed.

• [QUOTE=ffaspector; 1393264]Confirmed, sort of....
  
  When you pull the clutch in with a 5th or 6th gear (Static / non Auto / Smart) TRE fitted, you still get the Neutral Ignition Map, but you don't get the Neutral Fuel Map. You get the IAP Idle Map, which is noticeably richer...
  
  Should explain the starting change....[/QUOTE]

So if you don't want to run the neutral ignition mapping, don't bridge out the clutch switch unless you do the mod.

Here is a short video in windows media format Right click "Save as" (133kb). To prove the point, I fooled the ECU that the bike was in 6th gear using a static TRE, I wound the idle up to approximately 2000rpm, then pulled and released the clutch. There is a definite map change depending on the clutch position. I also did it with the bike temperature under 80C as this is when the effect is the greatest. Maybe I'm wrong, but the ECU definitely does something with the clutch switch signal. Bridge it out if you want, but the ECU is fooled into thinking that the clutch is permanently pulled. There is some reason for the ECU receiving a signal from the clutch switch and it definitely does something with it.

Here are some pictures and descriptions on how to bypass the clutch switch for starting the bike but leaving the clutch switch / ECU operation the same. Tracing out this part of the wiring diagram is not too difficult. One side of the clutch switch goes to earth / ground, the other goes to the earth side of the starter relay and "T"s off and goes to the ECU. So when the clutch lever is pulled the switch closes and and earth is available at the starter relay. Why the ECU needs to know if the clutch is engaged / disengaged I'm still not sure of at this point. The video shows running differences of the engine depending whether the clutch switch is engaged / dis-engaged. The next diagram is a rough drawing of the wiring and the change needed.

Click for a larger image

Another Option

You could always bridge out you clutch switch so the bike starts without pulling the clutch in, just cut the cable or pull the pin that goes to the ECU, (Yellow/Black #16 CLT) right at the plug, then whether I'm right or wrong the ECU operation will remain the same. - Wrong! Keep reading, the ECU uses the signal for starting as well.

Don't Blame Me!

If you do this modification, don't blame me if you drop the bike when you try and start it in gear. I've only put this information up as I think it is a better fix than to just bridge out your clutch switch.

Back to the top

Update: (Oct 2009)

Juat leave the clutch switch alone - or do the following mods to keep things working as designed. (Although I think if it's too much trouble to pull the clutch switch to start the engine, it'll be too much trouble to do the following recommended modifications).

More than 13 years since the TL was released we're still discovering some simple basic starting tricks Suzuki used on its EFI bikes. This doesn't just apply to the TL, but is the same for other fuel injected Suzuki's, i.e. the GSXR range too.

When the starter button is pressed, not only is +12V applied to the starter solenoid coil it is applied to the STA input of the ECU. The ECU knows the starter is depressed.

The starter solenoid coil doesn't energise until the other side of its coil has a ground connection. This is done via the clutch switch - or if the clutch switch is bridged the ground connection is permanent. Activating the clutch switch doesn't just ground the starter solenoid coil, it applies a ground signal to the ECU CLT input as mentioned above.

When the ECU gets the STA input from pressing the starter button, and the CLT input is low because it is grounded the ECU 'knows' you're trying to start the engine and it re-primes the fuel pump. It also expects to 'see' crank and cam sensor pulses, if it doesn't it will throw the code for the missing sensor. Who knows what else the ECU does to the timing and injectors when it knows the engine is being started? Something for the ECU hackers to discover. If the clutch switch input is disconnected from the ECU, it doesn't know you're trying to start the engine and if there is a faulty crank sensor, the ECU can't tell you as it doesn't know it should be expecting it. (The cam position sensor may or may not wake the ECU up into realising the crank signal is missing.

The ECU will turn the fuel pump on when it detects the crank sensor pulse - but the men in white coats though it was best to start the pump as soon as you try and start the engine, not when it is already turning over.

Therefore - for normal starting the ECU expects the STA input to go high and the CLT input to go low. End of story.

• Bridging the clutch switch causes the ECU to run a different map than what it should. Don't bridge out you clutch switch. In other models is severely affects power.
• Disconnecting the CLT input as I advised previously prevents the ECU from knowing the engine is being started and it wont reprime the fuel pump and do whatever else it does for starting. Not the end of the world, but it's not right.

Here is a solution as designed and drawn up by D'Ecosse who is an active member on several forums. Ken's also done some great write ups for the frequent mods forum on TLZone.net

Using a relay to simulate the same input to the ECU when starting allows for starting without pulling the clutch switch and the ECU can do what it's supposed to do when starting.

I've pulled all this info from Ken's thread here on TLZone.net In the thread Ken gives credit to the other sources where he learned different things about the clutch input to the ECU, I'm collating it all here too.

Over to Ken

The clutch switch provides more information than just an interlock for the starter. It also provides an input to the ECU which selects different map (this was further validated by the guys hacking the ECU.

Actually Suzuki uses this feature on all its FI systems.

So it is important that you do not simply jumper out the wires to the clutch switch in order to bypass the clutch switch (so that lever does not have to be pulled in). Shorting the switch permanently has negative consequences for 'normal' running - some are obvious and others may be more subtle.

For example on the SV650, it has a pronounced effect on gas mileage and you can see the effect of the switch on the injector pulse signal.

(Image from TeeRiver published on SVRider - click for a bigger one)

On a GSXR it makes a significant different to the power output!

(Image from EDR Performance published on WERA forums)

Convinced?

Bottom Line - not a good idea to just jumper out the switch! Personally, I am not at all inconvenienced by it, but a lot of people seem to not want to have to pull in the lever. So here is safe work-around.

You have a couple of options:

1. 1: You can jumper out the switch but cut the input at the ECM so that it does not see the 'permanent' clutch shorted condition. The downside is that the enhanced starting map will never be selected either.
2. 2. Connect a relay across the switch. This has the bonus of still temporarily selecting the appropriate enhanced starting map, while removi>ng the requirement to actually pull in the lever.

Here is option 2:

Drawing by D'Ecosse - Click for a big one.

As indicated, you can either connect this at the front/headlight harness, or directly at the starter relay itself. The wire colours are same at either location on both the TLS and TLR.

The signal starter button itself is what energizes the bypass relay.

Very simple mod to incorporate if you feel you really must have that bypass.

Back to the top