View Full Version : Stumble/Sputter at constant throttle position

07-10-2018, 11:22 PM
Hello All,

So I have only put around 600 km on my DD since buying it last year... long story... broken wrist...

When I pulled it out this spring it ran great... no issues.

A few rides ago it started to studder when I ride at a constant throttle position... and pop/backfire a lot on deceleration. Runs fine under acceleration... just sputters if I cruise on the highway at a constant throttle position.

I have googled the situation and found one other thread dealing with similar issues. The guidance was to check for latest ECU programming, check Lambda sensor, then MAP sensors then injectors.

I live several hundred km from the nearest dealership... and do all of my own work... so I started with the Lambda.

While removing the tank to access the Lambda connector I disconnected the two breather/emissions tubes from the barbs on the right side of the tank. The barb for the vapour canister vent (with a check-valve just below the radiator) seemed to be corroded to i turned the tube toward the floor and out ran a few teaspoons of water. The entire tube was full of water.

I gently applied some vacuum and got a bunch more water out.

Does anybody know why this might be happening? There's no way that any emissions hose connected to a fuel tank should be full of water!?!

If it matters this is a Canadian bike with a soup-can sized vapour canister on the left side of the engine.


Has anyone ever experienced or heard of this?

My only though is that the tube on the tank barb did not have any form of clamp. Perhaps it was drawing air and/or water in at the barb and it accumulated in the tube!?!

I didn't check if there was any water in the tube that leads to the rear throttle body... i'll check that tomorrow. I sure hope it wasn't getting water in one cylinder when the throttles closed (maximum throttle body vacuum).

07-12-2018, 10:46 PM
Okay... so I'm not much for needless modification to bikes... but after thinking about it and doing some research I opted to remove the vapour canister.

I don't know why the line from the fuel tank to the canister was full of water... but I do know that the one-way valve in the line was not working at all. I plugged the line to the throttle body and capped the barb on the tank.

Since I suspected that the lambda sensor was not working properly i opted to clean it (soak overnight in gasoline - dry well) and re-install to see if there was any difference.

After re-assembling everything it ran fine for a short time then started to run like crap. It now sputters and misses and surges at constant throttle position.

Kinda runs like a carbed bike with the choke on... and judging by the smell of the exhaust she was running rich.

Since there was such a drastic change I think I'm going to order a new sensor tomorrow.... $170 CAD +shipping.. :eek:

Since the tank removal is such a headache maybe I'll replace the plugs while I'm at it.

Fingers crossed!

07-13-2018, 10:42 AM
I think you need to open that capped barb on the tank too. That is the tank vent/overflow and needs to be able to breathe. Without that, you might get tank vapor lock. Just run a long hose from the barb down to the ground like on an old bike.

07-13-2018, 03:39 PM
There are two barbed fittings on the tank... one is an open breather that terminates below the engine and the second connected the tank to the vapour canister.

The remaining breather tube is more than adequate to vent the tank.

Thanks for keeping an eye out for me though... much appreciated.

Micah / AF1 Racing
07-15-2018, 11:17 PM
Look for air leaks in the piping at all distances before and immediately after the oxygen sensor. Sounds like classic lean surge. First OEM mapping that came here ran that way stock, Arrow mapping was light years better but not in terms of the bike seeming to guzzle fuel.

07-21-2018, 10:55 AM
I think you need to open that capped barb on the tank too. That is the tank vent/overflow and needs to be able to breathe. Without that, you might get tank vapor lock. Just run a long hose from the barb down to the ground like on an old bike.

This. It may seem like the other vent is adequate, but if one of the two are blocked or capped, the bike will run like crap. Same as with the 750's. I would bargain that your initial issue was due to the bad one-way check valve, effectively blocking the line. Common issue.

07-22-2018, 10:37 AM
So I ordered a new Lambda sensor... which took a week to travel 400km. It arrived Friday evening.

Installed it last night... finished at 1am and didn't feel like heading out for a test ride.

Rode 100km this morning... it runs better than ever.

I'm just grateful that my $200 gamble paid off! i have read horror stories about dealerships changing parts at random for weeks without solving this type of problem.

i guess I got lucky!?!

Micah / AF1 Racing
07-25-2018, 06:27 AM
Glad it worked, there are only a few mv separating a smooth running motor from a rough one.