KLX250S EFI snorkel mod


With carburated klx’s it’s very common mod to change airbox snorkel or remove airbox lid. Removing the top makes horrific intake noise and exposes airfilter to water and dirt. So i tried the kdx200 snorkel. Actually it was ordered as KLX650 C1 snorkel.

To make sure the EFI can adjust it and engine doesn’t run lean i wired a multimeter to bike own lambda sensor wires under the seat. (the meter shows nothing in the picture as the engine is not running). It seemed to show only millivolts at idle, which would indicate lean, but idle was fine and strong. Also after starting cold engine the meter shows 4,7volts?, but after running for 30seconds or so it started to show ok values, and during driving the values are ok.


Riding the bike like this shows the mixture. 0,3 is lean and 0,8 or higher is rich. It was running over 0, 8 most of the time so its not lean st least. It felt like it adjusted the mixture and readings developed to better the longer i rode it.

it’s best to block the SAI before doing this so lambda shows correct values.

The intake noise is noticably louder and power seems to have increased considerably. It accelerates faster. This has to be one of the best mods ever.

I will check the mixture with wideband lambda in near future.



4 responses to “KLX250S EFI snorkel mod

  1. Hello,

    This weekend I tested the A/F mixture by measuring the voltage on the oxygen sensor. So the test was not so straight forward, but at the end I got the results as they should be. Just to note, the bike has KDX200 snorkel, stock exhaust, and no fuel programmer. Outside temperature – about 9C. Attitude about 250m.

    1. Connect two additional thin wires directly in the connector where the oxygen sensor is connected. Close the connector. Connect the additional wires on the multi-meter.

    2. Start the engine. At the very beginning the multi-meter is indicating 4.7v. After ~30 seconds, the voltage starts to fall to 800-890mv. I guess that is the time where the Oxygen sensor heater start to work, and after 30s it will disengage. The oxygen sensor is not working properly before it is heated for good. After disengaging the heater, the sensor is heated by the exhaust itself.

    3. The 800-890mv stays for about 3-4 minutes, and after that it starts to fall to 80mv and stays there. Here I was thinking that the AF mixture is extremely lean. Twisting the throttle was not helping, the 80mv stays on 2000rpm, 4000rpm, 6000rpm. After that I quit measuring, thinking that something is wrong, maybe wrong points where I had measure the voltage… or faulty multi-meter maybe (the bike is running fine, pulls ok and everything seems good, no lean indications).

    4. The other day I opened up again just to confirm the wiring and with another multi-meter as a backup. Again the measurements were the same. Than I opened the bypass on the clutch side.. and again the same results. Hmm. ok I start to ride, and again the same 80mv. BUT, after 3-5 minutes of riding normal, the multimeter started to indicate 890mv, and it stays there. Maybe on the throttle responds it goes down or sometimes up, but stabilize on the 890mv. Than stopped, and after some time in neutral (about 5 min) the indicator shows 80mv again.

    5. My conclusion is that you have to put some load on the bike and than the sensor gives the best results (real driving, like in real life). If it stays on idle for more than 5 min, it falls to 80mv. That is, not that the mixture is lean, I think the ECU knows that the bike is in idle and stays on the current AF mixture, ignoring the oxygen sensor. The sensor is not measuring the real AF mixture because it is not on the needed temperature. After some riding the sensor start to work as expected, and starts to measure the mixture, and the result is by giving voltage to the ecu, which in my case was 890mv.

    • Do you still have the SAI. Secondary air injection. It could make that lean as it is letting air to the exhaust when there is no load?
      Also I think most EFI systems have trim tables, so if it notices lean it will adjust the table up to 10%. So some driving may correct initial lean. At least this happens with my speed triple. It Triumph trim table values are visible with odb connection if I remember correctly.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s