Thanks to the forum for a lot of useful information about mods. I've started with just KDX snorkel and since I have access to cheap dyno, I've decided to check what it provides compared to stock snorkel.
Here are 2 runs:

1. blue - KDX snorkel + clutch switch
2. red stock

Interesting observations:

1. clutch switch mod doesn't offer anything apart from speed limiter removal. I.e. no power gain
2. KDX snorkel offers some mid range torque increase.
3. At the same with stock snorkel bike runs at AFR 13:1. With KDX snorkel it's 13.8:1. So providing more fuel should improve power and torque.

After this dyno I've already changed the end cap pipe on stock muffler and added more ~6% fuel with 2.5 kOhm NTC in IAT sensor circuit. I will post the next dyno soon.

 

You just added more useful information.
This was my totally bone stock 300SM with only one mod... a slip on exhaust.
The readings are only different because different dynos yield different results.

Yours likely doesn't have a secondary throttle plate as it has no dip in the middle of the power curve. I've since removed mine and it helped midrange a lot.

 

Can you show us how you did the 6% fuel increase with the 2.5K Ohm resistor?

Thanks!

 

Yes this would great! Thanks!

 

The logic behind adding 2.5kOhm in series to IAT sensor is following. More resistance makes ECU think the air is colder, hence it's density is higher. Which means ECU needs to add more fuel. By adding 2.5 kOhm NTC we fool ECU that air is about 18 degrees Celsius colder. It translates into ~6% more fuel. By adding more or less resistance (NTC) you can manipulate how much fuel added. So 2.5kOhm is not magic number - it just translates into 6% more fuel which I needed . Table with calculation can be found here. I've checked resistance-temperature mesaurments of the IAT sensor on KLX and it's same as in the table above for Vulcan. So I thing the sensor is exactly the same.

The IAT connection on the bike look following


:

What we want to achieve is following:


I didn't want to cut bike wires, so I spent some time looking for suitable 2-pin connectors to make it plug and play as per circuit below. I found 2-pin female connector - it's widely used in honda jass/fit/civic to connect IAT sensor($1). But I've struggled fiding male connector. In the end I've bought IAT sensor from honda jazz ($4) and cut the sensor from the plug.




Also I only found 10kOhm NTC, so I connected 4 of them in parallel to get 2.5kOhm. For 4 NTC I paid $1.3 total.
I drilled hole in airbox lid and put NTC sensors there. I sealed the hole with silicon.




Then I've mounted everything to the bike, connected bikes and checked that new sensor signal is what was expected.

 

@tooter I didn't disassemble throttle body, but I'm pretty sure there are two plates there. I believe klx250's ECU works differently from klx300 to operates secondary throttle plate. That was also one of the dyno reasons - I wanted to see if there is a similar to klx300 drop in torquer power in the mid rpm range. My bike is 2022 asian spec (Indonesia).

 

The idea is that we trick ECU to thing the air is colder (hence more dense) to add more fuel. 2.5kOhm sensor makes ECU think the air is ~18 degrees celsius cooler and hence ~6% more dense. So It adds ~6% more fuel. Amending IAT sensor signal you can add fuel (within small reasonable range). You can check calculations here. I've double checked klx IAT sensor temperature-resistance signal and it matches Vulcan from link above. So sensors should be the same.

Now the connection:

The stock IAT circuit:



What we want:



I didn't want to cur bike wires, so I spent some time looking for sockets. Female socket is the same which is used at honda jazz/fit/civic for IAT sensor connections ($1). It was easy to find (even though I ordered couple of wrong ones before I found correct). The male socket was difficult. I ended up buying honda jazz assembled IAT sensor ($4) and cutting sensor from the socket to have male socket:




Also I was able to find only 10kOhm NTC, so I connected 4 in parallel to get 2.5kOhm total ($1.3 total). Then I drilled hole in the airbox lid, put sensors and sealed it back with silicon:




Then I assembled and connected everything:

 

It's good you don't have the dreaded "dip".

Had you ecperimented with removing the primary throttle plate control spring? You might not have that either with the benefit of not having stupid US emissions controls, or it might also operate differently. Mine had a rabid fast idle on cold start and throttle "hang" when closing the throttle. It cold starts exactly as easily except there's no fast idle. Shifting is crisper because engine rpms drop immediately between shifts.

(with spring)

(without spring)



Those two mods in combination completely remove ALL ECU controls from the throttle, so it both opens and closes ONLY by the cable.

 

Gotcha thanks for the visuals! Wow ok that's quite a bit of work but how does it feel on the butt dyno?

 

Today I've made another dyno with following mods on top of KDX snorkel

1. 2.5kOhm NTC in series in IAT to add more fuel.
2. Modified stock exhaust muffler. I've removed "straw" end pipe and replaced it with 2.5cm stainless steel pipe. I followed the process shown in this
   video
   . It was quite difficult to remove the stock end pipe from the muffler - took about 30 minutes back and forth. I'm very happy with the sound though - it's still quiet (I measured 90dB), but more pleasant. You can check my
   video here
   . I do feel exhaust flows much better - the air stream pressure after exhaust tip is much stronger compared to stock straw.

The dyno (top-down on the chart):

1. Red - KDX snorkel, modified exhaust, NTC
2. Blue - KDX snorketl, modified exhaust (no NTC/IAT modification)
3. Pink - KDX snorker

Observations:

1. According to AFR chart NTC added less fuel than I expected. AFR is basically just little bit rich.. Maybe the relation between IAT and additional fuel is not that linear I assumed.
2. Without fuel correction there is no power gain in modifying exhaust. Power torque lines are basically the same.