so I took some logs this morning. The first one is a 2-3-4-5.25 pull. the second one is a 4-5-6.25 pull.

you can see that in the first pull there isn't as much knock as the second. Why would that happen? Just a little more heat soaked?

 

1. Set up your wheel size so we can tell what gear you are in in the log.
2. In the second log your biggest spikes of noise are happening in areas of the map that you never hit in your first log.
4800rpm, max boost in 4th. In your first log you aren't at max load in 4th until 5400rpm.
The other spike is at 5600rpm in 5th. Which you don't hit in the first log.

That being said i've never played with the knock sensor and I don't know what your noise threshold values are set to, or even what percentage is considered actual knock.

 

Just to clarify, you mean my actual wheel and tire size? And where would I find my noise threshold?

 

You need to set that up so that we can tell what gear you are in. I don't know how it is hooked up on an MS3Basic.

Don't know, never played with the knock sensor on MS. My board is in the mail. Andrew might be more helpful.

 

ok I pulled that up but I wanted to double check before I started to change anything. Ill ask Andrew about the knock sensor. Thanks for the help!

 

Mine looks like this:



The "1.9" value is empirical, I have no idea why I need to tell it that the wheels are 6 feet in diameter in order to get the VSS to register something that's approximately correct, but that's what it took. I dug into the source code to try to figure out WTF it's doing and OMG the VSS-related calculations are a heap of undecipherable spaghetti.

(edit): Maybe it's lying about the units and it means "feet" when it says "meters"? Wouldn't be the first time TS/MS3 have gotten their units totally confused. That would actually make sense, 1.9 feet is pretty close to the ~23 inch nominal tire diameter.

--Ian

 

The settings I posted above used to work great. Not sure what could be different.

 

My settings were the same as Ian's, I punched in Aidan's numbers so I'll see how it does this afternoon.

Any reason why the digital inputs are different? Like why did you change it to datalog Adian?

 

Just looked again. I think your gear detection is wrong.

VSS looks like it might be right. 120mph in 6th.

So check out Ians gear detection table.

 

That's the input pin on the HC11 micro controller, so it depends on how the board was made. I have an early MSLabs MS3, IIRC yours is a current MSLabs, so it's probably wired to the same pin (check with Reverant). Aidan has an MS2 that he put together himself, IIRC?

--Ian

 

ms3. Its all how you wired it. Mine is wired to datalog because thats where the signal goes.

 

ok. well for the record. my input is PE1 for anyone else that may be wondering.

 

VSS is in the factory harness, so it's all how Reverant wired it.

 

Sounds like he is using PE1 now. Which makes sense. Flex is probably used for flex.

 

Also, Aidan, my gear detection table is the same as Ians. My final drive is a 3.9 which is the only difference.

 

Post a screenshot. Your log shows it not working.

 

 

I look for knock by analyzing all four individual cylinder graphs at the same time. If all four rise and fall at once, it's engine noise. On my built motor which is a bit noisier, I can actually see the center two cylinders rise and fall (closest to the knock sensor). If one cylinder jumps to 2-3x the others, that's a pretty clear indication of knock.

From "morning3":


That's knock on cyl#2. Not a lot, not very bad, but a tiny bit. You can also see a little on cyl around the 2sec mark. Could be as simple as running the car too lean there (12.4:1 in that range). I would fatten it up to 11.8 and pull 2deg of timing at that boost level. There's also a little puff of knock at 9.413s in the same log.

 

Your Gear detection values are still incorrect.

Use Flyin' Miatas' Gear Calculator, https://www.flyinmiata.com/tech/gearing.php, to get/set the correct values.

You need to define your Transmission, Rear End and Tyre Size as a minimum to get the correct values.

 

Did you compare the numbers to the ones on the FM site? For a 6-speed (which is what I believe MiataMan's car has), those are correct.

--Ian