Gear ratio Calibration & Speed Signal noise

[Bud]

Dear all,

I have one link G4X plug in on my Subaru GT 1999-2000 GC8

I try to set up correctly gear ratio for use on boost control strategy 

i not arrive to made correctly the calibration of Gear Ratio and my speed signal is really noisie.

Can you help me? I share with one log and print screen of speed set up.

Regards

 

PC Datalog - 2023-10-3 7;17;24 pm.llgx

[Vaughan]

Does it have a speedo cable or just a speed sensor?

[koracing]

That sort of noise sure looks like it could benefit from using an averaging function in math channel as the source for speed and gear calculations.

[Vaughan]

4 minutes ago, koracing said:

That sort of noise sure looks like it could benefit from using an averaging function in math channel as the source for speed and gear calculations.

each individual DI input has a filter setting built into it, no need to use a math block.

[koracing]

When I've had noise on the speed signal in the past, the filter settings didn't appear to do much of anything.  Perhaps in this case that would be the first thing to try, however, as it looks way nicer than when I've had noise (random 1400mph readings on a Nissan SR20 equipped 240sx).

[RyanG]

13 hours ago, koracing said:

When I've had noise on the speed signal in the past, the filter settings didn't appear to do much of anything.  Perhaps in this case that would be the first thing to try, however, as it looks way nicer than when I've had noise (random 1400mph readings on a Nissan SR20 equipped 240sx).

I had the same issue when grabbing the speed signal from the back of my dash (cable driven driven). Worked okay, but quite noisy and random 400kph spikes from memory. Filter didn't really fix it. Can you share how your average math channel works? I had put this issue to the side until I got time to actually fixed what I assume is a hardware issue in the dash (30yo electronics).

[Confused]

I too am getting really "noisy" speed signals on my G4X

I'm using M8 hall-effect sensors, one picking up from the back of the wheel hub on the front driven wheels, and one on the prop->diff bolts for the driving wheel speed.

Exactly the same hardware, exactly the same wiring, but with the G4+ I had before...

 

 

 

(As an aside, I get really confused with the "Driven" and "Driving" terminology in the G4X, and have to really think about which is which, "Driven" and "Non Driven" was much easier!)

[Adamw]

10 minutes ago, Confused said:

one picking up from the back of the wheel hub on the front driven wheels, and one on the prop->diff bolts

This is likely your problem, target wheel edge accuracy is critical with hall effect sensors.  

G4+ had heavy hardware filtering and a hardcoded software filter, which worked ok for most, but does have a negative effect on control accuracy of synchronous functions like VVT where timing is critical.  G4X has a user selectable software based filter on DI's when used as speed inputs.  Does the filter not take care of it?

 

32 minutes ago, Confused said:

(As an aside, I get really confused with the "Driven" and "Driving" terminology in the G4X, and have to really think about which is which, "Driven" and "Non Driven" was much easier!)

I agree, the G4+ terminology seemed more intuitive to me and even now I still occasionally need to pause and think about it before changing a setting or whatever.  

[Confused]

My "feeling" is that the G4+ speed seemed to reflect what I'm seeing/feeling, whereas as soon as I enable any level of filtering here, it just feels "laggy".

With the G4+ when I came to a stop, it seemed that the speed showed 0 as soon as I stop. With the G4X, it seems like it takes "a second" before it registers 0.

 

Your documentation is usually absolutely on point, but this is unfortunately missing as far as I can find  

The only reference to this input Filtering I've seen is in the VVT Settings page, which I have completely ignored until now, as my engine doesn't have VVT!

 

[koracing]

3 hours ago, RyanG said:

I had the same issue when grabbing the speed signal from the back of my dash (cable driven driven). Worked okay, but quite noisy and random 400kph spikes from memory. Filter didn't really fix it. Can you share how your average math channel works? I had put this issue to the side until I got time to actually fixed what I assume is a hardware issue in the dash (30yo electronics).

Averaging doesn't really help with large spikes as they tend to just raise the average, or you have to take such a large time interval that it doesn't respond fast enough.  Small noise values that nominally look to be correct should be able to be improved via averaging.  A root mean square or standard deviation equation would probably do a better job to elimintate outliers/spikes.

You would do something like this:
 

It would require playing with the GP speed 2 calibration numbers, as well as the time interval on the averaging function (0.1 sec in the example pic). 

[Bud]

6 hours ago, Adamw said:

This is likely your problem, target wheel edge accuracy is critical with hall effect sensors.  

G4+ had heavy hardware filtering and a hardcoded software filter, which worked ok for most, but does have a negative effect on control accuracy of synchronous functions like VVT where timing is critical.  G4X has a user selectable software based filter on DI's when used as speed inputs.  Does the filter not take care of it?

 

I agree, the G4+ terminology seemed more intuitive to me and even now I still occasionally need to pause and think about it before changing a setting or whatever.  

Thanks Adam, 

I will test tomorrow 

do you have advice which level i need to test?

regards