Login   
C6owners :: Forums :: C6 Support :: The Garage

Advice on diagnosing an on-going "Suspension Faulty"

Home   Forum Rules    Forum Help  Conversion Tools
   
Please Register to enjoy additional Member Benefits
Top Thanked Forum Posts: Today | Week | Month | Year | All time | Most Thanks Given: To member | By user
Author Post
mixolydian   
Tue Jan 14 2020, 06:06pm
Joined: Jan 03 2018
Member No: #3382
Location: South
Folks,

Success!

I'm pleased to say that I now have the suspension working correctly. The error is no longer being thrown and the variable damping / travel sensors appear to be functioning as intended. The breakthrough was on Sunday, but I confess to holding off on this update until I'd put some mileage on the car. ~250 miles later, she's in good shape.

The 'programming failed' message proved somewhat erroneous. Even though I received this after setting the reference heights, I would see / feel the car moving when the new values had been set. As per @cruiserphil mentioned (many thanks for this), the car accepts relative values providing they're in a range it can take.

To be honest, I'm still going through the data as I worked through quite a lot different combinations for the values at each corner (~50). I took and re-took the actual measurements, but ended-up settling for something more idealised. The set-up appears to be based on optimising three attributes:

1.) The actual values input for the R and HxM variables;
2.) The per-axle variation from left-to-right; and
3.) The front-to-back variation.

As has been mentioned, the CSS ECU calculates the 'right' value based on the values you give as input.

Obviously we don't really know what this function is, or indeed that it's purely a function of the data we put in. I used so many combinations as I wanted to try to model what impact these variations have on the values the ECU finally applies - as visible in the 'learnings' data in the Standard Parameter sub-menus.

It's hardly rigorous regression analysis, but I started to get a sense of how manipulating the FL, FR, RL, RR settings influenced the final calibration. The ECU will throw other errors if it the front axle left-right delta is too high, similarly so with the front-rear. I suspect, but don't know, this is in proportion to the total amount of correction the system can apply through the stepper motors - viz. if the total variance exceeds the tolerance, then the system will fail to a 'safe' setting.

In my idealised values I worked with the actual values as a mid point, correcting either side of this according to the positive / negative values I'd seen on the travel sensors. Effectively, treating it as a weighted mean. The important part is to review the 'learnings' data from the parameter measurement sub-menu and not making large changes in one go. I think this has been mentioned elsewhere on the forum, but the system possibly guards against large changes to protect it from errors / typos when programming. I found this to be the case when testing the hypothesis with my car.

I may still make a few other changes, but I'm enormously pleased to have everything working. The ride really is sublime. Once I have things finalised, I'll look at sharing the settings / trace data as before.

Thanks again to everyone who helped out - it's massively appreciated.

Mixolydian.
5 User said Thank You to mixolydian for this Post :
 C6Dave (14 January 2020) , FraserG (14 January 2020) , Hattershaun (15 January 2020) , cruiserphil (15 January 2020) , Eric (15 January 2020)
cruiserphil   
Wed Jan 15 2020, 09:56am

Joined: Jan 24 2010
Member No: #38
Location: Celbridge
Hello Mixolydian,

Well done. Thanks for posting your success and analysis.

Best regards,

Phil C.
Go to page   <<       

Jump:     Back to top

User Colour Key:
Head Administrator, Administrator, C6 owner, Technical Expert, C6 Premier Discount Club