DBW tuning

Discussion and Support for MoTeC's M1 series ECUs

DBW tuning

Postby greenamex2 on Tue Mar 12, 2024 8:37 am

Hi all

My Throttle servo PID setting seem pretty good EXCEPT on full throttle. Any idea what I need to do to reduce the throttle fluctuations...and it not hitting the target? I am guessing that the two issues are related!

See attached for my data logging which (I hope!) includes the device configuration.
Attachments
Throttle fluctuation.ld
(2.91 MiB) Downloaded 49 times
Motec CDL3+M130+LTCD+MDD+PDM15+PDM16M
Nissan VQ30DE fitted to an AM Sportscars EX2 with a Hewland HP 2000
greenamex2
 
Posts: 353
Joined: Fri Sep 12, 2014 7:06 am
Location: England

Re: DBW tuning

Postby Stephen Dean on Tue Mar 12, 2024 10:06 am

Hi,

Are you using a calibration supplied by MoTeC for the DBW Servo? What is the Part Number for the Servo?

If you look at the Throttle Area table, you will see that at 98.3% Throttle Pedal Position, you are requesting a Throttle Position of 88%, so this lines up with what you are seeing in the logging. You need to recalibrate the Pedal to give you 100% when it is fully open, this will have the Throttle servo also going to 100%. You can also change the Throttle Area table to make the operation more linear, but this can have a negative effect on driveability.
Stephen Dean
MoTeC Research Centre Melbourne, Australia
User avatar
Stephen Dean
Site Admin
 
Posts: 1614
Joined: Tue Dec 02, 2014 10:29 am
Location: Melbourne

Re: DBW tuning

Postby greenamex2 on Tue Mar 12, 2024 6:51 pm

Thanks Stephen. The servo is the Jenvey ETA unit attached to Jenvey ITB throttle bodies on a V6. The calibration was derived from the two that Jenvey supply on their site.
I'll have another look at the throttle map, I have obviously misunderstood something there!
Motec CDL3+M130+LTCD+MDD+PDM15+PDM16M
Nissan VQ30DE fitted to an AM Sportscars EX2 with a Hewland HP 2000
greenamex2
 
Posts: 353
Joined: Fri Sep 12, 2014 7:06 am
Location: England

Re: DBW tuning

Postby NathanB on Wed Mar 13, 2024 8:32 am

Hi Denis,

The throttle area table serves to normalise the throttle servo position so that the throttle position can be made to represent the available airflow through the throttle, so in effect the throttle servo position represents the opening of the throttle blade in the throttle bore, and the throttle position represents the % of available airflow. This was introduced into M1 with the R35 package for the purpose of torque control. In non torque based packages, it can be used to assist in ease of tuning for certain throttle applications where a need to linearise the throttle is required, such as throttle servos with scalloped throttle bores. In other cases people may choose to make the throttle area table linear, so that throttle servo position == throttle position.

This is often a chosen method, as in the M1, almost everywhere a throttle aim is commanded, it is commanding a throttle position, so the system does a reverse lookup of the throttle area table to determine how much throttle servo opening that is.

Throttle area.png
Throttle area.png (23.35 KiB) Viewed 379 times


In the case of your throttle area table, this means 100% throttle aim will target 100% throttle servo opening (100% throttle position), however a throttle aim of 96% will aim for 80% throttle servo opening (96% throttle position).

This isn't what is the cause of of your throttle behaviour. You pedal sensor is only deviating 0.2%, which in your throttle area table would be around 0.8-1.0% throttle deviation.

In your instance the throttle servo control PID is chasing itself.

I would suggest reducing your feed forward at 100% throttle to 17-18. I would also suggest reducing the throttle servo control integral gain down to 10-15%. In multi throttle applications, where there are linkages and non linear behaviour, integral can cause 'ringing' in the throttle as it changes based on error over time.

I also notice (although the channels are logged a little bit slow for this type of diagnosis) that your servo tracking is moving more than the main, which indicates to me you have a sensor on the motor and a sensor on the throttles (or one on each bank). The servo control will only work on the main provided there is no fault detected, however the difference between the two could lead to aim tracking faults and tip outs.

The Throttle servo position sensor tracking linearisation exists for correcting any non linearity between the sensors in such applications when needed.
NathanB
MoTeC
 
Posts: 190
Joined: Thu Oct 15, 2020 1:50 pm

Re: DBW tuning

Postby greenamex2 on Wed Mar 13, 2024 9:55 am

Thanks Nathan
The whole throttle area vs throttle servo position thing finally clicked today. I had been misunderstanding it for sometime and mixing up the two sets of values.
I redid the pid values from scratch and now have a much more stable servo, as a result I also came to a similar conclusion on the feed table and integral value.
The throttle sensors are on each bank. I'll have another look at linearisation now I have a more stable servo.

Learning all the time!
Motec CDL3+M130+LTCD+MDD+PDM15+PDM16M
Nissan VQ30DE fitted to an AM Sportscars EX2 with a Hewland HP 2000
greenamex2
 
Posts: 353
Joined: Fri Sep 12, 2014 7:06 am
Location: England

Re: DBW tuning

Postby greenamex2 on Sun Mar 17, 2024 4:07 am

Think I have got the hang of it, how does this look?

When people say that sorting out just the throttle setup etc can be a whole day job for a pro, I am now beginning to understand!
Attachments
Final calibration.ld
(2.93 MiB) Downloaded 43 times
Motec CDL3+M130+LTCD+MDD+PDM15+PDM16M
Nissan VQ30DE fitted to an AM Sportscars EX2 with a Hewland HP 2000
greenamex2
 
Posts: 353
Joined: Fri Sep 12, 2014 7:06 am
Location: England


Return to M1 ECUs

Who is online

Users browsing this forum: No registered users and 5 guests