Jump to content

Traction control on AWD


NickRP

Recommended Posts

Yes of course it will be perfect for this, a basic out line is that each wheel will have a speed sensor and once calibrated 'slip' can be measured and then torque reduction applied.

You can use physical sensors or in later cars you can read the abs CAN information and use that.

Link to comment
Share on other sites

Actually, traditional wheel slip based traction control doesnt work with an AWD where all 4 wheels are spinning.  In high end motorsport they use Lidar/Radar or high speed GPS to get a more true ground speed but most of that is costly.

Maybe acceleration based launch control will be an easier option?

Link to comment
Share on other sites

2 hours ago, Adamw said:

Actually, traditional wheel slip based traction control doesnt work with an AWD where all 4 wheels are spinning.  In high end motorsport they use Lidar/Radar or high speed GPS to get a more true ground speed but most of that is costly.

Maybe acceleration based launch control will be an easier option?

Other ECU manufacturer use ROC to make a traction control. What I read it does work well. That would just need an additional axis option in a future release. 

I would like to have GPS function in the next gen Link Ecu. 

Link to comment
Share on other sites

21 hours ago, mapper said:

Other ECU manufacturer use ROC to make a traction control. What I read it does work well. That would just need an additional axis option in a future release. 

I would like to have GPS function in the next gen Link Ecu. 

You can already put acceleration on the traction and launch table axes.  Is this not what you mean?

 

19 hours ago, NickRP said:

Thanks for the replies.

Would it be possible to use high resolution GPS derived speed (available over CAN or as variable frequency) for traction control on Fury (or Thunder) ECU?

GPS is ok for high speed traction control as it is can be quite accurate when moving (provided you have very good signal), but it is very poor at low speed, standing starts etc.  Since you were originally talking about "in 1st and 2nd gear", for decent results I suspect you would need one of the high-end units that incorporate accelerometers and other physics into their speed output calculation to overcome the large lag GPS normally inherits from standing start. 

Something like this would probably give better results:  http://www.gmheng.com/speed_sensor.php

Link to comment
Share on other sites

  • 5 months later...
On 2/12/2018 at 10:51 AM, Adamw said:

You can already put acceleration on the traction and launch table axes.  Is this not what you mean?

 

GPS is ok for high speed traction control as it is can be quite accurate when moving (provided you have very good signal), but it is very poor at low speed, standing starts etc.  Since you were originally talking about "in 1st and 2nd gear", for decent results I suspect you would need one of the high-end units that incorporate accelerometers and other physics into their speed output calculation to overcome the large lag GPS normally inherits from standing start. 

Something like this would probably give better results:  http://www.gmheng.com/speed_sensor.php

Hi , how can you use this speed sensor input with Thunder ecu and confiqure so you can have a speed comparisson with all 4 wheel speed sensors ? And then how can i get the ecu to trigger the traction control  based on this speed diference /slip? Is this possible o do with the Thaunder ecu? Or do i need some other external module ? Thanks in advance ! Mike.

Link to comment
Share on other sites

You would have this radar sensor connected to a DI and assigned as a wheel speed.  You would set that up as non-driven wheel speed in the speed sources menu.  You would then need at least one of the driven wheels connected to a DI and assigned as driven wheel speed.  Slip will be the difference between the Radar (road speed) and the driven wheel speed.

Link to comment
Share on other sites

17 hours ago, Adamw said:

You would have this radar sensor connected to a DI and assigned as a wheel speed.  You would set that up as non-driven wheel speed in the speed sources menu.  You would then need at least one of the driven wheels connected to a DI and assigned as driven wheel speed.  Slip will be the difference between the Radar (road speed) and the driven wheel speed.

Hi Adam , thank you  very much for your reply ! I have a couple of more questions (I am New with Link ecu) , is this sensor compatible with the speed GPi of Thunder ecu ? Please find the sensor specs below :

  • 0 to 5 V square wave, differential or single ended
  • 62.1 pulses per second for every kilometer per hour of speed measured (100 pulses per second for every mph of speed measured)

And what would be the aproximate starting calibration setting and other parameter (pull up ressistor) settings so it will read correctly the speed ?

Thanks in advance ,

Mike.

17 hours ago, Adamw said:

You would have this radar sensor connected to a DI and assigned as a wheel speed.  You would set that up as non-driven wheel speed in the speed sources menu.  You would then need at least one of the driven wheels connected to a DI and assigned as driven wheel speed.  Slip will be the difference between the Radar (road speed) and the driven wheel speed.

Link to comment
Share on other sites

12 hours ago, jmcpower said:
  • 0 to 5 V square wave, differential or single ended
  • 62.1 pulses per second for every kilometer per hour of speed measured (100 pulses per second for every mph of speed measured)

Hmm, this output is a bit higher frequency than I expected, with the thunder DI's limited to 6500Hz they will only be capable of measuring up to a max speed of about 105kmh.  You are going to need a frequency divider to make this work up to decent speed.  Something like this would be needed:  http://www.vems.com/vr-to-hall.html

Whats the cost on the radar sensor?

Link to comment
Share on other sites

  • 5 months later...

Hi Adam,

My appologies for late reply ,

The Sensor was 1600 us$  when i last aked.

I was hoping that Thunder would have an option at the  speed sources conf.  for  each of the 4 wheels max to be compared with the speed sensor input ?  Pobably will need to have a Dinput separately for this radar sensor ...this way probably would have covered any possible wheelspin on a 4wheeldrive car  from any of the 4 wheels.

Is there Anyway we can configure the Thunder ECU as i mentioned above ?

Thanks ,

Mike.

 

Link to comment
Share on other sites

Under what circumstances is this wheelspin ?

 

Straight line, corners, launch, other ? The simplest means of preventing wheelspin...is power management to prevent it in the first place.

 

Is one wheel a problem, one axle ? all wheels ? A decent GPS unit shouldn't have any lag that would cause major problems and if for whatever reason traction is so terrible, it'd still be better than nothing at all. 

 

but as others mentioned, some companies offer TC based on rate of change of a prop/driveshaft which can be viable if the software allows it. Not sure it'd be suitable for all applications, and as with any TCS, would require development time

Link to comment
Share on other sites

 

For example , a tarmac hill climb event with parts of the road being dusty, others dry with good traction and some  parts of the road wet,!

These conditions are very often seen on Hill climb events on the same run , and can be at a launch , straight line , high speed turns or even low speed turns and hair pins !

Power management for Dry conditions would be good for dry but not good for wet conditions and vise versa...so we are trying to get the best out of it as much as we can !

On a 4wheel drive the rate of change of the probshaft would indicate  slip on the rear axle if the speed of the axle increases more than the acceptable rate...but the same can happen if all four wheels are slipping at the same time. This is where you need a reliable and acurate speed sensor to have as a reference so the system can compare each wheel speed to the true vehicle speed. ( this is my personal opinion as i do not know excactly how these sytems work)

If we could have a reliable speed reference and also each wheel speed sensor to be compare to that speed reference for any possible difference during the race, then we could probably manage-reduce the torque to reduce the slip to an acceptable rate.

Thanks ,

Mike

 

 

 

 

Link to comment
Share on other sites

Sounds like you need to be using better tyres or diffs. 

I've worked on 600hp+ 4wd cars for hills etc, and ok yes the wet can sometimes pose challenges...but rarely the dry.

And other than at launch, which would have it's own specific requirements anyway outside of traction control, very rarely are there full 4 wheel losses of traction.

Anyone crazy enough to fully exploit that sort of power on a 4wd car on proper tyres....the last thing most would want is TCS intervention.

 

Although the ROC of a driveshaft under these circumstances would be different than a more rapid ROC for loss of traction via 1 or 2 wheels. But again all of that is still down to fine tuning the system, which would be the case regardless.

Taking the ROC back a step further.....ROC of rpm is another means of determining what is normal acceleration vs a loss of traction. But again there are variables and it could get quite complicated. But when you look at logs of any run where there is grip, vs traction loss....it's always apparent in the rpm trace. Making use of it would be difficult though, but probably not impossible.

 

Even a 4 wheel system would still have some functionality for TCS, except the circumstances where all 4 wheels slipped at the same rate, which lets be honest would be rare. So that conventional setup culd probably still be utilised to a degree in the absence of any other means.

 

Even with a GPS based "undriven" speed reference, you'd still need to spend a lot of time tuning the system for each track/surface/driving style. GPS can and does work for a speed reference, although for hillclimbs, especially if there is tree coverage etc this can affect the signal quality, although any GPS I've used tend to read lower under these circumstances, so TCS simply wouldnt activate so it shouldnt really impact too much on driving. I'm sure a good external aerial would help matters here though and the modern 50Hz stuff is probably a lot better than the 10Hz ones I've used.

 

No matter what route...it will all be a lot of trial and error. But the easiest is power management from the outset for the track and conditions, with perhaps a power setting or two the driver can adjust if need be.

Getting a TC system wrong though, can be utterly rubbish for a driver though and can affect the controlability of a car, especially if it is one the driver might throw or slide about.

 

But log the 4 wheelspeeds first and see how they actually are.

Link to comment
Share on other sites

Thanks for the nice detailed explanation .

Engine Power used to be 700+hp with 680nmtorque  and now aiming for 800+ hp . with 800nm torque

On tire size i cant go any wider ... using 12.5inch wide  Avon slicks.

Center diff is upgraded with 450nm viscous enough to transfer 56% of the engine torque to the  rear axle ( probably not as good as the Evos center diffs with electro hydr. control) ...front and rear diffs are disc type LSd !

The car handles the 700hp very nicely on a nice tarmac road with bearly any straight line wheelspin on 2nd gear ...but as i said when the tarmac is dusty , uneven with bumps, or wet , or even a combination of all these then things can easy get out of hand and reducing power /torque makes things easier and the car is faster during the race ! the ultimate would be a nice traction control (:

But as you said and totaly agree with you i have to get some data of each wheel speed and slip and then decide which way to go !

Thanks once more ,

Regards,

mike.

 

Link to comment
Share on other sites

I think you have too high expectations of what traction control might do especially on a 4wd application.....unless it was some extremely sophisticated setup, which a Link ecu isnt going to offer, or indeed most ecu's.

 

As said, straight line stuff is fairly easy, but the impact TC can have on the driving dynamics...or if you get it wrong, can make them horrible to drive and in some cases unsafe and unpredictable.

Much the same way many people really dislike OEM style traction control, which are very well developed.

 

But yes, you need data, and lots of it. Wheel speeds, g forces, steering angles, brake inputs etc to all get a picture of how the driver handles the car and what action might need taken.

 

Although those slicks sound huge, even 8" wide Avon's are enough to take a 4wd car onto two wheels around some corners.

 

Link to comment
Share on other sites

Hi, I've been working on a solution with a micro controller connected over canbus to make various calculations about what the car is intended to be doing vs what it's actually doing based on wheels speed, accelerometer, which I can concur is very complicated to even arrive at a predictable result. Even if all goes well, we're still talking 1/10ths difference in consistency (not ability) compared to a good driver in an analog car.

This is only straight line stuff too, the very last thing you ever want to encounter is a power cut when wheels break loose in a 4wd, IMO.

Link to comment
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

×
×
  • Create New...