Haltech Forum Area is for discussing Haltechs

Haltech Elite traction control strategies

Thread Tools
 
Search this Thread
 
Old 04-11-17, 10:25 PM
  #1  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
Elite traction control strategies

I started playing with traction control, and it's a whole new car!! I can punch it with confidence in the corner exits and bumpy roads, and there's no more anticipation of the second gear traction loss when turbo builds boost before shifting into the third... I can use the entire rev range all the way to 8500 in the second gear before shifting into the third well into the powerband of the turbo which is whole lot faster than shifting into the third at a lower rpm.

That being said, traction control seems to be some combination of art and science, and there is no discussion about it. I want to start one.

I have 4 abs wheel sensors routed to the Elite 2500 ECU SPI's, and yaw & lateral G sensor to AVIx2. My turbo is gt4202r feeding 20b in an RX8. I'm using ignition retard for TCS which increases boost mildly (1-2psi) when it's kicking in alot. Overall I think the ignition retard works fantastic. It just sounds like crap when it's kicking in, but it feels very smooth.

I realize that upto 10% slip feels like normal grip driving, and it doesn't interfere normal natural slip of the wheels during cornering which seems to be less than 5% with my log. This means that if you set the slip at less than 5%, the ignition retard will kick in even when you're not actually "slipping" the driven wheels in a corner just because the driven radius of outer and inner wheels will be different.

My slip table is based on yaw and lateral G readings. I have it set up so that it slips upto 10% when yaw and lateral G is minimal, and I have them tapering off as lateral G and yaw increase. It's way too aggressive right now, so it would kick in even during aggressive cornering with minimal ~10% TPS off-boost power (sounds like rapid stuttering of the engine) but this setting would never let me kick out the rear tire even in sandy unpaved parking lot.

My next goal is to tune the closed loop boost control to set up the boost by gear. 8psi in the 1st gear, 10psi in the 2nd gear, and 12-14psi in the third gear and beyond. After that, I plan to incorporate target boost cut with high lateral G force and high yaw angles to minimize ignition cut in those dicey situations.

I would love to read what others may have done. love to hear about any pitfalls and lessons learned. My goal is safe power delivery over a faster lap time but I would love to read about some strategies to make the car faster on the tracks as well.

Last edited by stickmantijuana; 04-12-17 at 01:34 AM.
Old 04-12-17, 01:05 AM
  #2  
Rotary Enthusiast

 
Havoc's Avatar
 
Join Date: Aug 2004
Location: Australia - Perth
Posts: 1,326
Received 9 Likes on 8 Posts
mate you have done more than me now. Very Jealous

what Yaw and G meter did you use to work with 2500.

Also the ABS input, does that detect each wheel? Or really just know what speed the car it traveling (combines all 4 corners to = 1 speed output?)
Old 04-12-17, 01:30 AM
  #3  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
Bosch yaw/lateral G sensor p/n is 0265005254. You can get it on ebay with harness attached for less than $20 shipped.

I can take screen shots of calibration page and wiring diagram once I get home tomorrow.

I am detecting individual speed of all 4 wheels. You can get away with one front and one rear wheel. Or 1 front and transmission speed sensor. I did all 4 because I didn't know how much slip LSD will allow under acceleration when one rear wheel has traction and the other rear wheel doesn't.

yea it's a team effort. I would love to get the discussion started. We have a solid platform that can do alot. My current setup is far from perfect. It's way too intrusive and I am not even pushing the car hard in the corners. It will take the community effort to figure out what works the best.

Last edited by stickmantijuana; 04-12-17 at 01:42 AM.
Old 04-12-17, 03:39 AM
  #4  
www.lms-efi.com

iTrader: (27)
 
C. Ludwig's Avatar
 
Join Date: Mar 2001
Location: Floyds Knobs. IN
Posts: 5,234
Received 127 Likes on 83 Posts
Just made this post on FB last week.
Facebook Post

Id' recommend using the trans speed sensor or a driveshaft sensor for the rear. The ECU, at this time, doesn't have a function to average axle speeds between the inside and outside tire, so you'll have inconsistency between left and right turns. By using driveshaft speed, you'll get a more consistent speed for the rear. If you're ABS is functional, you'll need a signal conditioner between the ECU and the ABS computer. Simply trying to split the signal between the two will result in the ABS computer throwing fits. If the ABS has been removed, you can feed the wheel speed and/or trans speed sensors directly in to the ECU. If you're running the stock dash with speedo, bring the output from the cluster that would go to the OE ECU into the Haltech.

I also like to use aggressive timing retard as a first step before you start cutting cylinders. It's much less intrusive. Overall, compared to a high end system like the Syvecs, I've been pretty darn impressed with how well the Haltech TC works.

Good find on the yaw sensor! We've been building them but can't build them for that price. We do have a 3G, 2-axis sensor available that we build in-house. https://lms-efi-llc.myshopify.com/co...meter-g-sensor

Last edited by C. Ludwig; 04-12-17 at 03:44 AM.
Old 04-12-17, 03:51 AM
  #5  
Rotary Enthusiast

 
Havoc's Avatar
 
Join Date: Aug 2004
Location: Australia - Perth
Posts: 1,326
Received 9 Likes on 8 Posts
Will have to start following you Chris Looks good.

I have no ABS on my car... so no timed inputs
Old 04-12-17, 05:57 PM
  #6  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
Everything you need to hook up that yaw sensor to haltech elite 2500.
Attached Thumbnails Elite traction control strategies-capture2.png  
Old 04-12-17, 08:22 PM
  #7  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
hi chris, sent you an email.
Old 04-13-17, 01:35 AM
  #8  
Rotary Enthusiast

 
Havoc's Avatar
 
Join Date: Aug 2004
Location: Australia - Perth
Posts: 1,326
Received 9 Likes on 8 Posts
what pins do you use connecting back the Elite? timed intputs?
Old 04-13-17, 01:38 AM
  #9  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
AVI, it's variable voltage input.
Old 04-14-17, 11:52 AM
  #10  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
Learned that traction control by cutting target boost is a bad idea.. the boost recovery is really slow and always non-recoverable especially with a large turbocharger. I set it up so that it will cut target boost by 2 psi when wheel slip is >10% and after the target boost gets cut in the mid rpm range (5-6k) in the beginning of the wheel slip, the MAP kept tapering off to wastegate spring pressure even after regaining traction. I think it's because I set up boost controller solenoid base map with WOT from low rpm. Once you cut boost in the mid-rpm range, the base map gets thrown off and can't reach the target boost anymore. Either that or regaining boost is too slow. I took it off and now just use ignition retard with different target boost by gear. Boost by gear works awesome though. It just takes a long time to set up boost controller solenoid map for many psi settings for different gears.
Old 04-17-17, 11:59 PM
  #11  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
This map works pretty well for me. Nice power slide out of a turn with TCS kicking in when I floor it right before the apex. Come on guys, I can't be the only one playing with haltech tcs..

I am adding a potentiometer to increase slip for more fun... I'll share that map when I get around to it.
Attached Thumbnails Elite traction control strategies-1.png  
The following 2 users liked this post by stickmantijuana:
Havoc (06-19-17), RGHTBrainDesign (01-26-18)
Old 04-18-17, 01:10 PM
  #12  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
I found this unnamed government owned parking lot yesterday night and had a blast. I need new tires..

I did bunch of datalogging while doing donuts, drifting, and purposely loosing control, and I think I am using wrong axis for the slip table. The angular velocity/yaw of the car tend to stay below 60deg/second as long as I'm in control of the car. When I loose control of the car with TCS off, the yaw reads over 100deg/second briefly... I would like 3D slip map but since we only have 2D option, I think the best option would be to use yaw angular velocity as threshold value to cut boost when I'm about to loose control of the car.

I am now thinking of using vehicle speed vs lateral G or MAP vs lateral G for the slip table. when I'm going faster than 45mph or in boost, I can't think of a scenario where I would want more than 10% slip... but not sure what's the best solution.

In terms of physics, I would think lateral G and longitudinal G would be the ideal way to control the slip... it should be a more direct way of measuring how much grip the tire has left?

would love to hear other thoughts.
Old 04-20-17, 12:51 AM
  #13  
Mac Attack

iTrader: (5)
 
MaczPayne's Avatar
 
Join Date: Dec 2005
Location: California
Posts: 5,668
Received 19 Likes on 9 Posts
This is cool. Thanks for sharing your experience with this feature! I've been thinking of doing the same thing on my new setup as well...
Old 04-21-17, 03:10 PM
  #14  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
I am really liking vehicle speed vs. lateral G so far.

I'm bit limited in my experience because I have yet found a good clean paved parking lot that I can lose control of the car. The big parking lot I was at is in pretty bad shape.

Most of my experience comes from this one onramp to freeway that has wide 3 lane bank leading to it.

I didn't like MAP vs lateral G because it's hard to replicate MAP in any corner.

Also, another tip: make sure to calibrate your wheel speed often... I balded out my rear tires while testing all these variables.. and noticed that the TCS kept on kicking in when it shouldn't. When I saw the log, I saw that my slip % was 1-2% when going straight. I recalibrated the wheels, and no problems.
Old 06-14-17, 02:44 AM
  #15  
Rotary Enthusiast

 
Join Date: Mar 2001
Location: Walnut, CA
Posts: 761
Likes: 0
Received 16 Likes on 12 Posts
hi thank you for sharing.

have you tried simply using the front/rear wheel % difference to map out the TC?
using gforce and yaw seems rather complicated and requires a lot of testing.

and for the boost by gear, do you have H pattern shifter, can you share how you plan to get the gear position into Haltech?


Originally Posted by stickmantijuana
I started playing with traction control, and it's a whole new car!! I can punch it with confidence in the corner exits and bumpy roads, and there's no more anticipation of the second gear traction loss when turbo builds boost before shifting into the third... I can use the entire rev range all the way to 8500 in the second gear before shifting into the third well into the powerband of the turbo which is whole lot faster than shifting into the third at a lower rpm.

That being said, traction control seems to be some combination of art and science, and there is no discussion about it. I want to start one.

I have 4 abs wheel sensors routed to the Elite 2500 ECU SPI's, and yaw & lateral G sensor to AVIx2. My turbo is gt4202r feeding 20b in an RX8. I'm using ignition retard for TCS which increases boost mildly (1-2psi) when it's kicking in alot. Overall I think the ignition retard works fantastic. It just sounds like crap when it's kicking in, but it feels very smooth.

I realize that upto 10% slip feels like normal grip driving, and it doesn't interfere normal natural slip of the wheels during cornering which seems to be less than 5% with my log. This means that if you set the slip at less than 5%, the ignition retard will kick in even when you're not actually "slipping" the driven wheels in a corner just because the driven radius of outer and inner wheels will be different.

My slip table is based on yaw and lateral G readings. I have it set up so that it slips upto 10% when yaw and lateral G is minimal, and I have them tapering off as lateral G and yaw increase. It's way too aggressive right now, so it would kick in even during aggressive cornering with minimal ~10% TPS off-boost power (sounds like rapid stuttering of the engine) but this setting would never let me kick out the rear tire even in sandy unpaved parking lot.

My next goal is to tune the closed loop boost control to set up the boost by gear. 8psi in the 1st gear, 10psi in the 2nd gear, and 12-14psi in the third gear and beyond. After that, I plan to incorporate target boost cut with high lateral G force and high yaw angles to minimize ignition cut in those dicey situations.

I would love to read what others may have done. love to hear about any pitfalls and lessons learned. My goal is safe power delivery over a faster lap time but I would love to read about some strategies to make the car faster on the tracks as well.
Old 06-14-17, 07:25 AM
  #16  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
I already did the testing and posted the map. Feel free to try it and post any changes to make the behavior better.

from your questions, you haven't tried TCS yet. you'll answer your own questions after you try the haltech tcs.

as for gear, haltech does it automatically based on rpm and speed.
Old 06-14-17, 06:38 PM
  #17  
Rotary Enthusiast

 
Join Date: Mar 2001
Location: Walnut, CA
Posts: 761
Likes: 0
Received 16 Likes on 12 Posts
it'll be awhile before I get to try it =) car is in pieces right now.
but I like to do my homework while waiting =)

can you please share your experience with just using front/rear wheel speed to map the TC?
why is it not possible?


Originally Posted by stickmantijuana
I already did the testing and posted the map. Feel free to try it and post any changes to make the behavior better.

from your questions, you haven't tried TCS yet. you'll answer your own questions after you try the haltech tcs.

as for gear, haltech does it automatically based on rpm and speed.
Old 06-14-17, 07:57 PM
  #18  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
when you're doing high G cornering, and computer lets the rear to slip upto 10%, you will loose traction.

so why not leave it at 3% or something.. I found that accelerating in a straight line at less than 5% leads to excessive ignition cut and slow acceleration.

I basically wanted the haltech to figure out the corner exit after I navigate the car through the apex. I also didn't want to oversteer into a corner and wreck the car.

so I used the sensor to tell the car when it's in a corner. low slip in a corner and high slip in a straight line.

again, it's something you just have to try out. your car may be different depending on your tires, suspension, car, etc. mine is rx8 with ohlins coilover, staggered setup with RE71R. you also may be a much better driver that may prefer to modulate the pedal and just set the max slip at 10%. I may get there some day. for now, I consider it as a training wheel.
Old 06-14-17, 10:33 PM
  #19  
Banned. I got OWNED!!!

 
RGHTBrainDesign's Avatar
 
Join Date: Aug 2012
Location: San Jose, CA
Posts: 1,666
Received 82 Likes on 75 Posts
Originally Posted by stickmantijuana
when you're doing high G cornering, and computer lets the rear to slip upto 10%, you will loose traction.

so why not leave it at 3% or something.. I found that accelerating in a straight line at less than 5% leads to excessive ignition cut and slow acceleration.

I basically wanted the haltech to figure out the corner exit after I navigate the car through the apex. I also didn't want to oversteer into a corner and wreck the car.

so I used the sensor to tell the car when it's in a corner. low slip in a corner and high slip in a straight line.

again, it's something you just have to try out. your car may be different depending on your tires, suspension, car, etc. mine is rx8 with ohlins coilover, staggered setup with RE71R. you also may be a much better driver that may prefer to modulate the pedal and just set the max slip at 10%. I may get there some day. for now, I consider it as a training wheel.
Just ordered my $18 Sensor with pigtail on eBay (don't be afraid to counter-offer those guys, they'll usually accept it!).

Now my front end is where I'm going to have to spend some money. I have a 1st Gen with FC Front Suspension grafted in. So it's either off to the wrecking yard to hopefully find a S5 Rx-7 chassis with ABS and snag that hub, or pick up a generic wheel/sensor from any car and adapt it to what I already have.

Damn it. I knew I should have gone with an Rx8 front suspension. :P Naw, I wanted a challenge to optimize the FC platform on the front end.
Old 06-17-17, 06:05 PM
  #20  
Rotary Enthusiast

 
Join Date: Mar 2001
Location: Walnut, CA
Posts: 761
Likes: 0
Received 16 Likes on 12 Posts
Originally Posted by stickmantijuana
when you're doing high G cornering, and computer lets the rear to slip upto 10%, you will loose traction.

so why not leave it at 3% or something.. I found that accelerating in a straight line at less than 5% leads to excessive ignition cut and slow acceleration.

I basically wanted the haltech to figure out the corner exit after I navigate the car through the apex. I also didn't want to oversteer into a corner and wreck the car.

so I used the sensor to tell the car when it's in a corner. low slip in a corner and high slip in a straight line.

again, it's something you just have to try out. your car may be different depending on your tires, suspension, car, etc. mine is rx8 with ohlins coilover, staggered setup with RE71R. you also may be a much better driver that may prefer to modulate the pedal and just set the max slip at 10%. I may get there some day. for now, I consider it as a training wheel.
hi !

do you mind sharing what channel you are using when testing the TC? is it just individual wheel speed channel (plus your g sensors) or is there a slip % channel?
Old 06-17-17, 07:50 PM
  #21  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
4xwheel speed, vehicle speed, longitudinal and lateral g, and yaw. that's under main setting. I added vertical g also for those momentary lift-off WOT @ 150+mph.

you set % slip vs. cut in the main screen based on any of the above inputs. you can pull timing based on gear, speed, ignition cut, boost cut/decrease, increase fuel, whatever you want.

Last edited by stickmantijuana; 06-17-17 at 08:23 PM.
Old 06-19-17, 08:19 AM
  #22  
Can Post Only in New Member Section
 
Datsun610's Avatar
 
Join Date: Jun 2017
Location: Sydney
Posts: 1
Likes: 0
Received 0 Likes on 0 Posts
HI stickmantijuana,


I have to say reading your post made me pull my finger out and try out the TC on the Haltech. so I created a an ignition correction table mapping wheel slip percentage against front wheel speed as a start for a trial and it worked a treat at roll racing on Friday! and now I have come to realise the possibilities are endless in regards to tables that can be created

Thank you

I just ordered a Yaw/Lateral G sensor and wanted to ask what are you connecting Pin 1 (reference voltage) and Pin 2 (Bite) to ??? or do you leave disconnected ?


Regards


Fez
Old 06-20-17, 04:21 PM
  #23  
Rotary Enthusiast

 
Join Date: Mar 2001
Location: Walnut, CA
Posts: 761
Likes: 0
Received 16 Likes on 12 Posts
Originally Posted by stickmantijuana
4xwheel speed, vehicle speed, longitudinal and lateral g, and yaw. that's under main setting. I added vertical g also for those momentary lift-off WOT @ 150+mph.

you set % slip vs. cut in the main screen based on any of the above inputs. you can pull timing based on gear, speed, ignition cut, boost cut/decrease, increase fuel, whatever you want.
when you datalog, how do you monitor what change are caused by the TC strategy?

have you noticed the EGT getting too hot with ignition being retarded?

has retarding timing suddenly then back to normal cause any issue with boost control?

thanks again !
Old 06-21-17, 08:50 AM
  #24  
Rotary Enthusiast

Thread Starter
iTrader: (1)
 
stickmantijuana's Avatar
 
Join Date: Dec 2006
Location: east
Posts: 927
Received 11 Likes on 9 Posts
Originally Posted by Datsun610
HI stickmantijuana,


I have to say reading your post made me pull my finger out and try out the TC on the Haltech. so I created a an ignition correction table mapping wheel slip percentage against front wheel speed as a start for a trial and it worked a treat at roll racing on Friday! and now I have come to realise the possibilities are endless in regards to tables that can be created

Thank you

I just ordered a Yaw/Lateral G sensor and wanted to ask what are you connecting Pin 1 (reference voltage) and Pin 2 (Bite) to ??? or do you leave disconnected ?


Regards


Fez
you leave those 2 disconnected. Have fun with it! Feel free to share some strategies that may work better




All times are GMT -5. The time now is 10:58 AM.