[Diy_efi] Traction Control System (update 2)

Fri Apr 26 11:26:49 GMT 2002

hmmm. I think you need to do a bit more investigation. ABS 
sensors (All that I've seen) are simply inductive pickups, which use 
a toothed rotor, usually on the outer CV, or drum (2WD cars). Each 
sensor gives out an AC signal, of a set voltage, but varying 
frequency. The operation is clear. As a tooth on the rotor comes 
close to the inductive pickup, the voltage out of the sensor rises, 
as a gap approaches, the voltage falls, this gives the AC output. As 
the wheel rotates faster and faster, the teeth pass the rotor much 
faster, and therefore the output frequency is much faster.

I'm afraid that doing this on a PC is a bit OTT. All you need is what 
has been suggested. A simple micro (I'm an Atmel AVR man ;) ) 
Which has at least one timer on board. Use the LM339, or 
whatever other op-amp you wish to get the pulses, input these to 
four input pins on the processor, and use the timer to get the 
frequency of all four wheels. Then a simple differential algorythm in 
the processor will give you the resultant difference in wheel speeds. 

For reference, Vauxhall have two ABS sensor rotors. All available 
on CV joints. The early system used 29 teeth, and the later 
system used 43 teeth. This gives a pretty high resolution. I guess 
that they increased the number of teeth to get a higher accuracy. 
They started including traction control as an option on the cars 
which appeared with the later 43 toothed rotor

What are you measuring the voltage with? Are you using an 
oscilloscope? If you're majorly stuck, and it sounds like your 
hardware knowledge is very limited, then maybe look at getting a 
Picoscope or another add-in board for your PC. It's very OTT 
though, you may struggle to code quick enough code to deal with 
the data. A micro would be much neater, your code would be 
dedicated to the task of TC. I would suggest getting an eval board 
aswell TBH.

I'm off the internet ATM, but if you want some internet links, I'll get 
some together for you. Other than that, my system info will be 
available pretty soon, which you could try out on your car too. I'll 
try and get a couple of prototype boards done, and you can have 
the code... (be a couple of months though, as I'm v.busy at work)

Brian.
P.S. if your using a DVM or similar to get the voltage off the 
sensors, you are reading incorrectly. The response of the meter will 
change with frequency. Basically, using a DVM on the output of 
the sensor is inaccurate. Check the DVM, it will have a frequency 
range for measuring AC signals.
> Perry,
>   You are correct... Typo on my part.. the signals are actually 70mV at 1
> tire revolution per second.
> I hopped onto NSC website and looked at teh LM339. To be honest - its all
> cryptic to me <grin> I see that it has 4 inputs and thats about all I can
> get out it.
> 
> If I understand it correctly, the LM339 will take an analog AC signal and
> convert it to DC? Even at 70mV would it require a opamp to bump up the
> voltage still?
> 
> Also, what does this Hitachi H8 etc do?
> 
> Thanks.
> 
> Crescent Kao
> Director of Marketing and Sales
> www.c-speedracing.com
> 
> 
> 
> -----Original Message-----
> From: diy_efi-admin at diy-efi.org [mailto:diy_efi-admin at diy-efi.org]On
> Behalf Of Perry Harrington
> Sent: Thursday, April 25, 2002 5:44 PM
> To: Diy_efi at diy-efi.org
> Subject: Re: [Diy_efi] Traction Control System (update 2)
> 
> 
> First, your .7mv reading seems WAY too low to me.  I'd expect about
> 300-400mv.
> At full tilt, over 1v.
> 
> Second, you are going at this with much more complication than neccessary.
> 
> Go get the NSC datasheet on the LM339.  Look through the applications
> section
> and find the app for a single supply Zero Crossing detector.
> 
> Implement that circuit 4 times for an LM339.  Hook it to your ABS sensors.
> 
> You now have 4 digital pulses which mirror the ABS sensors.  Now you can use
> the PC parallel port directly for capture.
> 
> I would recommend a standalone controller, something like a Hitachi H8/3664
> evaluation board ($50 from Avnet, $99 with cables and power supply).
> 
> --Perry
> 
> On Thu, Apr 25, 2002 at 05:16:29PM -0700, Crescent Kao wrote:
> > Ok.. we've made a little more progress on the traction control system.
> > Here's a recap of what we have so far:
> >
> > . Test vehicle - Acura Integra with ABS
> > . Signals generated from the ABS sensors - 0.7mV AC (and up, depending on
> > speed)
> > . Probing the ABS wires at the ABS brain verified the signals. The AC
> > voltage increases as speed increases. (around 1.4mV at 35mph
> >
> > So... at this point we were looking into how to convert the signal to a
> > digital signal and read it via a serial connection (RS232) to an onboard
> > laptop.
> >
> > This is what we have so far hardware wise:
> >
> > . using an op-amp, bump up the 0.7mV+ input signal (above 1v?)
> > . then convert the AC signal to DC
> > . convert the analog signal using an AtoD processor
> > 	(perhaps NSC ADC08034 http://www.national.com/pf/AD/ADC08034.html 4
> > inputs -
> > 	  however it specifies that it interfaces using "Microwire". I have no
> idea
> > what this is. Should we use an 		  AtoD that uses parallel interface
> > instead?)
> > . format the signal to a usual datastream
> > . send the data via UART (9600 or 11500 baud) to the RS232 port
> >
> > This is our current challenge. Does this approach sound correct?
> >
> > Crescent Kao
> > Director of Marketing and Sales
> > www.c-speedracing.com
> >
> >
> >
> >
> > _______________________________________________
> > Diy_efi mailing list
> > Diy_efi at diy-efi.org
> > http://www.diy-efi.org/mailman/listinfo/diy_efi
> 
> --
> Perry Harrington             Linux rules all OSes.               APSoft
> ()
> perry at apsoft dot com 			                 Think Blue. /\
> 
> Those who would give up essential liberty to purchase a little temporary
> safety
> deserve neither liberty or safety. Nor, are they likely to end up with
> either.
>                              -- Benjamin Franklin
> 
> 
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> http://www.diy-efi.org/mailman/listinfo/diy_efi
> 
> 
> _______________________________________________
> Diy_efi mailing list
> Diy_efi at diy-efi.org
> http://www.diy-efi.org/mailman/listinfo/diy_efi
> 

_______________________________________________
Diy_efi mailing list
Diy_efi at diy-efi.org
http://www.diy-efi.org/mailman/listinfo/diy_efi