Instrumenting Prius brake problem
 

The good news is we are hearing stories that Toyota has a fix for the intermittent "brake problem" that is being applied at the factories since January. The bad news is I'd bought two, three axis accelerometers from Gulf Coast Data Concepts to study a problem I've never been able to reproduce.

Gulf Coast Data Concepts makes a USB interfaced, accelerometer and I bought the X6-2, Li-Poly powered, and X6-1A, AA powered units:
http://www.gcdataconcepts.com/images/x6-2.jpg
The don't include a 3-axis yaw but with two, located as far apart as possible, differential acceleration should give angular measurements, pick two of three:

• pitch
• yaw
• roll

Regardless, this is my first, unpolished measurement of our Prius hitting a speed bump while braking and hitting a speed bump:
http://hiwaay.net/~bzwilson/prius/Prius_brake_010.jpg
The salient data points:

• linear acceleration blue line - a positive acceleration values are braking, zero is no braking, and a negative value is acceleration. Obviously, I had the wrong end pointing forward.
• Two light taps are evident in the Ax line as I approached the bump trying to time brake application relative to the bump. I'll need to figure out a fixed object to initate braking.
• Hitting the speed bump, the brake system really has to do some modulations including a few excursions to zero but braking continues.
• The unit was configured for 80 samples per second and I'd experimented with "dead band" but I may have misinterpreted the timestamp. More work to do here.
• After leaving the bump, normal braking returns and evens out. This is normal braking behavior.
• Red line, verticle acceleration - intially a significant increase in 'gravity' as the car accelerates up.
• The first initial upward acceleration shows a slackening followed by a second upward impulse.
• There is a slight reduction in G as the car 'floats' past the peak and then back to another slight impulse. Then a final float before reaching the pavement with compressed struts and springs.
• The last bit shows the suspension extending up, gradually, and back to normal gravity.

I'm one of the 50% who have never experienced the 'brake problem,' reports of an intermittent fading after hitting a road surface anomaly ... railroad tracks, manhole covers, patched roads, potholes, ice/snow/leaf patches and even a rippled road. Understand that for the past month, I've tried to replicate the problem. Regardless, there are reports that there is a fix that is being applied at the factory and suddenly the interest in the 'brake problem' has faded.

Any new tool takes a little time to master. Issues like drift, calibration, timestamps, and data handling can be investigated with Excel. But for operational use, I'll write some software to handle the mass of data. The device can sample 320 per second so the data can grow quite fast and Excel is so ... limited.

Bob Wilson

Friday morning in a drizzle, I finally experienced the brake anomaly while testing the accelerometer at 160 samples per second:
http://hiwaay.net/~bzwilson/prius/Prius_brake_030.jpg

The Martin Road gate into Redstone Arsenal has speed bump just before the badge check. The first set gave a brief illusion of 'sliding', which is an all too common impression in Huntsville AL. I don't know why but ever since we moved here, every rain is followed by a 'slip and slide' road surface. In fact we lost our '91 Camry on a rain-slick, descending off-ramp. Had it been dry, I probably would have stopped instead of hitting the guy who was stopped at the bottom of the off ramp.

This is a little more detailed view of the speed bump:
http://hiwaay.net/~bzwilson/prius/Prius_brake_040.jpg

Regardless, expanding the data around the first speed bump, I was surprised to find:
http://hiwaay.net/~bzwilson/prius/Prius_brake_050.jpg
I was surprised to see that slight speed increase after integrating the acceleration. Well below the threshold I might detect, the accelerometer data is fairly definitive.

I would be happy to share the spreadsheet and data but at 3.6 MB, it is too large for the 900 KB quota. It can be uploaded from my PriusChat posting in the Technical Forum. I'll also upload it to the YahooGroup "Prius Technical Stuff."

Bob Wilson

Hi,

I had the brake patch, A0B, applied on Saturday. Fortunately, I found this narley pothole near the dog park before the patch:
http://hiwaay.net/~bzwilson/prius/Prius_bump_240.jpg

• 4-5 inches deep, near side
• 7 feet long

I captured accelerometer data before and after the fix and the brake pause symptoms disappeared. Sunday morning we had a mix of snow and rain so I retested the speed bump ... no more brake pause. It looks to be a good fix.

Bob Wilson

Breaking news:

America announces a recall of two million miles of road which is causing braking problems for Toyota Prii. The fix will require some softwareing asphalt repairs to the roads. Toyota has not approved the repair plan yet pending review. To help pay for the fix, Corporations may actually have to pay taxes for their share of road use.

In related news, God is recalling 4 billion humans for repairs to their brains. A programming glitch causes them to destroy the planet that gives them life.

LOL! I wished I'd thought of it first. Reposted with full attribution. Moderators moved it ... darn it.

Bob Wilson

Bob - thanks again for a quality post.

...and Marcus for telling it like it is!