Parking-brake follies

[hobbit]

I had a weird experience with the parking brake yesterday, while playing
around in the snow we got in New England. Basicalliy, it got applied and
*wouldn't release* for a while, leaving me sitting in a Walmart parking lot
unable to move. I could hear it *trying* to release, and if I held the
rocker switch down it would keep trying to run the motors to the release
position but stall early in the cycle and back off, repeatedly.

I power-cycled the whole car and that didn't seem to make a difference.
I tried going forward and reverse a little and basically trying to drag
the rear wheels, but the brake still wouldn't release. I started wondering
what bits I would need to get under there and disassemble to let the car move
again, but a bit of persistence on the switch and they finally released for
real and stayed that way. Without touching the switch again, I drove home.

Now, this was a heavy, wet snow and I was out deliberately abusing things
a bit, trying to use the e-brake to lock up the back end and spin the car
around in an empty unplowed part of the lot. The way the switch works is a
suboptimal approximation of what you can do with a real cabled e-brake; at
any speed above close-to-zero you get that frantic BIP-BIP-BIP-BIP-BIP-BIP
from the dash speaker and it applies momentarily until you let the switch off,
but you can't really regulate it well. The temp was also heading down, dipping
below freezing after the snow had fallen. My best guess is that the brake
calipers were all slushed up, and then a pad momentarily froze to the disc
somehow. The attempts to drag forward and back and/or repeated use of the
motors may have warmed them up just enough to un-stick.

If that's a correct theory, then it would seem prudent to put the car in
Neutral once in a while and do some no-regen slowing/stopping with the normal
service brakes in that kind of weather, just to keep the calipers clear and a
little warm. It also shows that the EPB calipers have some sufficient notion
of their own motor position to determine if they've successfully moved,
although I can't quite see how that works from the schematics. I like this
setup even less than I did before, I wish they'd just gone with a traditional
mechanical system that you can actually control in realtime.

_H*

 

[srkz]

Now, this was a heavy, wet snow and I was out deliberately abusing things
a bit, trying to use the e-brake to lock up the back end and spin the car
around in an empty unplowed part of the lot. The way the switch works is a
suboptimal approximation of what you can do with a real cabled e-brake; at
any speed above close-to-zero you get that frantic BIP-BIP-BIP-BIP-BIP-BIP
from the dash speaker and it applies momentarily until you let the switch off,
but you can't really regulate it well.

I believe this is mentioned in the owner's manual, but if you're driving and you pull the parking brake lever momentarily it won't do anything, but if you pull the lever and hold it will bring the car to a stop quickly-ish and then apply once you're no longer moving. I've tested it on the warm dry asphalt of Southern California and even holding the lever does not actually apply the parking brake as you'd assume, it just applies the vehicles brakes to bring you to a stop. No locking of the rear wheels is allowed, no engagement of the ABS happens, it's just the car applying an intelligent braking routine to bring you to a stop, and not even as quickly as a full-panic ABS stop would.

If that's a correct theory, then it would seem prudent to put the car in
Neutral once in a while and do some no-regen slowing/stopping with the normal
service brakes in that kind of weather, just to keep the calipers clear and a
little warm.

I could be wrong here but I believe the Kona has blended braking even when the drive-regen mode is set to 0, unless putting it in neutral is the trick? My understanding is that any braking force is always handled by the regen up to the point that regen isn't sufficient and then the mechanical brakes kick in. The only way to guarantee you're using mechanical brakes is to brake suddenly and hard from speed, or to be at 100% state of charge coasting downhill and apply the brakes, since it can't regen if the battery is full.

 

[KiwiME]

... unless putting it in neutral is the trick?...

Apparently...

 

[hobbit]

The Kona does have fully blended braking, as I've described on my site and here, but
that has nothing to do with the servo-operated emergency / parking brake. Pulling and
holding the lever runs the rear-caliper motors in the apply direction, which is why the
resultant braking force builds up, but it's all from the rear wheels. Regen is entirely
from the front wheels since that's what the main motor is connected to.

Activation of the parking brake at speed is a last-ditch backup, but in most front-wheel
drive cars is necessary to have a little fun spinning around in the snow. The delay and
imprecision of the motor-actuated setup makes it far less useful for that,, and completely
useless if it's just going to arbitrarily lock up or malfunction.

_H*

 

[electriceddy]

The Kona does have fully blended braking, as I've described on my site and here, but
that has nothing to do with the servo-operated emergency / parking brake. Pulling and
holding the lever runs the rear-caliper motors in the apply direction, which is why the
resultant braking force builds up, but it's all from the rear wheels. Regen is entirely
from the front wheels since that's what the main motor is connected to.

Activation of the parking brake at speed is a last-ditch backup, but in most front-wheel
drive cars is necessary to have a little fun spinning around in the snow. The delay and
imprecision of the motor-actuated setup makes it far less useful for that,, and completely
useless if it's just going to arbitrarily lock up or malfunction.

_H*

My gen 1 Leafs were good for that (on rare occasion) with the standard foot operated pedal.

 

[srkz]

Pulling and holding the lever runs the rear-caliper motors in the apply direction, which is why the
resultant braking force builds up, but it's all from the rear wheels.

[...]

Activation of the parking brake at speed is a last-ditch backup, but in most front-wheel
drive cars is necessary to have a little fun spinning around in the snow. The delay and
imprecision of the motor-actuated setup makes it far less useful for that,, and completely
useless if it's just going to arbitrarily lock up or malfunction.

_H*

If it was as simple as a servo that fully actuated the rear calipers then you'd expect the rear wheels to lock up eventually, no? I can pull the parking brake lever from at least 45mph and the rear wheels never lock up nor is there any sort of ABS activation. Are you suggesting that the reason the rear wheels don't lock is only because the servo takes too long to actuate? Because I suspect instead that the car is smart enough to know you're in gear and moving forward and is not fully applying the parking brake specifically to prevent you from losing control.

Time to go pull the parking brake at 60 and see what happens!

 

[srkz]

Yep, I just tested it and I can confirm that at least on my car it works exactly as I described. If the car is moving, the parking brake WILL NOT engage, instead it uses four-wheel blended braking (really just regen since it doesn't seem to slow down too much faster than holding the left paddle does) to bring itself to a safe halt before the parking brake pistons are engaged. As soon as you apply the throttle to move the car off again the parking brake will disengage even if you're still holding the lever. The rear wheels do not lock up even when holding the lever at 60mph on a cold wet road, because the car is just using its normal brake blend to slow you down to a halt before it even tries the parking brake servo. Incidentally, the amount of braking force that it applies when you pull the lever is (like I mentioned before) firm but not nearly as much as is possible if you really stamp on the brake pedal and actually get the car to the limit of its braking potential.

Here's a video of me testing it for proof:



The car is too smart to allow you to lock the rear wheels in any circumstance using the parking brake lever. It is not possible. It's really only a parking brake for us, not an e-brake.

Of course this is a different issue than them getting frozen with ice/snow! I've never had that happen to me but it does sound like our cars have a not-uncommon issue with the rear pads wobbling, clunking, and sticking under different circumstances. Since the parking brake uses the same pads as the normal service brake it seems like it would fall victim to the same problems.

 

[electriceddy]

Yep, I just tested it and I can confirm that at least on my car it works exactly as I described. If the car is moving, the parking brake WILL NOT engage, instead it uses four-wheel blended braking (really just regen since it doesn't seem to slow down too much faster than holding the left paddle does) to bring itself to a safe halt before the parking brake pistons are engaged. As soon as you apply the throttle to move the car off again the parking brake will disengage even if you're still holding the lever. The rear wheels do not lock up even when holding the lever at 60mph on a cold wet road, because the car is just using its normal brake blend to slow you down to a halt before it even tries the parking brake servo. Incidentally, the amount of braking force that it applies when you pull the lever is (like I mentioned before) firm but not nearly as much as is possible if you really stamp on the brake pedal and actually get the car to the limit of its braking potential.

Here's a video of me testing it for proof:



The car is too smart to allow you to lock the rear wheels in any circumstance using the parking brake lever. It is not possible. It's really only a parking brake for us, not an e-brake.

Of course this is a different issue than them getting frozen with ice/snow! I've never had that happen to me but it does sound like our cars have a not-uncommon issue with the rear pads wobbling, clunking, and sticking under different circumstances. Since the parking brake uses the same pads as the normal service brake it seems like it would fall victim to the same problems.

Thanks for that courageous video demo.
I don't think I would have tried that myself, but age has gotten the better of my adventurous side. It is a great benefit however to know the reaction to expect if a situation should ever arise where it may be required to use the parking brake as a back up.
If you are interested in more experimentation and research in these areas, perhaps get ahold of our own Kyle Conner/Out of Spec Motoring (see the Podcasts in the General section) who has been known to also push EVs to their limits (and sometimes beyond)

 

[hieronymous]

If you are all feeling really keen, you might turn off traction control and repeat...

 

[srkz]

If you are all feeling really keen, you might turn off traction control and repeat...

My intuition is that the result will be the same, since the TCS blinks a light on the dash and makes a tell-tale clicking sound when it activates. Also, any time I've had the TCS activate mid-corner in other situations I can pretty clearly feel it grab the brake on at least one wheel to regain stability. Neither of those things are present when you pull the EPB at speed, though. The EPB slow-down routine feels very similar to pulling and holding the left regen paddle, only it brings you to a full stop and uses maybe 10-25% more braking force in total.

But yeah I'll go do another test without TCS later today.

 

[srkz]

If you are all feeling really keen, you might turn off traction control and repeat...

Here you go, as requested:



I actually noticed while doing the first test yesterday that if you keep your foot on the accelerator while pulling the EPB, the car does slow down but won't stop completely, and can still be accelerated or slowed with the pedal for as long as you like.

In fact, on further reflection, it's exactly the same feeling as holding the left regen paddle, so I'm pretty sure that holding the EPB at speed is identical in the ECU to holding the left regen paddle until the car comes to a full stop. It does feel a little stronger than holding the left regen paddle, but I think that's probably just because I don't use the left-paddle-hold very often and I've never yanked my foot off the accelerator and held it all the way down from 60 before!

(If you're wondering why my dash shows 2.9 mi/kWh, it's because I coasted down a mountain in neutral on my way to this stretch of road to test for myself whether neutral really does force the car to use only friction braking - it sure does.)

 

[hieronymous]

Nice work. Where would we be without citizen scientists!!

 

[hobbit]

I also went out for a bit of testing today, adding my OBD2 gauges and
logging to the mix. It turns out that srkz and I are both half-right about
parking-brake lever operation at speed. And the various traction/stability
control settings matter.

In my snow experiments I was near a somewhat noisy highway environment
and didn't think to listen for the parking-brake motors one way or the
other. Srkz is right about that point; the *normal* hydraulic brakes are
used in the rapid e-brake slowdown until full stop or very low speed is
achieved, and then the motors run to clamp the calipers. But there is *no*
electric regeneration during a stop like that, under any of the TCS/SC settings.

The violet line is my battery current, indicative of drive or regen. The
first big wiggle is accelerating across a long parking lot and then a normal
heavy regen braking event right afterward. The second wiggle is the same,
but using only the e-brake lever to stop. Note that there is NO negative
motor current; it was all mechanical braking. Here's where I also confirmed
that I too didn't hear the rear caliper servos running on the pull.

This image is a series of accelerations followed by e-brake pulls. I drove
around normally for a bit in between when the images were taken, to let all
the rotors cool down from the previous abuse, because here I wanted to try
and figure out *which* wheels were being stopped by the e-brake lever. Since
I don't have front/rear hydraulic pressure monitors, my finger would be the
test instrument. I started with all the rotors fairly cooled down and then
did several cycles of accelerate/slowing, and tested temps again. Turns out
that the car was doing normal balanced hydraulic braking across all four,
with the front rotors slightly warmer than the rears but all of them clearly
having been worked during the e-brake usage. [The one little negative blip is
from turning around, normal pedal use]

Turning off traction control doesn't seem to affect straight-line behavior.
Turning off *stability* control, i.e. hold the button, completely disables
normal regeneration -- brake-pedal stopping is then all hydraulic, like you
were in Neutral. At that point, the car behaves like a Tesla -- the regen
you set or invoke with the paddles works as usual, but the service brake
only sends hydraulic pressure and wastes the stopping energy. Those favoring
one-pedal driving would be fine with this; I certainly wouldn't.

The parking-brake actuators don't only operate at a standstill; they also
kick in when rolling at 2 MPH or lower, lurching the car to a fairly abrupt
stop from what Hyundai must have considered a safe speed to do that. I'm
not sure what happened in my snow experiments; I'll have to wait for another
appropriate day. In the meantime, it might be worth digging into how the
rear brakes are constructed, both to understand that rattling problem and
what might have caused the freeze-up. The Kona will eventually get worked
into my usual once-a-year clean-out-all-brakes cycle anyway, I imagine,
usually in the spring after they've gotten their big dose of winter salt.
The Kona has four more slidepins than the Prius to keep lubed up...

So this leaves me wondering what happens if you completely lose your
hydraulics, i.e. would the system sense that well enough to try and use
the rear-rotor servos as a true emergency backup even if it would risk
some control loss. Hmm, how to simulate such a condition??

_H*

 

[electriceddy]

I also went out for a bit of testing today, adding my OBD2 gauges and
logging to the mix. It turns out that srkz and I are both half-right about
parking-brake lever operation at speed. And the various traction/stability
control settings matter.

In my snow experiments I was near a somewhat noisy highway environment
and didn't think to listen for the parking-brake motors one way or the
other. Srkz is right about that point; the *normal* hydraulic brakes are
used in the rapid e-brake slowdown until full stop or very low speed is
achieved, and then the motors run to clamp the calipers. But there is *no*
electric regeneration during a stop like that, under any of the TCS/SC settings.

View attachment 9883

The violet line is my battery current, indicative of drive or regen. The
first big wiggle is accelerating across a long parking lot and then a normal
heavy regen braking event right afterward. The second wiggle is the same,
but using only the e-brake lever to stop. Note that there is NO negative
motor current; it was all mechanical braking. Here's where I also confirmed
that I too didn't hear the rear caliper servos running on the pull.

View attachment 9884

This image is a series of accelerations followed by e-brake pulls. I drove
around normally for a bit in between when the images were taken, to let all
the rotors cool down from the previous abuse, because here I wanted to try
and figure out *which* wheels were being stopped by the e-brake lever. Since
I don't have front/rear hydraulic pressure monitors, my finger would be the
test instrument. I started with all the rotors fairly cooled down and then
did several cycles of accelerate/slowing, and tested temps again. Turns out
that the car was doing normal balanced hydraulic braking across all four,
with the front rotors slightly warmer than the rears but all of them clearly
having been worked during the e-brake usage. [The one little negative blip is
from turning around, normal pedal use]

Turning off traction control doesn't seem to affect straight-line behavior.
Turning off *stability* control, i.e. hold the button, completely disables
normal regeneration -- brake-pedal stopping is then all hydraulic, like you
were in Neutral. At that point, the car behaves like a Tesla -- the regen
you set or invoke with the paddles works as usual, but the service brake
only sends hydraulic pressure and wastes the stopping energy. Those favoring
one-pedal driving would be fine with this; I certainly wouldn't.

The parking-brake actuators don't only operate at a standstill; they also
kick in when rolling at 2 MPH or lower, lurching the car to a fairly abrupt
stop from what Hyundai must have considered a safe speed to do that. I'm
not sure what happened in my snow experiments; I'll have to wait for another
appropriate day. In the meantime, it might be worth digging into how the
rear brakes are constructed, both to understand that rattling problem and
what might have caused the freeze-up. The Kona will eventually get worked
into my usual once-a-year clean-out-all-brakes cycle anyway, I imagine,
usually in the spring after they've gotten their big dose of winter salt.
The Kona has four more slidepins than the Prius to keep lubed up...

So this leaves me wondering what happens if you completely lose your
hydraulics, i.e. would the system sense that well enough to try and use
the rear-rotor servos as a true emergency backup even if it would risk
some control loss. Hmm, how to simulate such a condition??

_H*

Hope you can add this to your blog:
http://techno-fandom.org/~hobbit/cars/ev/
I like to refer to this on occasion, thanks