1 2
iansane
iansane Dork
2/9/22 12:09 p.m.

I need help with my Jaaaaaag. I have an idea and I don't know how possible or ridiculous it sounds.

The car ('03 S-type R) has an electric parking brake and obviously the associated modules/motors to go with it. From what I've gathered it uses one motor on the car and two cables like a regular parking brake setup to a specific parking brake caliper on each wheel, because the main brake is a big brembo. 

Well a long while ago it stopped working. Which sucked because it stopped working when it was engaged. The motor still functions fine, I've jumpered the motor wires a few times to actuate the brake either direction with no issues. I replaced the switch and the module and still have no functioning brake. Unfortunately I think this car is new enough that the module has to be coded or paired to the vehicle to function properly. Well Jaguar themselves doesn't work on these and I haven't been able to get the software/hardware to work correctly as I would rather avoid spending $1500 on just the hardware adapter to talk to the car.

Anyway, with the module unplugged I get no error codes and obviously no function. How likely is it I can make my own module?

At first I thought the switch was CAN controlled but it looks like just a specific signal to the module and the module gives its network signal back to the car and kicks 12v to the motor in a specific polarity to control apply/release. Can I just use a motor controller like this; https://www.pololu.com/product/2995 and use the switches signal to actuate? Obviously the first iteration I'd have to consciously not engage while driving or anything weird but if this works I could probably through an arduino back there to control it, right? Like make sure it doesn't enaged with wheel speed or auto engage on engine shutdown/disengage on engine startup, etc.

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
2/9/22 12:51 p.m.

Do you have anything that identifies any of those wires? For example, the white wire that's connected to the motor, module and switch. 12v? 5v? ground? Do you know what the letters/symbols in the triangles mean in a Jaaaaaaaag wiring diagram? I can't quite make them out.

I would be tempted to give the motor a twitch of 12v down the power/ground wires and see what happens to the other two. They're probably some sort of position report. Looks like the switch itself is easy enough.

iansane
iansane Dork
2/9/22 1:17 p.m.

I've got some identifiers. I reupped the image to a better quality/larger but there is also a generic key at the bottom that I had cut off;

The white looks like a ground at the module but confusingly looks like the power supply at the switch? I guess I don't quite understand that section. 

Good suggestion on measuring the other wires for function. I presumed aside from the switch inputs and the notquiteCAN network everything would be 12v but I guess that assumption may fry some things.

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
2/9/22 1:59 p.m.

Good info.

White is signal ground. Y is probably a regulated 5V reference voltage, that's fairly common in my experience.

RW and GW are the motor control, probably just 12v with switched polarity to change direction. I'll bet they're bigger than the others.

Which leaves the feedback. YB is probably just a 0-5v output showing position. It could also be a 0 or 5v switch.

So I'd probably put 5v on Y, ground W and twitch 12v down RW/GW and see if it moves and what happens to the voltage at YB. It's unlikely to be digital.

That motor controller looks like total overkill. Check the wire gauge and the fuses, it's unlikely you have that kind of power need.

SCP sure sounds like CAN but isn't. If you're willing to accept a dumb parking brake that simply works off the switch, you can probably ignore it.
http://www.jagrepair.com/images/TSB/Training%20Information/Multiplex%20Systems.pdf

 

LOL yeah SCP is a pre-CAN network method.

motor current is monitored during EPB apply, as the hardware has been characterized by the OE for clamp force vs current draw.  the system is fused to protect the wiring in case a motor fault caused excessive current draw.  so one very dumb way to control a DIY EPB would be to have some kind of switch (resettable breaker? idk E36 M3 about electrical stuff LOL) that opened at the same amperage as the OE fuse.  i guess you'd want to be able to allow some level of inrush current above that fuse amperage value.  again, i know how EPB's work but IDK E36 M3 about controllers.

 

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
2/9/22 2:19 p.m.

I wonder if the signal wire is reporting current? That's a good point on current vs clamp force. I'd hook it up to my bench power supply and watch what happens as it pushes on the pads.

Any idea what kind of clamping force we should look for on a parking brake?

I know Wilwood has some electric parking brake spot calipers. Here's the installation instructions/details for the controller I didn't know existed: https://www.wilwood.com/PDF/DataSheets/ds1477.pdf

bearmtnmartin (Forum Supporter)
bearmtnmartin (Forum Supporter) GRM+ Memberand UltraDork
2/9/22 2:21 p.m.

I have one on my Land Rover LR3. I have always been terrified to use it and i have no idea if it works.

Paul_VR6 (Forum Supporter)
Paul_VR6 (Forum Supporter) SuperDork
2/9/22 2:48 p.m.

Looks like it's a servo. Motor +/- on RW/GW and you have 5v likely on Y, signal ground on W/WU (which makes sense as it goes to the switches) and motor position signal on YB

Likely logic is that Apply button causes the motor to start turning, once the position signal stalls it decides it's "engaged" and goes into hold mode. Release button moves the motor back a set amount.

Not sure if there is a mechanical aspect to it that holds with 12v power totally removed.

iansane
iansane Dork
2/9/22 3:15 p.m.
Paul_VR6 (Forum Supporter) said:

Looks like it's a servo. Motor +/- on RW/GW and you have 5v likely on Y, signal ground on W/WU (which makes sense as it goes to the switches) and motor position signal on YB

Likely logic is that Apply button causes the motor to start turning, once the position signal stalls it decides it's "engaged" and goes into hold mode. Release button moves the motor back a set amount.

Not sure if there is a mechanical aspect to it that holds with 12v power totally removed.

Is it likely, just to get it working, that I wouldn't need the 5v reference on Y since that's probably what senses the motor bottoming out and tells the system to stop? When the brake was stuck engaged I unhooked the connector from the module to the motor and applied b+ to the motor side at RW and b- to GW (or vice versa I don't remember exactly) to disengage the brake just listening for when the pitch of the motor changed. Functionally, I could do the same thing via the controller or some relays to get it working and then slog at some logic later, correct?

Keith Tanner said:

Good info.

White is signal ground. Y is probably a regulated 5V reference voltage, that's fairly common in my experience.

RW and GW are the motor control, probably just 12v with switched polarity to change direction. I'll bet they're bigger than the others.

Which leaves the feedback. YB is probably just a 0-5v output showing position. It could also be a 0 or 5v switch.

So I'd probably put 5v on Y, ground W and twitch 12v down RW/GW and see if it moves and what happens to the voltage at YB. It's unlikely to be digital.

That motor controller looks like total overkill. Check the wire gauge and the fuses, it's unlikely you have that kind of power need.

SCP sure sounds like CAN but isn't. If you're willing to accept a dumb parking brake that simply works off the switch, you can probably ignore it.
http://www.jagrepair.com/images/TSB/Training%20Information/Multiplex%20Systems.pdf

 

I'm totally going after a "dumb" brake for V1. After swapping the manual trans in, it's a bit unsettling not having a parking brake at all.

It looks like the motor gets sole power through the module and the module was fused at 30a so I was just trying to get close to that capacity.

iansane
iansane Dork
2/9/22 3:17 p.m.
AngryCorvair (Forum Supporter) said:

LOL yeah SCP is a pre-CAN network method.

motor current is monitored during EPB apply, as the hardware has been characterized by the OE for clamp force vs current draw.  the system is fused to protect the wiring in case a motor fault caused excessive current draw.  so one very dumb way to control a DIY EPB would be to have some kind of switch (resettable breaker? idk E36 M3 about electrical stuff LOL) that opened at the same amperage as the OE fuse.  i guess you'd want to be able to allow some level of inrush current above that fuse amperage value.  again, i know how EPB's work but IDK E36 M3 about controllers.

Yeah the SCP seems like what Jag came up with before CAN and then just used to for sub systems along side CAN later.

Paul_VR6 (Forum Supporter)
Paul_VR6 (Forum Supporter) SuperDork
2/9/22 3:38 p.m.

You can either use some current sensing on the motor, or watch for no position change to stop applying signal. Would be interesting to see if if you just apply 12v or -12v to the motor does it still have a hold tq on the brake or not.

iansane
iansane Dork
2/9/22 3:45 p.m.

In reply to Paul_VR6 (Forum Supporter) :

It does stay in whatever position you actuate it to. Currently open, allowing me to drive around. If I apply 12v to engage it and just leave the connector unplugged/not jumpered it will stay there.

Paul_VR6 (Forum Supporter)
Paul_VR6 (Forum Supporter) SuperDork
2/9/22 4:23 p.m.

If it works that way, and you can control your hand, it's likely possible that you can just use a pair of relays. Hold it down until your brake holds, hold it down till released. 

iansane
iansane Dork
2/9/22 4:35 p.m.

In reply to Paul_VR6 (Forum Supporter) :

If I get some time tonight I think I'll try that. If it works then I'll play with what the output signals are from the motor when powered different directions.

Paul_VR6 (Forum Supporter)
Paul_VR6 (Forum Supporter) SuperDork
2/9/22 4:43 p.m.

You can get off the shelf reversing relays for exactly this application.

Keith Tanner said:

Any idea what kind of clamping force we should look for on a parking brake?

FMVSS 135 requires park brake to hold vehicle stationary on 30% slope at GVW. Since there are 2 brakes contributing:

Fpb = 1/2 * GVW * sin (atan 0.3)

that's the force at one contact patch. If you know your tire rolling radius, your brake effective radius, and the mu of your pad / rotor, you can calculate the required clamp force.

getting from clamp force to motor torque is the tricky part. Short of having a consistent measurable way to apply torque, I guess you could:

  • stop on a movable ramp that is horizontal
  • apply the EPB to some known motor current
  • raise the ramp until the car starts rolling
  • measure the angle of the ramp

and repeat those steps for increasing motor currents until you reached 30% grade

No Time
No Time SuperDork
2/9/22 9:35 p.m.

Massachusetts uses a simplified test for annual inspections.
 

Basically the parking brake needs to keep the car from moving when rpms are 1200-1300. Much easier to do on an automatic, but might be a way to test you holding force with a manual if you aren't abusive to the clutch. 

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
2/9/22 10:28 p.m.

In reply to No Time :

I have two Miatas that would require vastly different parking brake abilities for that test :)

Maybe we can test clamp force with a lead cell, but I think finding a 30 degree slope and increasing the clamp force until it holds the car is probably easiest. Although you could perform the tilting ramp trick with a flatbed tow truck easily enough.

No Time
No Time SuperDork
2/9/22 10:57 p.m.

In reply to Keith Tanner :

I've always thought that it was too much of a "one size fits all" when doing the pre-check before the annual inspection, but figured it might be the parking brake equivalent of a butt dyno to get the adjustment in the ballpark. 

Some steel bar, a drill bit, a torque wrench (or scale) and some math could be pretty accurate for applying the proper torques to the wheel, but like you said probably more work than finding a suitable test location. 

Paul_VR6 (Forum Supporter)
Paul_VR6 (Forum Supporter) SuperDork
2/10/22 11:14 a.m.

I love that this is one of those things that most people aren't trying to figure out, yet everyone here is on board with the complete lunacy of it all.

iansane
iansane Dork
2/14/22 12:15 p.m.

I got around to messing with a reversing relay and came up empty handed. I get no actuation of the motor using the switch to actuate the relay. If I hit the relay's trigger with 12v to moves the motor. I'll have to get a second set of hands so I can probe the connector when the switch is activated to see what I get on those stock trigger pins 5/6.

iansane
iansane Dork
2/15/22 10:52 a.m.

Alright, so I tried measuring voltage at pins5/6 that go from the switch to the module that are supposed to activated the engage or disengage function and got absolutely bupkis. Frustrated, I went back inside with the wiring diagram in hand and realized I missed something on the next page. 

And looking at the diagram I guess I understand it. Without activating the switch in any way, the module sees the resistance of 600ohms (I think?) and when the switch is pulled/pushed it bypasses the 450ohm resistor and the module sees only 150ohms. I think I'm out of my element here. I haven't the faintest idea how to activate a relay based on the change in resistance. 

No Time
No Time SuperDork
2/15/22 12:39 p.m.

If I recall correctly the change in resistance is actually different than the 450 ohms because the resistors for apply/release are in parallel. The resistance when the switch is activated is (603*150)/(603+150), and from there you can calculate current and then the voltage drop across each leg. 

But that isn't really important, other than the fact that it appears they use that resistance change to create a differential signal from the switch that provide information on the desired direction. 

How to take that info and run a pair of relays is outside of what I can answer, but I think it might be possible to work with the differential signal to activate the relay, just not as easily as if it was an on/off switch. 

Keith Tanner
Keith Tanner GRM+ Memberand MegaDork
2/15/22 12:42 p.m.

Maybe the change in resistance is so it can also detect no signal or a shorted wire. No resistance = problem.

That's the sort of thing a little microcontroller could handle easily. If you can figure out the draw of the motor at peak (use the fuse on the circuit, perhaps?) I can give some suggestions.

iansane
iansane Dork
2/15/22 1:30 p.m.

That would be most appreciated. I have a small pile of arduinos sitting around I could use. That was my initial plan for a v2 or v3. Switch from a dumb setup to something that has a bit more safeguards.

The the motor and module are on the same 30a fuse.

1 2
Our Preferred Partners
TZw3IlYNmJEIbPO4zXIGtJTykEBDHSu5JaBQovkDiG1ygpze3iR7BFMXGM2LEbag