electric brake booster?

[ericjon262]

I had looked into this a few years ago, and didn't see anything about the Tesla/Accord electric brake booster, today on OE someone posted a link to this video:

https://youtu.be/SRZ8XDNz2vU

the video outlines how the booster works, and how to install it in a vehicle. from a form factor standpoint, it doesn't appear to be any larger or smaller than a vacuum booster, weight is still an unknown factor.

Any thoughts on using one of these? I found them online for ~$170 seems reasonable enough, and should perform consistently regardless of engine vacuum, which could be important for a car with a rowdy cam.

[Emc209i]

There's beauty in simplicity. The vacuum powered diaphragms are just that. I've been around several hydraulic and electric powered boosters, they work well, but they're a pain in the ass when they need service. Our hybrid Lexus uses its 24volt circuit to power a bunch of accessories including the ABS when the engine shuts off at stop lights. It works, and I've not had a problem with it so far (omg finger's crossed), but I've seen several videos of people hunting down problem after problem after problem when that system starts to go. Funny enough the brake booster in that car remains vacuum - I guess it's got enough residual in the vacuum canister to go for a while.

Turbo cams don't usually tend to be that unruly. Your idle is that rough?

[pmbrunelle]

It's good that the booster is a Bosch part, and not Tesla. Hopefully it becomes a generic part that is found everywhere, such as the ZF transmissions. The fact that it can be found in the Honda is a good sign.

While the wiring diagram in the video is simple, I would bet that there is also the possibility to connect it to a CAN bus. Then, in an electric car, the booster assist could be reduced as the regen comes on. The unit may also broadcast the pedal force value on the CAN bus, so the car knows how hard the driver is pressing on the pedal, and how much regen to apply.

In case of CAN bus failure (or absence, as in the case of a hot rod), it looks like the booster simply falls back to its "safe state" of simplified operation based only on the sensor.

The sensor appears to be a Melexis or similar:
https://www.melexis.com/en/products/mag ... sensor-ics

I think I would trust the Melexis, we have used them at work. The package with many pins suggests that it is a dual-die redundant version.

I am kind of baffled that Teslas come with huge disc brakes and calipers. Looks backwards. I thought the idea of electric cars is that they depend on regenerative braking, not friction. They should have small vestigial friction brakes if they want to look like the car of the future.

For an everyday A-B car (with performance/fade not being a primary concern), I think I would prefer 4-wheel drum brakes. With the self-energizing effect that can be built into drum brakes, it is possible to avoid the brake booster altogether while keeping brake force reasonable.

If you want to use a different master cylinder, you probably want to make sure that the Bosch screw has enough travel to push through the entire stroke of your selected master cylinder.

The booster is a closed-loop control system. If you change the spring to increase the gain of the booster (as suggested in the video), you may cause the booster to oscillate. You can increase the gain to the oscillation point, then back off a bit.

[Emc209i]

Regenerative braking doesn't stop a vehicle, it causes parasitic drag, but it doesn't stop the car. I come off the highway sometimes and shift into full regenerative braking mode to speed charge the batteries. If I didn't apply the brakes the vehicle would coast for a mile.

Tesla's are not Prius'. People who buy Tesla's like to stop light drag them because of how much torque they generate. The car is a performance vehicle and requires large brakes. It's not uncommon to drive a stock Tesla on the Nürburgring.

[pmbrunelle]

That Teslas are (optionally) AWD and powerful makes them good candidates for strong regen braking.

In theory, whatever current you force through the electric motor in the forward direction, you can also force through the motor in the reverse direction.

In the forward direction, gearbox inefficiency reduces wheel power relative to motor power. When braking, gearbox loss is actually a plus! The loss in the gearbox means that wheel power is greater than motor power.

In a Tesla with strong acceleration, that implies the potential for pretty strong regen braking. A lot more than a parasitic slowdown.

I don't think that Teslas have the motor power to break all four tires loose at highway speeds, so they would still need some kind of friction brake to make up for the deficit. However, I think that the role of the friction brake could be reduced, even considering the cars as semi-sports vehicles.

[Emc209i]

That's not how they work.

When leisurely cruising, you can use the regen to effectively brake, but during high speed and emergency braking, those big brakes are necessary. You might remember a few years ago the model 3 had the worst braking of any car in its segment.

[ericjon262]

There's beauty in simplicity. The vacuum powered diaphragms are just that. I've been around several hydraulic and electric powered boosters, they work well, but they're a pain in the ass when they need service. Our hybrid Lexus uses its 24volt circuit to power a bunch of accessories including the ABS when the engine shuts off at stop lights. It works, and I've not had a problem with it so far (omg finger's crossed), but I've seen several videos of people hunting down problem after problem after problem when that system starts to go. Funny enough the brake booster in that car remains vacuum - I guess it's got enough residual in the vacuum canister to go for a while.

Turbo cams don't usually tend to be that unruly. Your idle is that rough?

I wasn't necessarily meaning to post this as something I planned to do to my car, just something I found interesting and worth sharing. One day I may adapt it, but it's way down on the priority list. the video actually claims that early Teslas use a vacuum pump to run a standard booster.

It's good that the booster is a Bosch part, and not Tesla. Hopefully it becomes a generic part that is found everywhere, such as the ZF transmissions. The fact that it can be found in the Honda is a good sign.

While the wiring diagram in the video is simple, I would bet that there is also the possibility to connect it to a CAN bus. Then, in an electric car, the booster assist could be reduced as the regen comes on. The unit may also broadcast the pedal force value on the CAN bus, so the car knows how hard the driver is pressing on the pedal, and how much regen to apply.

In case of CAN bus failure (or absence, as in the case of a hot rod), it looks like the booster simply falls back to its "safe state" of simplified operation based only on the sensor.

The sensor appears to be a Melexis or similar:
https://www.melexis.com/en/products/mag ... sensor-ics

I think I would trust the Melexis, we have used them at work. The package with many pins suggests that it is a dual-die redundant version.

I am kind of baffled that Teslas come with huge disc brakes and calipers. Looks backwards. I thought the idea of electric cars is that they depend on regenerative braking, not friction. They should have small vestigial friction brakes if they want to look like the car of the future.

For an everyday A-B car (with performance/fade not being a primary concern), I think I would prefer 4-wheel drum brakes. With the self-energizing effect that can be built into drum brakes, it is possible to avoid the brake booster altogether while keeping brake force reasonable.

If you want to use a different master cylinder, you probably want to make sure that the Bosch screw has enough travel to push through the entire stroke of your selected master cylinder.

The booster is a closed-loop control system. If you change the spring to increase the gain of the booster (as suggested in the video), you may cause the booster to oscillate. You can increase the gain to the oscillation point, then back off a bit.

I don't care for four wheel drums, I've driven just about every brake system combination, and and disc/disc stops best, and is easiest to maintain IMO. if I did this, I would start with the system as is, I imagine it would be more than enough for a Fiero, and hopefully not too much, but it would really depend on the master cylinder too.

it would be interesting to see what some of the other pins could be used for, there may be a 0-5v output for force monitoring?

[pmbrunelle]

That's not how they work.

I was talking about how things could be, with sufficient development effort. Not how things are.

it would be interesting to see what some of the other pins could be used for, there may be a 0-5v output for force monitoring?

If you can find a service manual for the Accord, wiring diagrams there might give some hints. Otherwise, poke around with a multimeter.

For force monitoring, probably the most likely chance of success would be to tap into the 4-wire sensor harness. If you're lucky, two of the wires will have an analog voltage or PWM; easy enough to interface to the ECU for datalogging.

If you're less lucky, the wires may carry the information digitally with some protocol such as SENT. It would not be impossible to decode (especially if the markings on the sensor chip allow for it to be identified, and for its datasheet to be found), but this would be difficult for the average garage guy without a background in this sort of thing.

On the main connector there is most likely a CAN bus. I think the MegaSquirt might be able to receive CAN messages (one possibly containing brake pedal force), which could then be datalogged... assuming that you can make sense of the CAN message.

[ericjon262]

it would be interesting to see what some of the other pins could be used for, there may be a 0-5v output for force monitoring?

If you can find a service manual for the Accord, wiring diagrams there might give some hints. Otherwise, poke around with a multimeter.

For force monitoring, probably the most likely chance of success would be to tap into the 4-wire sensor harness. If you're lucky, two of the wires will have an analog voltage or PWM; easy enough to interface to the ECU for datalogging.

If you're less lucky, the wires may carry the information digitally with some protocol such as SENT. It would not be impossible to decode (especially if the markings on the sensor chip allow for it to be identified, and for its datasheet to be found), but this would be difficult for the average garage guy without a background in this sort of thing.

On the main connector there is most likely a CAN bus. I think the MegaSquirt might be able to receive CAN messages (one possibly containing brake pedal force), which could then be datalogged... assuming that you can make sense of the CAN message.

if I go any further with the electric booster, I will definitely look into a service manual for the vehicle using it.

[The Dark Side of Will]

Regenerative braking doesn't stop a vehicle, it causes parasitic drag, but it doesn't stop the car. I come off the highway sometimes and shift into full regenerative braking mode to speed charge the batteries. If I didn't apply the brakes the vehicle would coast for a mile.

Tesla's are not Prius'. People who buy Tesla's like to stop light drag them because of how much torque they generate. The car is a performance vehicle and requires large brakes. It's not uncommon to drive a stock Tesla on the Nürburgring.

That Teslas are (optionally) AWD and powerful makes them good candidates for strong regen braking.

That's not how they work.

When leisurely cruising, you can use the regen to effectively brake, but during high speed and emergency braking, those big brakes are necessary. You might remember a few years ago the model 3 had the worst braking of any car in its segment.

Teslas are really heavy and really fast, so they need big brakes to back up the regen. I wouldn't have said this a couple of years ago, but a Model S Plaid is starting to look like my kind of lead sled.

The bottleneck on regen is not motor generating capacity, but rather battery charge current limitations. With a rack of resistors on the roof as with a diesel electric locomotive, a Tesla could sink a heck of a lot more regen current and the regen braking performance could approach the acceleration performance as Pat suggests. Clearly dumping regen current as heat is not smart for a BEV. Concurrently, pushing that much current to the battery is very hard on the battery and dramatically reduces battery life. High current Supercharging reduces battery life more quickly than slower charging, but most Tesla owners don't Supercharge frequently enough for that to matter. A BEV can have dozens or hundreds of regen events per charge cycle, so high current regen would suck the life out of the battery quickly.

It would be interesting to pair a storage mechanism like a flywheel or super-capacitor which CAN handle high charge currents as an intermediary between the traction motors and the battery... but that's adding weight and complexity to a vehicle that's already heavy and expensive.

I don't know about Teslas and Prii, but the Chevy Bolt had a hill-hold function that would use traction motor torque to keep the vehicle stationary on a hill when the driver isn't giving any input.

//

As far as boosters go, if your priority is light weight, stick with a vacuum booster. Note that the video doesn't mention how much the electric booster weighs. You also need to make sure you have enough alternator capacity to keep up with your brake use. If you're building a track car with an electric booster, you'll need a bigger alternator to keep up with the higher duty cycle on that 30 amp motor vs. building a cruiser with a very low duty cycle of high torque from the booster motor.

[pmbrunelle]

Teslas are really heavy and really fast, so they need big brakes to back up the regen. I wouldn't have said this a couple of years ago, but a Model S Plaid is starting to look like my kind of lead sled.

The bottleneck on regen is not motor generating capacity, but rather battery charge current limitations. With a rack of resistors on the roof as with a diesel electric locomotive, a Tesla could sink a heck of a lot more regen current and the regen braking performance could approach the acceleration performance as Pat suggests. Clearly dumping regen current as heat is not smart for a BEV. Concurrently, pushing that much current to the battery is very hard on the battery and dramatically reduces battery life. High current Supercharging reduces battery life more quickly than slower charging, but most Tesla owners don't Supercharge frequently enough for that to matter. A BEV can have dozens or hundreds of regen events per charge cycle, so high current regen would suck the life out of the battery quickly.

It would be interesting to pair a storage mechanism like a flywheel or super-capacitor which CAN handle high charge currents as an intermediary between the traction motors and the battery... but that's adding weight and complexity to a vehicle that's already heavy and expensive.

I don't know about Teslas and Prii, but the Chevy Bolt had a hill-hold function that would use traction motor torque to keep the vehicle stationary on a hill when the driver isn't giving any input.

I don't know if charging with regen is any harder on the battery than discharging during "WOT" (what do we call that on an electric car, or diesel for that matter?).

Or, do the batteries actually have some sort of asymmetry between charging and discharging? Should probably edjumacate myself.

If there's nowhere for the regen current to go, I'd still rather see waste heat be dumped in resistors than in the large rotating and unsprung weight of brake rotors, which also require large wheels. However, large wheels might be considered a styling requirement anyway, and not a consequence of big brakes.

As far as boosters go, if your priority is light weight, stick with a vacuum booster. Note that the video doesn't mention how much the electric booster weighs. You also need to make sure you have enough alternator capacity to keep up with your brake use. If you're building a track car with an electric booster, you'll need a bigger alternator to keep up with the higher duty cycle on that 30 amp motor vs. building a cruiser with a very low duty cycle of high torque from the booster motor.

In the video we see that the Bosch booster is using a high-pitch backdriveable screw, surely to allow for the system to revert to unassisted hydraulic brakes in case of booster failure.

However, there is a downside to the backdriveable screw. When you are sitting still in traffic with your foot on the brake pedal, the electric motor of the booster will have to apply a constant torque, and hence draw current non-stop.

The vacuum booster only affects the engine when you move the brake pedal, not when the pedal is held stationary.

[The Dark Side of Will]

Teslas are really heavy and really fast, so they need big brakes to back up the regen. I wouldn't have said this a couple of years ago, but a Model S Plaid is starting to look like my kind of lead sled.

The bottleneck on regen is not motor generating capacity, but rather battery charge current limitations. With a rack of resistors on the roof as with a diesel electric locomotive, a Tesla could sink a heck of a lot more regen current and the regen braking performance could approach the acceleration performance as Pat suggests. Clearly dumping regen current as heat is not smart for a BEV. Concurrently, pushing that much current to the battery is very hard on the battery and dramatically reduces battery life. High current Supercharging reduces battery life more quickly than slower charging, but most Tesla owners don't Supercharge frequently enough for that to matter. A BEV can have dozens or hundreds of regen events per charge cycle, so high current regen would suck the life out of the battery quickly.

It would be interesting to pair a storage mechanism like a flywheel or super-capacitor which CAN handle high charge currents as an intermediary between the traction motors and the battery... but that's adding weight and complexity to a vehicle that's already heavy and expensive.

I don't know about Teslas and Prii, but the Chevy Bolt had a hill-hold function that would use traction motor torque to keep the vehicle stationary on a hill when the driver isn't giving any input.

I don't know if charging with regen is any harder on the battery than discharging during "WOT" (what do we call that on an electric car, or diesel for that matter?).

Or, do the batteries actually have some sort of asymmetry between charging and discharging? Should probably edjumacate myself.

If there's nowhere for the regen current to go, I'd still rather see waste heat be dumped in resistors than in the large rotating and unsprung weight of brake rotors, which also require large wheels. However, large wheels might be considered a styling requirement anyway, and not a consequence of big brakes.

My understanding is that charging and discharging have huge asymmetries.

As far as boosters go, if your priority is light weight, stick with a vacuum booster. Note that the video doesn't mention how much the electric booster weighs. You also need to make sure you have enough alternator capacity to keep up with your brake use. If you're building a track car with an electric booster, you'll need a bigger alternator to keep up with the higher duty cycle on that 30 amp motor vs. building a cruiser with a very low duty cycle of high torque from the booster motor.

In the video we see that the Bosch booster is using a high-pitch backdriveable screw, surely to allow for the system to revert to unassisted hydraulic brakes in case of booster failure.

However, there is a downside to the backdriveable screw. When you are sitting still in traffic with your foot on the brake pedal, the electric motor of the booster will have to apply a constant torque, and hence draw current non-stop.

The vacuum booster only affects the engine when you move the brake pedal, not when the pedal is held stationary.

The motor must be PWM controlled, so even if it pulls 20-30A at max assist, it should pull an amp or less at light brake application.

When you're blasting down the front straight and smash the brakes for Turn 1, the engine is at high RPM overrun making so much vacuum it doesn't even notice vacuum booster application. Of course that's a different situation than idling in traffic. Brake boost at idle is way less important with a manual transmission than it is with an automatic transmission.