Ceci n’est pas une pipe
In this post, we shall spend some time exploring the topic of brake by wire, and look at the various flavours of decoupled brake systems already in the market, as well as the prospects for Brake-by-Wire using electrical energy as the energy transmission system. We will also discuss the advantages of such a system, as well as some of the trade-offs.
Peeling the onion…
When we talk about Brake-by-Wire, what we are really considering is a change in brake actuation architecture, that is, how we take a (human or robot) driver brake request, and translate that into deceleration. When you first consider this, most likely you’d imagine a brake request as a pedal being pressed, and deceleration resulting from a disc and pad (or drum and shoe) being pressed together. And for an awful lot of braking, that is correct.
But with the advent of terms like Single Pedal Driving, Remote Parking Assist and Traffic Sign Recognition, the topic of Brake-by-Wire becomes a good bit broader. Brake-by-Wire could also be brake by Wi-Fi, brake by geolocation or brake by tweet. (OK, hopefully not the last one, but you get the idea). So we need to stratify and refine the topic somewhat.
Firstly, when we talk about by-Wire, we should agree that a primary facet of a by-Wire system is the ability to execute a variety of brake responses to a given brake request. In the main use case today, we’re therefore talking about a system which can vary the deceleration response to a driver’s pedal input. This might mean offering a variety of pedal characteristics, or overlaying additional brake pressure in a panic situation, or modulating applied brake pressure for wheel slip control or indeed generating some deceleration blend in response to a driver lifting off the accelerator (the Single Pedal Driving characteristic that is popular in some EVs). For all these, there is still a direct hydraulic connection between the brake pedal and foundation brake, but these overlays or offsets are generated by a control mechanism and actuated by electronic means.
But Brake-by-Wire can also mean a hydraulic disconnect between two complimentary circuits – the typical operational mode of a modern 1 Box system. This means that in normal operational conditions, a driver pedal input is fed into a pedal feel simulator, made up of a piston working against a spring in dedicated chamber. The driver input is recreated in the second circuit through electronic interpretation, and a pump or plunger is used to generate the required pressure. In certain safety scenarios the circuits can be brought together, so the driver is directly hydraulically connected to the foundation brakes.
This system is increasingly common among mass market vehicles as the single unit can incorporate the functions of a brake booster, brake modulation and volume blending for regenerative braking. It sacrifices some brake pedal feel granularity, but provides a pretty good solution for the vast majority of drivers in the vast majority of cases while saving weight and cost.
The 1 Box system can also be modified in a number of ways to bring about a more ambitious Brake-by-Wire layout. A second unit built into the vehicle can provide a back-up brake controller for Autonomous Driving. In this case, the unit will dispense with the physical driver interface, and maintain the pump, reservoir and electrical connections. The unit must also have a discrete pedal travel sensor to interrogate driver requests. This can even be combined with a redundant power source to bring an added layer of back-up to a vehicle. This combination of hydraulic elements connected through electronic means is perhaps the truest Brake-by-Wire in popular use today, but it is still primarily a hydraulic system, albeit augmented with electrical brains and muscle.
Down to the wire
To move beyond that, and remove the hydraulic element means there must be an alternative method to transmit braking effort from the driver to the brakes. While we will focus on electric systems further below, it’s worth first taking a wider view. At this point (if not only to placate the many motorcyclists who have suffered in silence to now) we must point out that cable-operated brakes are still used in the vast majority of motorbikes and of course bicycles. So in that sense, most vehicles on the planet are already Brake-by-Wire, and have been for some time. And if we remember back in our brakes history, we’ll see that Brake-by-Long-Handle was replaced by cable-pull brakes late in the 19th Century as the performance braking set-up of choice, regardless of what was decelerating the wheels. Hydraulic systems became popular as vehicles got heavier and faster, but cables were still in use for parking brakes until the rise of electronic parking brakes in the last decades.
And so to the idea of a fully electric connection between driver’s foot and wheel. Here, we remove any hydraulic element from the vehicle, and instead use electromechanical systems to create braking force. In terms of brake hardware, that will mean replacing pistons in our brake callipers with some manner of actuator and power electronics. While a number of competing ideas are vying for acceptance in this area, one of the most promising developments to date is known as Smart Brake from the Hitachi Automotive Systems Group (developed by Chassis Brakes International, before acquisition by Hitachi Automotive Systems). The system incorporates an ECU at each wheel, a motor to control the pad movement, a wheel speed sensor to achieve modulation and control tasks. The advantages offered by the system include lighter and simpler cabin components, lower energy consumption, better pad control and faster response dynamics. This last point is of particular importance in emergency braking scenarios, where the so-called Time-to-Lock (TTL) can be decisive in avoiding accidents.
In order to dispense fully with either cables or hydraulics, however, we also have to talk about redundancy. For a hydraulic system, including the fancy by-Wire systems we looked at earlier, the system can handle a wide variety of faults in the electrical systems by relying on a fall-back hydraulic connection between the driver and brake. Even a full loss of electrical power doesn’t prevent the driver from decelerating the vehicle. Of course, a fault in the hydraulic system is possible, so circuits are split from the master cylinder, such that even with a leak in one circuit, the remaining circuit can still achieve a minimum (mandated) level of deceleration. Like many conventions in automotive, we’ve done it this way for a good reason, and it may take a long time before the engineering teams want to move beyond this solution.
But to achieve the equivalent with an electromechanical brake system, we need to recreate this dual circuit approach, so that a failure of one doesn’t hamper the other. A similar requirement exists for steering systems, whereby to remove the mechanical connection between driver and steering gear, a redundant transmission circuit is required – effectively two independent power supply systems. This means that both safety systems would come from a shared power source, and in the event of constrained supply, steering systems must be given priority. Where we can envisage these systems being most relevant is in those autonomous vehicles which dispense with driver controls altogether. And as we saw previously, these systems are already well advanced and in everyday use in certain parts of the world. They promise to reinvent many aspects of how we interact with mobility, but will be built upon the building blocks of what has come before.
The future is (still) coming…
Looking through a slightly longer lens, the story of an electromechanical brake in the modern sense is already a number of decades old. In researching this article, I came across a report focusing on early efforts at Continental, as well as an op-ed wondering whether Bosch would still be a relevant player in the coming age of electromechanical braking, and a heady prediction from Siemens VDO from 2006, foretelling of a production offering before the end of the decade.