Editor’s note: David Swanson proposes a pretty neat concept for the modern door lock in automobiles. Enjoy and get those creative juices flowing.
It is amazing to me. Not so long ago there was no such thing as a BCM (Body Control Module). Lights and other loads were switched on by, of all things, mechanical switches. Higher current loads were sometimes driven by relays that were driven by switches. Protection came in the form of a fuse. A faulted load would blow a fuse. Those niggly momentary shorts would completely disable a system and were virtually impossible to find. As a mechanic, back in the 70’s a momentary fault in the harness was detectable only by the repeated replacement of fuses over time. The fix often resulted in running a new wire from, you guessed it, the switch to the load.
(Packard was the first to successfully introduce electric door locks in 1956)
Relays were distributed according to their need. The high beam relay/switch was located on the steering column (even earlier they were on the floor left of the pedals), the High blower relay was on the firewall near the HVAC blower motor, and the horn relay was under the hood. Even the alternator voltage regulator was a relay on the fender well. The advent of smart switches brought on the ability to centralize the load driving in one box, the BCM. In order to protect the silicon from the hazards of an automotive electrical system, intelligence was needed. Intelligence in the form of overcurrent protection, thermal protection, and inductive energy clamping. Not to mention things like reverse battery and the standard automotive transients. Most of which a relay either generated itself or didn’t care that much about.
In the process of adding all these protections things got quite a bit more sophisticated. Now momentary faults do not kill a system or systems. A shorted load did not take out several other, sometimes seemingly unrelated, things tied to the same fuse.
Now shorts, even momentary faults, are intelligently detected and dealt with by providing fault codes and higher level diagnostics. It is not just nice to have today. It is required… for most loads that is.
In this age, sophisticated electronics safety is less of a feature and more of a requirement.
When it comes to door locks safety and security are becoming more central… European standards require secondary, or Thatcham, locks that provide a double lock of sorts. Thatcham locks prevent interior door handles from opening doors to thwart thieves. Also, European standards require that doors must unlock during an accident to allow first responders faster access. I would think a fairly important safety feature that should focus on reliability and action at all costs.
Electronic child lock features allow the option for the rear seat interior door handles to be disabled while driving. Thus, keeping little ones from accidentally opening doors while on the road.
The introduction of passive entry systems puts pressure on unlock timing. So much so that one car manufacturer has implemented a third lock motor in their door latch mechanism just to speed up the unlock response time when accessing a door handle. They don’t want the customer to have to wait even 100ms to unlock a door. Those few milliseconds of speed are a feature that they are paying for.
Other security issues might include in the future individual door unlock capability so that passive entry can be done one door at a time… one pull of the handle only unlocks that door… keeping all other doors locked. A second pull on that same handle might then unlock the remaining doors or just the doors on that side of the vehicle. Unlocking the fuel door can keep the doors on the opposite side of the car locked for security. Any number of combinations could be considered. All with safety and security in mind.
Thief stealing a pocketbook while the victim is pumping gas
Watch the video here
Door locking can get quite sophisticated and secure. Alas, the cost of a relay is so low these days that such features could be possible except for the fact that relays are loud, heavy, unreliable, and take up a lot of precious circuit board space in the BCM. When you count up all the relays needed to drive the doors individually, the secondary locks front to rear separately, the fuel door, trunk/liftgate etc. the relay solution can get quite onerous.
No, wait! This is the 21st century. Cars have sophisticated body control modules that regulate the RMS voltage in the exterior lamps to extend the life of the bulbs. There is a lot of sophistication to detect the presence of a bulb or an LED array or a shorted or open load. Yet still door locks are driven by pre-1950’s technology. With pre-1950’s reliability… and pre-1950’s performance for what I would consider a safety and security related function. Mmm… what to do?
If there could be a solution. If there was a central door lock driver IC that can do all of this and more in a footprint of a single relay. This device could answer all of the above needs and then some. It could drive the door locks individually, drive the secondary locks individually or by group (front/rear), drive the trunk/rear hatch lock, and drive the fuel door lock. How can a single IC do all this and not be the size of gas cap? Simple, two words, current regulation.
Today door lock drivers (relays) focus their requirements on the maximum amount of current the door lock motors can demand. How demanding are they you ask? Very. Some door lock systems need multiple fuses and can drive as much as 60 Amps. So, the relays, wires, circuit board traces, and connectors that are required to drive these lock motors must be able to handle this monstrous current.
To make a solid-state switch handle this much current takes a bit of silicon. Facts of life, fire is hot, water is wet, and silicon cost money. There is not much money in body electronics. Car makers are reluctantly willing to pay for a feature that adds wow factor to a car or for something the consumer would appreciate enough to pay for… with margin. i.e. there has to be money in it for the car company. Adding cost and sophistication to a feature that is imperceptible to a consumer has two chances of being implemented, slim and none.
So, we can reduce the cost of the silicon by taking advantage of motor theory and at the same time we can provide features that improve reliability and increase safety, security and convenience for the consumer.
DC Motor Theory
Brush type DC motors translate motor current into rotational force or torque. The equation is fully relational; Torque is directly proportional to current. The relationship is described by the following equation:
- T is the motor torque
- N is the number of coil turns
- P is the number of poles
- φ is the flux
- I is the current in the motor
Note that all of the parameters are virtually fixed. The number of windings, the number of poles and the flux are motor parameters that are governed by the construction of the motor itself. That means that we can simplify the equation to:
T = k x I
Where k is a constant. The conclusion that can be made is that regardless of temperature or voltage or motor speed, the torque in the motor is solely governed by the current in the motor.
To be sure, the torque in the motor at the lowest operating voltage and highest ambient temperature is sufficient for actuating the lock mechanism. So, the trick is to provide that current regardless of battery voltage or ambient temperature.
Automotive charging systems charge the battery at different voltages depending on the ambient temperature. That voltage varies from as low as 12.5V at hot to close to 16V at extreme cold.
Typical battery voltage over temperature for one car manufacturer.
Interestingly enough, the motor winding resistance works in the opposite direction. At cold, the resistance due to the copper windings is the lowest making the stall current (something a relay would be required to handle) the highest. That very high stall current makes for a very high motor torque. As a result, there is a lot of unnecessary stress on a lock mechanism at low temperatures when driving the lock motors with a relay.
At high temperature, the run current of the motor is limited due to the lower battery voltage and the higher winding resistance. Fortunately for us that current is a fraction of the worst-case stall current at cold.
So, now we have shifted from a lock motor driver requirement based on what the motor can demand to a lock motor requirement based on what the lock motor needs to actuate. What might have been a 60 Amp requirement is reduced to something around 12 Amps. Also, taking into account that door lock actuation takes somewhere between 100ms and 300ms this becomes much more affordable.
How it could be done
To provide a system where one side of the lock motors is served by a simple half bridge and the other side is served by current regulated half bridges. For the most basic North American three relay circuit it would look something like Figure 4:
Simple current regulated circuit
Here, one current regulated output drives the Driver’s door lock and the other current regulated output drives the other three lock motors in parallel. The first thing that this schematic brings our attention to is the paralleled motors. What happens if one or two motors are already in stall when actuated?
How well can motors share current?
Let’s examine the case where one lock motor unlocked and the other two locked. The node that shares four motors is unregulated (just a switch). A lock function will see two stalled motors and one (hopefully) running motor. The concern is that in a heavily current regulated system with three motors in parallel the two stalled motors will hog all of the current keeping the unlocked motor from moving.
However, when the system is first actuated, until the unlocked motor begins to move, all three motors are in stall. As a result, the current is shared quite nicely getting the unlocked motor to move. It is only after the motor begins to move that the current to it can be siphoned off by the other two stalled motors. The “siphoning” of the regulated current by the stalled motors is proportional to the speed of the moving motor. As a result, the unlocked motor moves a tad slower. Still well within the timing requirements.
Figure 5 below illustrates the currents shared by (in this case) four motors where only one is not in stall. The magenta shows the stalled motor total current and the green shows the current in the un-stalled motor. Even with this worst case study the motor still moves.
Sharing currents in parallel motors
This example shows a heavily regulated 4A total current system. The stall current each motor wants to see is ~2A each. We are regulating to half of that in this example. The four motors are then sharing the limited current between them. Three stalled motors can easily take all of the regulated current if the fourth motor was not present. In the end, the run time for the un-stalled motor in this heavily regulated system is still less than half of the system specification to fully actuate.
A Universal Door Lock Solution
Here is where we can make a difference to the consumer both in convenience and in safety/security. There could be a universal door lock IC designed for central door locking. It could have, let’s say, six current regulated outputs and two external Half Bridge MOSFET controllers. And it could look something like this:
Example Application diagram for a Universal Door Lock IC
To make this a worthwhile endeavor we would want to make something that would add value to the consumer and to the engineer. Something that would bring distinction to a car manufacturer without too much pain.
For the consumer:
For the consumer reliability is always at the top of the list. A door lock driver that minimizes the stress on the entire lock system, from fuse to the mechanical bits, it is easy to see that reliability is improved. If we can improve reliability while enhancing safety and security as well as have added convenience, then we have definitely have made a “better mousetrap”.
- Safety is an important issue with respect to door locks. Malfunctions at the worst possible moment can cost lives or loss or property.
- In emergency mode, door lock actuation can continue even when harness integrity is in question. A relay will blow a fuse if actuation occurs when a harness is in the process of being severed as in a severe accident. This solution will continue to retry applying an unlock command working to unlock the remaining doors without blowing a fuse.
- During an accident, speed is of the essence. A relay takes ~10+ milliseconds to begin to actuate. That is 10 milliseconds lost at the worst possible moment. Solid state solutions are effectively instant in actuation, buying back those precious moments.
- In minivans where door lock current is not present unless the sliding door contacts are touching, this IC can determine of a minivan door lock was properly actuated.
- In Locking doors from a key fob and one door does not lock, a warning can be provided telling the consumer that a door or doors are not locked.
- In a passive entry system unlocking a single door whether it be a driver’s door or one of the other passenger doors makes for a much more secure vehicle. A consumer can now put their groceries in the back seat seamlessly without having to unlock the entire car.
- Having individual door lock drivers provides flexibility, security, and convenience.
- The increased speed that a door lock actuator can be actuated compared to a relay improves the passive entry experience.
- Electronic child locks allow parents to automatically disable interior rear door handles while the car is in motion to help keep little ones safe.
- Momentary shorts in the harness do not disable the system by blowing a fuse. A momentary short, for the most part, can be both virtually invisible to the consumer as well as reportable to the mechanic. This can potentially (silently) identify and fix a problem before it is even noticed by the consumer as an issue.
Safety, security, reliability, and convenience. Features that help make a car maker stand out from the rest.
For the engineer:
The idea would be to make this fully programmable and fully automatic. Lock time durations and dynamic braking times could be fully programmable from 100ms to indefinite. The current regulation programming could be set at 200mA steps from 1A to 4A for each output. Outputs could be driven singly or in groups of two or three for up to 12 A per output group. This would allow for extreme flexibility and usability for any number of door lock configurations. There could even be PWM frequency programmability with or without dithering for improved EMC.
Actuation could be initiated either by a single SPI command or by a rising edge on an external pin for increased security.
Since this solution would have current regulation digitally controlled that current information could be available on the SPI bus. With that, load integrity could be discerned almost for free. If for nothing else that information could be used at least for enhanced load integrity checks at vehicle assembly end of line.
The external MOSFETs could easily be protected by programmable drain-source voltage monitoring. The internal, current regulated, half bridges would have more protection in that they could be protected by overcurrent and programmable thermal protection.
This would make for an extremely flexible and powerful solution taking a little space as a single relay, freeing up micro ports and reducing overall circuit board burden.
We are not in the 1950’s. Our door locking systems shouldn’t be either.