The basic concept of a laser rangefinder is straightforward: A laser pulse is emitted from the rangefinding device. When the pulse reaches an opaque object, it is reflected back to the sender. The sender detects the reflected pulse and determines the duration between the emitted pulse and the received pulse. The system then uses the speed of light to convert the duration into a distance.
As you might expect, actually implementing a laser rangefinding system is a bit more complicated. The primary difficulty here is the speed of light—which, as you know, is a very large number. Since light travels at 3×108 m/s, it covers 1 m in 3.3 ns. A clock period of 3.3 ns corresponds to a clock frequency of 300 MHz, which is certainly not unreasonable. If we want to resolve something on the order of 0.1 m, though, we’re talking about a clock frequency of 3 GHz, and a digital system clocked at 3 GHz is by no means a trivial design task.
The easiest way to deal with extremely-high-speed clock design is to avoid it altogether, and that’s exactly what a chip like the AS6500 allows you to do. It’s a time-to-digital converter, i.e., a device that is specifically designed to measure the duration between two events and convert this into a binary number.
This diagram conveys the internal architecture of the TDC7200, a time-to-digital converter from Texas Instruments. As pointed out in the TDC7200 datasheet, a time-to-digital converter is essentially a highly sophisticated digital stopwatch.
When a system does not need to have extremely high precision, the time measurement can be based on typical counter functionality: a clock signal drives a digital counter that starts counting when the pulse is emitted and stops counting when the reflected pulse is received.
However, this approach becomes impractical when the goal is to resolve very small distances—the necessary clock frequency is simply too high. The AS6500, for example, has a maximum time resolution of 10 ps. A clock period of 10 ps corresponds to a clock frequency of 100 GHz—I don’t know the details of the internal measurement architecture of this IC, but I’m willing to bet that it does not have a 100 GHz clock driving a 100 GHz–compatible digital counter.
I’m definitely not an expert on high-precision rangefinding techniques, but apparently there are several methods that can be used to overcome limitations associated with excessively high clock frequencies. This paper on picosecond-resolution time-interval measurements mentions the Vernier method, interpolation methods, and tapped delay lines.
The AS6500 from ams
The ams AS6500 is a highly integrated IC that has four separate time-to-digital converters. It requires a single low-frequency reference clock (the acceptable range is 2 MHz to 12.5 MHz), and its measurement data is delivered to a microcontroller or FPGA via SPI.
Diagram taken from the AS6500 product page.
The standard single-shot resolution is 20 ps, though it has a mode that allows for 10 ps resolution. This means that the device supports measurement resolution on the order of 1 cm.
Measurement resolution is certainly an important specification for devices such as this, though you may also have to consider the rate at which distance measurements can be performed. In some applications, such as a basic handheld laser measuring device, there is no need for frequent measurements. However, in other applications, such as object detection and avoidance, the goal is not to make a single distance measurement but rather to create the equivalent of a 3D image based on numerous measurements. The AS6500 can deliver up to 1.5 megasamples per second, and the folks at ams think that it would be a good choice for applications that require high-frequency distance measurements.
This gives you a basic idea of the timing involved in an AS6500 time measurement. Diagram taken from the datasheet.
The AS6500 comes in a QFN40 package that measures 6 mm × 6 mm. A previous part from AMS, the TDC-GPX2, offers similar functionality but in a QFN64 package. This represents a 56% reduction in required PCB real estate, and it seems to me that with the AS6500 you really are getting quite a bit of functionality in a 6 mm × 6 mm form factor.
Do you have any experience with laser rangefinding systems or time-to-digital converters? If so, feel free to leave a comment and share your expertise with other AAC readers.