“High-speed circuit” has become a term often mentioned by electronic engineers today, but what is a high-speed circuit? This is indeed a “familiar” and “fuzzy” concept. In fact, the industry does not have a unified definition of high-speed circuits. Usually, there are many views on the definition of high-speed circuits: Some people think that if the frequency of digital logic circuits reaches or exceeds 45MHZ~50MHZ, and work above this frequency The circuit has already accounted for a certain share of the entire electronic system (for example, 1/3), which is called high-speed circuit; some people think that high-speed circuit and frequency have no big connection, whether high-speed circuit depends only on their rise time; Some people think that high-speed circuits are circuits that we have not touched in a few years, or that can produce and take into account the skin effect; more people have quantified the definition of high-speed, that is, when the digital signal in the circuit is on the transmission line When the delay is greater than 1/2 rise time, it is called a high-speed circuit, and this article also uses this definition as the standard for considering high-speed issues.
In addition, there is another concept of “high frequency circuit” that is prone to confusion. What is the difference between “high frequency” and “high speed”? For high frequency, many people understand the higher signal frequency. Although it cannot be said that this view is wrong, for high-speed electronic design engineers, the understanding should be more profound. In addition to the natural frequency of the signal, we should also consider The impact of higher-order harmonics that are accompanied by signal transmission, generally we use the following formula to define the signal’s emission bandwidth, sometimes called EMI emission bandwidth.
F=1/(Tr*π), F is the frequency (GHz); Tr (nanosecond) refers to the rise time or fall time of the signal.
Usually when F>100MHz, it can be called a high frequency circuit. Therefore, in a digital circuit, whether it is a high-frequency circuit does not depend on the signal frequency, but mainly depends on the rising and falling edges. According to this formula, it can be calculated that when the rise time is less than about 3.185ns, we consider it as a high-frequency circuit.
For most electronic circuit hardware design engineers, there is no need to stick to the difference in concepts. There should be a broad definition of “high speed” in mind, that is: if the circuit is still not stable under the premise of ensuring the correct electrical connection High-performance work requires special layout, wiring, matching, shielding, etc., then this is a “high-speed” design.