“We can see that digital isolators use transformers or capacitors to magnetically or capacitively couple data across the isolation barrier, and optocouplers use light from LEDs. This makes the quiescent current, data transmission delay, CMTI and other performance of the optocoupler scheme poor. Magnetic isolation is the use of transformer current pulses through one coil to form a small local magnetic field, thereby generating an induced current in the other coil.
Author: Tan, Yuan
Various types of isolators are topics that we often talk about when we are designing systems. This article will introduce the following three aspects:
1. Why is isolation necessary?
2. What is the difference between different isolation technologies?
3. What are the selection parameters of isolation devices?
Why is isolation necessary?
The answer is that isolation is about reliable protection. Galvanic isolation is a circuit design technique that allows two circuits to communicate, eliminating any unwanted direct current flowing between them.
Isolation is often used to:
• Protect operators and low voltage circuits from high voltages.
• Prevent ground potential differences between communication subsystems.
• Improve noise immunity.
Figure 1 Isolation Blocks unwanted DC and AC currents across a dielectric barrier
How are different isolation techniques different?
The three elements of insulation technology are: insulation material, structure and data transmission method. Designers introduce isolation to meet safety regulations or to reduce ground loop noise, for example. Galvanic isolation ensures that data transmission is not through electrical connections or leakage paths, thus avoiding safety risks. However, isolation imposes constraints on latency, power consumption, cost, and size. The goal of digital isolators is to meet safety requirements while minimizing adverse effects.
Let’s first look at the impact of insulating materials:
Table 1 Common insulating materials
This table lists common insulating materials and their corresponding isolation capabilities (Vrms/um). The requirements of the optocoupler device on the light transmittance of the material make air and epoxy resins are mostly used as isolation materials. Polyimide (polymide) is mostly used for magnetic isolation products. The capacitor and isolation devices are made of silicon dioxide (SiO2) material with the best dielectric strength.
The selection of isolation materials will affect the volume of the device. The following is a comparison chart of real shots. Comparing TI’s ISO series of capacitive isolation products and the common optocoupler isolation devices on the market, it can be seen that the capacitive isolation devices have obvious volume advantages.
At the same time, compared to polymide, the reliability of silicon dioxide (SiO2) will not be affected by working in a humid environment.
Let’s look at the circuit structure again. The circuit structure of the three isolation methods is as follows:
Figure 3.1 Optocoupler device circuit
Figure 3.2 Magnetic Isolation Device Circuit
Figure 3.3 Capacitor isolation device circuit
We can see that digital isolators use transformers or capacitors to magnetically or capacitively couple data across the isolation barrier, and optocouplers use light from LEDs. This makes the quiescent current, data transmission delay, CMTI and other performance of the optocoupler scheme poor. Magnetic isolation is the use of transformer current pulses through one coil to form a small local magnetic field, thereby generating an induced current in the other coil. The principle of electromagnetic induction makes the magnetic isolation products more likely to have noise interference in poor electromagnetic working environment. The capacitive isolation product uses a low-current electric field to couple the data to the other end of the isolation barrier, which is more anti-interference.
Finally, we look at the impact of the way the data is transmitted. The optocoupler uses the light from the LED to transmit the data to the other side of the isolation barrier: when the LED is on, it represents a logic high level, and when it is off, it represents a logic low level. Optocouplers consume power when the LED is lit; for applications where power consumption is a concern, optocouplers are not a good choice.
The data transmission method commonly used in magnetic isolation products is to encode the rising and falling edges into double pulses or single pulses to drive the transformer. These pulses are decoded as rising or falling edges on the secondary side. The power consumption of this method is 10 times to 100 times lower than that of optocouplers because, unlike optocouplers, there is no need to continuously supply power to the device. This also explains why the power consumption of magnetically isolated products is linear with the data rate.
TI’s isolation products mostly use on-off keying (OOK) modulation. The transmitter sends a high-frequency signal to represent one digital state, and does not send a signal to represent another digital state. After signal conditioning, it is sent out through the buffer. The power consumption of the isolation device basically does not change with the change of the transmission data rate.
Figure 4 Conceptual block diagram of a capacitive isolation device
Figure 5 ON-OFF keying (OOK) modulation
What are the isolation selection parameters?
Here is a list of parameters that you often see in data sheets and explanations. For more detailed parameter meanings, please refer to this link: http://www.ti.com/lit/pdf/slyy063
Table 2 Common parameters of isolation device selection
Instead of expanding them one by one, let’s focus on the meaning of three easily confusing distance parameters: creepage, clearance and DIT (Distance Through Insulation). As shown in the figure below, clearance is the shortest distance between the pins on both sides through the air, creepage is the shortest distance between the pins on both sides through the surface of the insulating material, and DTI is the shortest distance through the insulating filler material between the conductors. It can be said that DIT refers to the internal distance and creepage and clearance refer to the external distance.
Figure 6 Schematic diagram of clearance, creepage, DTI (sequentially)
It should be reminded that the target chip of the creepage distance parameter of the data sheet, if there is a shorter path in the system, then this shorter distance is the isolation creepage distance of this circuit. Here is an example. The following figure shows two ultra-wide packages ISO7841DWW (creepage is 14mm) working in series. Can it be understood that the creepage distance of the circuit is 28mm? The answer is no. This is because in addition to the data transmission path, there is also a path in the circuit, that is, the isolated power supply circuit. The creepage distance between VCC1 and Viso2 may be shorter than the creepage distance between D1 and Diso2, and the shorter distance value determines the creepage distance value of the circuit.
Figure 7 Creepage distance circuit example
This article starts with three frequently asked questions:
1. The use of isolation devices, discussing isolation and reliable protection.
2. Mainstream isolation technologies include differences in optical isolation, magnetic isolation, and capacitive isolation and the reasons behind them. The emphasis is on the description and comparison of insulating materials, structures, and data transmission methods.
3. The main parameters, meanings and precautions of isolation device selection, with emphasis on the meaning of creepage distance.
TI digital isolator star product selection:
TI’s digital isolator products use capacitive isolation technology, which is cost-effective and has a complete range of products that can cover many requirements for isolation levels, channel requirements and creepage distances.
Here are the ISO77xx and ISO78xx series selection tables for easy reference:
Figure 8 TI’s ISO77xx and ISO78xx selection comparison table