10

ISO124

SBOS074E – SEPTEMBER 1997 – REVISED JUNE 2018

www.ti.com

Product Folder Links: ISO124

Submit Documentation Feedback

Copyright © 1997–2018, Texas Instruments Incorporated

8 Application and Implementation

NOTE

Information in the following applications sections is not part of the TI component

specification, and TI does not warrant its accuracy or completeness. TI’s customers are

responsible for determining suitability of components for their purposes. Customers should

validate and test their design implementation to confirm system functionality.

8.1 Application Information

8.1.1 Carrier Frequency Considerations

The ISO124 amplifier transmits the signal across the isolation barrier by a 500-kHz duty-cycle modulation

technique. For input signals having frequencies below 250 kHz, this system works like any linear amplifier. But

for frequencies above 250 kHz, the behavior is similar to that of a sampling amplifier. Figure 9 shows this

behavior graphically; at input frequencies above 250 kHz, the device generates an output signal component of

reduced magnitude at a frequency below 250 kHz. This is the aliasing effect of sampling at frequencies less than

two times the signal frequency (the Nyquist frequency). At the carrier frequency and its harmonics, both the

frequency and amplitude of the aliasing go to zero.

8.1.2 Isolation Mode Voltage Induced Errors

IMV can induce errors at the output as indicated by the plots of IMV vs Frequency. It should be noted that if the

IMV frequency exceeds 250 kHz, the output also will display spurious outputs (aliasing) in a manner similar to

that for VIN > 250 kHz and the amplifier response will be identical to that shown in Figure 9.This occurs because

IMV-induced errors behave like input-referred error signals. To predict the total error, divide the isolation voltage

by the IMR shown in Figure 11 and compute the amplifier response to this input-referred error signal from the

data shown in Figure 9. For example, if a 800-kHz 1000-Vrms IMR is present, then a total of [(–60 dB) + (–30

dB)] x (1000 V) = 32-mV error signal at 200 kHz plus a 1-V, 800-kHz error signal will be present at the output.

8.1.3 High IMV dV/dt Errors

As the IMV frequency increases and the dV/dt exceeds 1000 Vµs, the sense amp may start to false trigger, and

the output will display spurious errors. The common-mode current being sent across the barrier by the high slew

rate is the cause of the false triggering of the sense amplifier. Lowering the power-supply voltages below ±15 V

may decrease the dV/dt to 500 V/M s for typical performance.

8.1.4 High Voltage Testing

TI has adopted a partial discharge test criterion that conforms to the German VDE0884 Optocoupler Standards.

This method requires the measurement of minute current pulses (< 5 pC) while applying 2400-Vrms, 60-Hz high-

voltage stress across every ISO124 isolation barrier. No partial discharge may be initiated to pass this test. This

criterion confirms transient overvoltage (1.6 × 1500 Vrms) protection without damage to the ISO124. Lifetest

results verify the absence of failure under continuous rated voltage and maximum temperature.

This new test method represents the “state-of-the art” for nondestructive high-voltage reliability testing. It is

based on the effects of nonuniform fields that exist in heterogeneous dielectric material during barrier

degradation. In the case of void nonuniformities, electric field stress begins to ionize the void region before

bridging the entire high-voltage barrier. The transient conduction of charge during and after the ionization can be

detected externally as a burst of 0.01–0.1-µs current pulses that repeat on each ac voltage cycle. The minimum

ac barrier voltage that initiates partial discharge is defined as the “inception voltage.” Decreasing the barrier

voltage to a lower level is required before partial discharge ceases and is defined as the “extinction voltage.” The

package insulation processes have been characterized and developed to yield an inception voltage in excess of

2400 Vrms so that transient overvoltages below this level will not damage the ISO124. The extinction voltage is

above 1500 Vrms so that even overvoltage induced partial discharge will cease once the barrier voltage is

reduced to the 1500-Vrms (rated) level. Older high-voltage test methods relied on applying a large enough

overvoltage (above rating) to break down marginal parts, but not so high as to damage good ones. Our new

partial discharge testing gives us more confidence in barrier reliability than breakdown/no breakdown criteria.