I wish to measure the phase voltages of an induction machine using Avago isolation amplifier (ACPL-790A).
(
http://www.avagotech.com/pages/en/optocouplers_plastic/plastic_miniature_isolation_amplifier/acpl-790a-000e/)
Figure 21 (attached) is copied from the datasheet and it is for current sensing. Note that the circuitry of R5 (10ohm) and C3 (47nF) forms an anti-aliasing filter.
The maximum phase voltage of the induction machine is expected to be not more than +/-50V. Since the recommended input voltage range of the ACPL-790A is only +/-200mV, a voltage divider is required to "step-down" the +/-50V to +/-200mV. To assure a current of not more than 1mA flowing through the "voltage divider" branch, R1=25kohm and R2=100ohm are selected (see attached figure).
According to datasheet (pp.14):
... The only restrictions are that the impedance of the divider be relatively small (less than 1kohm) so that the input resistance (22kohm) and input bias current (0.1uA) do not affect the accuracy of the measurement. An input bypass capacitor is still required, although the 10ohm (R5) series damping resistor is not (the resistance of the voltage divider provides the same function). The low-pass filter formed by the divider resistance and the input bypass capacitor may limit the achievable bandwidth.
Without the R5, and if R1=25kohm and R2=100ohm, what is the resistance of the voltage divider that will form the low-pass filter with the capacitor C3? Is it (R1*/R2)/(R1+R2)~=99.6ohm?
Do you have any idea to prevent the resistance of the "voltage divider" from affecting the bandwidth of the anti-aliasing filter formed with R5=10ohm and C3=47nF?
Cheers.
just having quickly looked at this, why have you limited current through the voltage divider to 1mA? This seems to have raised your divider resistor values above the recommended 1K?
I wish to measure the phase voltages of an induction machine using Avago isolation amplifier (ACPL-790A).
Yes, we know, because this is at least the third time you are posting this. If this is so difficult for you, why don't you just start with the schematic in the datasheet, and build it, instead of trying to "improve" it upfront? And it worries me that you have problems calculating a resistive divider but want to play with dangerous voltage. "Outsourcing" your design to the forum won't protect you from the dangerous voltage.
I wish to measure the phase voltages of an induction machine using Avago isolation amplifier (ACPL-790A).
Yes, we know, because this is at least the third time you are posting this. If this is so difficult for you, why don't you just start with the schematic in the datasheet, and build it, instead of trying to "improve" it upfront? And it worries me that you have problems calculating a resistive divider but want to play with dangerous voltage. "Outsourcing" your design to the forum won't protect you from the dangerous voltage.
Thanks for your reply. I do not have an ACPL-790A now. I wish to understand how the circuit works and to confirm that it will suit my application before purchasing several units to test. Since I've limited funding, I hope to get the right parts and do not waste too much money for getting the wrong parts and paying delivery charge several times. Thanks for reminding me about the dangerous voltage. The maximum dc-link voltage in my application will not be more than 50V at the moment, though in future development it may go further up. I must admit that my circuit analysis is weak. With the voltage divider, I'm not sure what is the effective value "seen" by the C3 capacitor to form the anti-aliasing filter. I will be glad if you could enlighten me.
Cheers.
The source impedance of a voltage divider is the parallel resistance of the two resistors -- assuming the input to the divider is low impedance. You can read about
Thevenin equivalent circuits. The equivalent circuit to your phase voltage + divider is a smaller voltage in series with a ~99.6 ohm resistor. You don't need R5, and it wouldn't materially change the behavior in any case -- the rolloff would be dominated by the 100 ohm resistance and the value of C3, which should be chosen to give the desired rolloff based on a 100 ohm resistance.
The source impedance of a voltage divider is the parallel resistance of the two resistors -- assuming the input to the divider is low impedance. You can read about Thevenin equivalent circuits. The equivalent circuit to your phase voltage + divider is a smaller voltage in series with a ~99.6 ohm resistor. You don't need R5, and it wouldn't materially change the behavior in any case -- the rolloff would be dominated by the 100 ohm resistance and the value of C3, which should be chosen to give the desired rolloff based on a 100 ohm resistance.
Thanks for your reply.
Without the voltage divider, the cut-off frequency of the anti-aliasing filter (R5=10ohm, C3=47nF) is approximately 338kHz.
With the voltage divider, since the rolloff will be dominated by the 100ohm, the cut-off frequency will be reduced to approximately 33.8kHz, which is not acceptable in my application.
Do you think it's possible to use an op-amp (configured as a voltage follower, inserted between the voltage divider and the anti-aliasing circuit) to minimize the influence of the voltage divider on the bandwidth of the "R5+C3" anti-aliasing circuit?
Cheers.
That will work, but I would just change C3 to 4.7 nF? Then you get back up to 338 kHz.
That will work, but I would just change C3 to 4.7 nF? Then you get back up to 338 kHz.
Thanks. According to the datasheet (see quotation below), a 47nF bypass capacitor (C3) is recommended at the input pins due to the switched-capacitor nature of the input circuit. I'm not sure if the bypass capacitor can go lower.
A 47 nF bypass capacitor (C3) is also recommended at the input pins due to the switched-capacitor nature of the input circuit. The input bypass capacitor also forms part of the anti-aliasing filter, which is recommended to prevent high-frequency noise from aliasing down to lower frequencies and interfering with the input signal. The input filter also performs an important reliability function – it reduces transient spikes from ESD events flowing through the current sensing resistor.