Find Linear Optoisolator for Measuring Voltage 0-3 Vac

  • Thread starter Thread starter Pagedown
  • Start date Start date
  • Tags Tags
    Linear
AI Thread Summary
The discussion focuses on selecting a linear optoisolator for measuring 0-3 Vac voltage, emphasizing the need for isolation when interfacing with an MCU. Participants suggest digitizing the AC signal before transmission and highlight the challenges of maintaining linearity and response time, particularly for a 60 Hz sine wave. Isolation transformers and galvanic isolators are recommended as alternatives for faster response times, although concerns about cost and availability are noted. There is also a discussion about using diode bridges and RMS to DC converters for accurate signal processing. Overall, achieving effective isolation and accurate measurement while considering the limitations of available components is central to the conversation.
Pagedown
Messages
92
Reaction score
0
Which linear optoisolator should I find, when I am want to measure voltage.

I want to isolate this voltage and transfer it to a MCU(linearly). This measure voltage has a range from 0 -3 Vac.
 
Engineering news on Phys.org
If it's possible, I'd recommend digitizing the information first and then transmitting the binary signal over the optoisolated / galvanically isolated link.

You'll have troubles enough maintaining linearity and proper operating conditions, especially and may have even more trouble keeping up with the signal, depending on how fast it changes (in my experience, most optos are fairly slow and have response times measured in 10s of microseconds, if not milliseconds)
 
you mean using an ADC then only to the isolation optocoupler?

I need to measure a sine wave of 60hz.
 
That's correct. That or use an isolation transformer to bring the signal across the gap.

Given that it's only 60 Hz, you could probably sample and send back every millisecond with appropriately chosen higher-speed optoisolator and ADC. Galvanic isolators (ones that use inductive coupling) are generally much faster (although a little pricier), so you may want to consider them as well.
 
I am living in M'sia thus find it hard and expensive get this special parts.
 
Okay,tq.

I don't really know how a isolation transformer. Can you kick-start me?
 
I think a transformer is too big.

and the optocouplers are abit pricey tto buy online from my place.

Using just a op-amp buffer is not appreaciated?
 
Last edited:
  • #10
Thanks bob for the informative link.

I am using an MCU to measure voltages of 240Vac dropped to around 2.4Vac via a voltage divider resistor circuit.

Thus for safety and line noise I need an isolation to send signals to the ADC of MCU. I know using an external ADC is better, but I have yet to learn to use an external ADC.
 
  • #11
I agree that you need 100% isolation between the 240V ac and the MCU. The first thing you need to consider is how to process the 60-Hz ac signal. You obviously don't want to digitize the ac signal directly, unless you do it at say 3000 Hz to get the full waveform. So you either want to use a diode bridge with a RC integrating circuit (RC ≈ 2 seconds) to get a good dc average, or use a special RMS to DC converter chip to get a RMS value. This Linear Technology app note discusses an isolated RMS measuring circuit. http://cds.linear.com/docs/Application%20Note/an106f.pdf
I have seen many circuits that produce pulses with the output frequency proportional to the input dc voltage (voltage to frequency converters). The optocoupler can receive the pulses (example, 240 Hz for 240 V ac), and your MCU would have to count the pulses in a defined time interval, and convert back to voltage.

Bob S
 
Last edited by a moderator:
  • #12
I need to do further analysis on the voltage itself like voltage sag, total harmonic distortions..etc. I think I might need to plot out the sine wave and further maths on it.

If I convert it to pulses with different frequencies, can I still do it accurately?
 
  • #14
It looks versalite.

But doesn't look good for AC, and is the information about the frequency of the voltages intact or only amplitude can be calculated?
 
Back
Top