0
\$\begingroup\$

I am working on a project where I need to have some sort isolation from a machine sending out highs and lows of 24VDC to 0VDC respectively. I'm using a microcontroller like an ESP32 that works on 3.3VDC to read these signals.

My current prototype setup is pretty straightforward. I'm using a PC817 optocoupler with a 4.7K resistance. It works pretty well.

My question is more related to having an improved design which can improve the reliability and durability of the whole set-up. From my research I've been able to summarise that optocoulers tend to wear out in long run and are prone to failure if they don't get a healthy environment to work with.

My production model will be using multiple optocouplers, over 20 and switching on and off about 3-5 times every second and there's literally zero margin of error.

I figured out that digital isolators like SI8422, or ADUM or ISO series from different manufacturers etc. are much more reliable than an optocouplers but sadly I'm unable to find any proper source to buy them, and 4-5 days of research gave me only a few options but then they are way more costly than expected.

Just wanted to understand if there are more reliable and "easy to get" options available. Any other device which can give me an isolation like I would get from the PC817. One way was to use Darlington array chips like ULN2003 or ULN2803 with proper resistance for each input but I don't think it'll be a real isolation.

\$\endgroup\$
2
  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$
    – Voltage Spike
    Commented Nov 3, 2020 at 23:14
  • \$\begingroup\$ Please edit the question to specify what kind of chips you can source. H11L1? RS-485 or CAN transceivers? Do the machine and the ESP have the same power supply or the same ground? \$\endgroup\$
    – CL.
    Commented Nov 4, 2020 at 10:30

2 Answers 2

1
\$\begingroup\$

4K7 will allow more than 0.5mA one or two volts before diode cut-off at about 2 volts. If the signals from the machine are slow (one to three per second) the coupler is on most of the time. 3V3 comparator at the input of the coupler might reduce its on-time and make it more reliable. Resistor, comparator and opto-coupler arrays may also simplify the circuitry.

\$\endgroup\$
1
\$\begingroup\$

A suggestion is to use an optocoupler, but protect it, too. You can use inverse diodes, TVS diodes and capacitors if your circuit allow some capacitance at that point. I am saying to put these components on the side away from the sensitive circuit.

\$\endgroup\$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.