0
\$\begingroup\$

I have a device that needs to detect the presence of voltages from 12V DC up to 240V AC RMS (Absolute maximum of 9V DC to 250V AC RMS) This device is going to have up to 8 channels that should be able to work independently with complete isolation between them.

I started working on a circuit that could accomplish this but can't seem to find a fit that wouldn't burn up with high heat dissipation or fail to trigger with low current.

The 2 options that I was thinking of were:

  1. Use a small value capacitor or high-value resistor to limit current to an optocoupler. A capacitor wouldn't work with DC voltages, (if I am not mistaken) and a high-value resistor would not be able to drive enough current to pull down a 100k resistor on the transistor side of the optocoupler at low DC voltages. If I lowered the resistance there would be enough current at low DC voltages but my calculations show it would be in the order of 2 watts of dissipation at the resistor.

  2. Use a high-value resistor with a Zener to limit voltage to around 9V on the sensing side. This should have the added benefit of triggering at a set voltage rather than varying based on the current transfer of the optocoupler. My issue here is that I then need to get the signal back over the optocoupler, there is not enough current left in the single to drive the Opto without some buffer that would require an additional isolated power supply. This could work but having an isolated supply per channel is quite of bit of wasted space and value. The stepping nature of the supplies would add a lot of noise too.

Possible solutions:

  1. Is there a better circuit or IC that could accomplish this task?

  2. Is there a low cost, low current (< 100mA), small footprint isolated DC-DC power supply available? I found the RFM-0505S, it an isolated 5V -> 5V supply at 200mA, much more than required and relatively large but could work.

  3. Would there be enough current available after running 12v DC through a 300k+ resistor to drive the base of a transistor or would I need an op-amp to buffer the voltage?

Thanks in advance for your help!

\$\endgroup\$
3
  • 1
    \$\begingroup\$ Are you simply detecting the existence? Or do you need to measure? If just detecting, I think you need to present a known load because you shouldn't want to detect static charges only. If measuring, how accurately? Fort both, how repeatedly between the inputs. \$\endgroup\$
    – jonk
    Commented Dec 29, 2020 at 8:26
  • \$\begingroup\$ @jonk that's what i want to ask too.. DC and AC are very different to measure but he said detecting 9VDC to 250VAC ??? which makes me confused af. and he added 8 channel isolated?? \$\endgroup\$ Commented Dec 29, 2020 at 8:44
  • \$\begingroup\$ @XCSource The idea is a universal digital input module that can accept 12V and 24V DC signals or 120V and 240V AC signals. The actual voltage level at the pin is not critical, just the presents of voltage to interpret as a digital signal. I have seen this product available as a CANBus module but I would like to implement just the isolated input part. For reference: Maretron "RIM100 run indicator module" maretron.com/products/rim100.php \$\endgroup\$ Commented Dec 29, 2020 at 9:15

1 Answer 1

0
\$\begingroup\$

Why don't you build a constant current sink that is set up at say 1 mA?

This circuit would be easy to design with discrete components. Worst case power dissipation is managed easily with SMD SOT223 packages.

When the proposed current sink is connected to the DC output terminals of a bridge rectifier AC or DC voltages can be detected.

The simple current sink can drive an LED and the input of a standard optocoupler like a 4N26 for an isolated digital output. 9VDC could easily be detected.

\$\endgroup\$
6
  • \$\begingroup\$ I like the idea, I assume I would need an op-amp capable of very high voltages as any methods of stepping the voltage down would limit the available current at lower voltages. \$\endgroup\$ Commented Dec 29, 2020 at 9:25
  • 1
    \$\begingroup\$ You do not need a rare onubtainium 500 Volt opamp .A 500V mosfet is easy to find and cheap because you do not need low on resistance. \$\endgroup\$
    – Autistic
    Commented Dec 29, 2020 at 9:30
  • \$\begingroup\$ Alright thanks, how about setting the current limit, would a zener suffice to create a fixed reference voltage level? \$\endgroup\$ Commented Dec 29, 2020 at 9:40
  • \$\begingroup\$ Yeah a zener could work .This 1mA does not need to be accurate .I have used a 680ohm source resister and connected this to a npn BJT .collecter to mosfet gate .VBE sense .Easy reliable . \$\endgroup\$
    – Autistic
    Commented Dec 29, 2020 at 9:46
  • \$\begingroup\$ Do you have a schematic of that, maybe post it as an answer and I can mark it as solved? \$\endgroup\$ Commented Dec 29, 2020 at 9:49

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.