2
\$\begingroup\$

What can I use as a silent "general-purpose" relay alternative?

There are various scenarios where I will need a relay-like device that allows me to control one circuit with a separate isolated circuit.
I have 3.3-5V microcontrollers and I want to use them to control, say...

A 12V DC 1A circuit (large-ish current)
A button on a radio (low-voltage and current)
A 24V AC circuit

A simple 5V magnetic relay+diode+transistor would work for nearly any situation (and no calculations necessary), but they are too loud.
I see two alternatives: Optocouplers and solid state relays. Unfortunately, most of them seem to have a maximum output of around 50mA (I generally need 500mA or more). They are also harder to find and more expensive. Finally, they are not a simple coil and contact setup, and there is much less documentation on how to implement them with a microcontroller.

Are optocouplers and solid state relays the best solution? I'm looking at these: http://www.mouser.com/ds/2/149/FOD3150-95595.pdf
http://www.mouser.com/ds/2/427/vo14642a-279692.pdf
http://www.toshiba-components.com/docs/opto/TLP152_en_datasheet.pdf

\$\endgroup\$
4
  • 1
    \$\begingroup\$ Do you require isolation between the thing you are switching and the controller? \$\endgroup\$
    – Gorloth
    Nov 17, 2014 at 3:29
  • 2
    \$\begingroup\$ For the DC scenarios, is there any reason a simple MOSFET switch won't work? \$\endgroup\$
    – The Photon
    Nov 17, 2014 at 3:50
  • \$\begingroup\$ I desire isolation because less can go wrong. Otherwise things can get complicated. I have to custom-tailor the components for possibly unknown values. As far as I know, activating a button on a radio/other device would require matching the correct voltage/current (thus more components), not to mention unknown data signals. \$\endgroup\$
    – Bort
    Nov 17, 2014 at 4:28
  • \$\begingroup\$ For isolation, optocouplers are good. For switching large currents and voltages, however, I'd suggest a triac (thyristor). \$\endgroup\$
    – Alan Campbell
    Nov 17, 2014 at 6:28

4 Answers 4

Reset to default
1
\$\begingroup\$

A reed relay should solve all of your problems.

\$\endgroup\$
3
  • \$\begingroup\$ I wouldn't call reed relays silent, though. \$\endgroup\$
    – Nick Alexeev
    Nov 17, 2014 at 4:25
  • 1
    \$\begingroup\$ The last time I heard one was when I made a concerted effort. Otherwise, they're pretty much just part of the background. \$\endgroup\$
    – EM Fields
    Nov 17, 2014 at 4:37
  • 1
    \$\begingroup\$ I bought a few reed relays (SIP-1A05) and they work great. They make no sound when closing, and the sound when opening is no louder than dropping a mechanical pencil lead onto carpet (I tested this). The added bonus is that I can operate them directly from a 5V microcontroller without needing a transistor or the like. I'm surprised I don't see these talked about more often! The switching current of the SIP-1A05 is 500ma (1A peak). I'll try to find reed relays with larger tolerances. \$\endgroup\$
    – Bort
    Nov 29, 2014 at 22:06
2
\$\begingroup\$

The simplest is to use a solid-state relay. You can find them from a couple of hundred mA up to 100 amps, in both DC and AC. I would look for one with a logic-level input; their input is an LED hooked to an opto-isolator, so you drive it just the way you would an LED; with a resistor hooked up to output from the microcontroller.

They aren't cheap, however.

You can use an opto-coupler to drive a transistor to switch most DC current.

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

A simple MOSFET can deal with most of your DC needs. They can switch (if you get the right ones) many amps (even hundreds of amps if you look around). For the higher currents you may need to have a multi-stage driver which uses a small BJT to switch the larger MOSFET, and of course there may be heatsinking needed.

Unless you really need galvanic isolation I wouldn't bother with relays (solid state or otherwise) and optocouplers and such.

\$\endgroup\$
0
\$\begingroup\$

You could perhaps use an operational amplifier configured as a voltage follower and use its output to drive a transistor capable of switching 500 mA.

Realistically, you can expect an op amp to provide about 2 Mohms of impedance across its inputs (a more accurate value should be stated in the data sheet for your particular op amp), and it shouldn't produce any acoustic noise.

If you're only concerned about digital outputs, you can use the op amp as a comparator, where your input signal is tied to the non-inverting input and some other threshold voltage (eg 1/2 Vcc) is tied to the inverting input (with a transistor again at the output). In theory this gives you complete isolation from anything attached to the output. What I mean by that is that regardless of what's attached to the output of the op amp, your input signal will just behave as if it was connected to ground through a fixed, and very large, resistance.

\$\endgroup\$
2
  • 2
    \$\begingroup\$ It would be helpful to explain the difference between signal isolation (which your answer provides and OP probably wants), and safety isolation (which your answer doesn't provide, but OP seems to think he wants). \$\endgroup\$
    – The Photon
    Nov 17, 2014 at 16:35
  • \$\begingroup\$ @ThePhoton I emphasized what I meant by isolation at the end. Perhaps OP should clarify though, because I can't tell that he's asking for safety isolation. \$\endgroup\$
    – Ponkadoodle
    Nov 20, 2014 at 7:30

Your Answer

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

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