Powering external devices with Arduino

The relays are an old and tried solution. There are many of those already. The second option, I am not familiar with. Is this the specs?

The NTE4066B (14−Lead DIP) and NTE4066BT (SOIC−14) are integrated circuits consisting of four independent switches capable of controlling either digital or analog signals. These quad bilateral switches are useful in signal gating, chopper, modulator, demodulator, and CMOS logic implementation. The NTE4066B/BT is designed to be pin−for−pin compatible with the NTE4016B/BT, but has much lower ON resistance. Input voltage swings as large as the full supply voltage can be controlled via each independent control input.

• VDD = −0.5 to +18.0V
• Vin = −0.5 to VDD to +0.5V
• Vout = −0.5 to VDD to +0.5V
• Iin = ±10mA
• Iout = ±10mA
• ISW = ±25mA
• PD = 500mW
  Temperature derating (from +65° to +125°C) = −7.0mW/°C
• TA = −55° to +125°C
• Tstg = −65° to +150°C
• TL = +260°C

Stresses exceeding maximum Ratings may damage the device. Maximum Ratings are stress ratings only. Functional operation above the Recommended Operating Conditions is not implied. Extended exposure to stresses above the Recommended Operating Conditions may affect device reliability.

  Can you explain how these would work?

Thanks,

Clem

Hi Clem,

thank you for the reply. Yes the NTE4066B is a digital switch. I have discovered it when I had to manage, possibly with a shift register a certain number of physical simulations of several buttons. I don't trust very much to use these kind of device to drive directly a power line, at least I should provide a opto-isolator, but are perfect to replace manual button switches because the only thing they need is to short-circuit the switch pin.

So based on your previous considerations I think that the alternative to a relais will be the 4066 + opt-isolation. This makes the circuit more complex.

The other way is a relay, I have already used the cheap finder 30.22.9.005.0010 (means size, voltage and other parameters in this dotted notation number). It is very small, fit in a 16/2 DIP socket and it is also sufficiently fast as it should only power the circuit of the two servo and the camera. Do you disagree this choice ?

I agree. Wishing digital switches could replace the clicking clack of a relay.

Also this finder relais is far away from the familiar click clack I remember on the 24V relais. A very small "tick"

So I plan to use it for power lines, contact will be opened one time and remains opened until the end of the process.

Instead, about the buttons, it will be very difficult to simulate them with a common relais because - for example - the power button has different functions with long-press, i.e. changing between shooting and video grabbing and also the shooting button has different behaviour if the camera is in video mode or photo mode.

Clem Martins wrote:

 

I agree. Wishing digital switches could replace the clicking clack of a relay.

There are solid-state relays: http://en.wikipedia.org/wiki/Solid-state_relay

An SSR is basically a power MOSFET or other power component controlled by a built-in opto-isolator.  Search for "solid-state relay" at your favorite distributor.

For the proposed application, it sounds like it's fairly low voltage and can get by using a common ground.  In this case, I'd just use NPN transistors controlled by the Arduino.  If you need more current, use Darlington pairs, e.g., something like a Texas Instruments ULN2003LV.

Thank you johnbeetem, the point is also to see what is cheaper and what more expensive. Frankly, I forgot the existence of the SSRs

None at all. Exclude the transistor.

First because the servos drain almost considerable current, second because I have already experienced the transistors and they are not reliable.

But what I ask is: why I should complicate the things - that are something like replacing a power switch with a relais - with a SSR + USN2003 ??? For about 3$ there are nice sub-micro relais that drive up to 2A from 5 up to 12 VCC... With one single component.

I wasn't suggesting both SSR and ULN2003 -- either use one or the other, depending on what you're doing.  My understanding is that your ultimate goal is to turn motors on and off, in which case the ULN2003 is a nice, cheap way to do it if you can use a common ground.  If you're just turning a separate board's power on or off, an SSR is nice.  If it's a lot of current, a mechanical switch normally has lower resistance but don't exceed the contact rating.

Transistors are usually reliable.  However, you do have to make sure you don't exceed their voltage or current limits.  For example, if you're driving an inductive load like a motor, relay, or speaker coil you must use a flyback diode: http://en.wikipedia.org/wiki/Flyback_diode

Otherwise when you switch the current off, the inductor will generate a high voltage spike that kills your transistor.  The ULN2003 has built-in flyback diodes.

Hey John,

thank you for your clarification. In fact I thought you want to use both

What you write is correct, I have already used ULN2003 to drive motors and as they have a limitation of about 1A, when I needed more power I have used ULN2004. In this case I have to manage with a separate power line low consuming devices (a micro camera consuming 180 mA) that need a separate power line for question of stability and a couple of micro servos. This is the reason that - in the need to automate all the manual commands (power, setting etc.) I though to use a micro-relay powering this part using the same 9Vcc battery-group (or wall-mount power) used to power Arduino but with a separate 5Vcc regulator. To drive the micro relais instead, I am using a N2222 NPN transistor. Maybe redundant or not, but there is the correct protection and no overheating in any place.

I have tried before making the circuit and the things seems working, leaving some hours under test. A protection diode (4001) has been put anyway between the power connectors of the microswitch.

Best, Enrico.