Arduino Micro voltage requirements

Greetings Arduino community,

I have built a prototype LED display taking advantage of an Arduino Micro and two TLC5947 breakout boards chained together. The project works, but my question involves the power requirements of the VI pin of the Arduino Micro board.

According to the Arduino Micro description:

VI. The input voltage to the Arduino board when it’s using an external power source (as opposed to 5 volts from the USB connection or other regulated power source). You can supply voltage through this pin.
AND
Input Voltage (recommended) 7-12V
Input Voltage (limits) 6-20V

When power is supplied via the USB cable to the microUSB port on the Arduino micro, the output is as expected. The USB 3.0 port on the laptop nominally delivers 5VDC, 500 – 900 mA max). Naturally, I don’t want to be tethered to a laptop all the time: I want to plug it into 120 VAC wall socket via an AC/DC converter (wall wart). I have a DC barrel plug soldered in to the prototype for this purpose. The 7-12 volts from the barrel plug on the prototype circuit board goes to the VI pin and the ground to ground pin.
When I power it through the barrel plug with a DC converter (rated: 9 VDC 500 mA, actual: 8.8 VDC by multimeter), the LEDs follow the sketch for about 5 seconds and then it freaks out: the LEDs flash very rapidly and randomly and eventually stop altogether. When I use another DC converter (rated: 12 VDC, 1000 mA, actual: 12.3 VDC) I see the same thing (rapid/random/shut down). And then, getting desperate, I spent the big bucks on a Jameco ReliaPro UL Listed DC converter (rated at 6VDC, 800 mA, actual: 10.5 VDC by multimeter, not a typo, but still within recommended voltage range) I get ditto result (rapid/random/shut down).

OK, the solution was to solder in a 5 V voltage regulator (L7805CV) onto the project circuit board between the DC barrel jack and the VI pin of the Arduino Micro board as shown in the attached schematic. Now ALL of the aforementioned AC/DC converters work fine. Problem solved! But I shouldn’t have to add the voltage regulator. The Arduino Micro has built-in voltage regulator (NCP1117-5) to allow the use of variable 7 – 12 VDC input.

So why do I need the extra voltage regulator?

THinRTP: (rated: 9 VDC 500 mA, actual: 8.8 VDC by multimeter), (rated: 12 VDC, 1000 mA, actual: 12.3 VDC) (rated at 6VDC, 800 mA, actual: 10.5 VDC by multimeter

When you measured these, were they connected to the Arduino? Measuring a switching supply like (or any supply actually) without a load, doesn't give a proper view of the voltage. In the case of a switching regulator, there is generally a minimum load required before the supply goes into regulation. (In the case of something like batteries, the voltage will drop do to its internal resistance.)

THinRTP: the LEDs follow the sketch for about 5 seconds and then it freaks out: the LEDs flash very rapidly and randomly and eventually stop altogether.

It would probably be a good idea to figure out how much current your project is drawing. Its very likely that there is enough current that the on-board regulator is heating up and going into shut down, causing the strange behaviors.

THinRTP: OK, the solution was to solder in a 5 V voltage regulator (L7805CV) onto the project circuit board

This also supports the over current idea. What package size is the regulator you added?

Hi James,

Thanks very much for your thoughtful suggestions and questions.

Let's take them one by one. 1) With regards to the DC converter: I bypassed the voltage regulator (that I had soldered in to the circuit) to measure voltage with load using my multimeter. The Jameco ReliaPro UL Listed DC converter (rated at 6VDC, 800 mA) delivered 8.0 V. The DC converter rated at 12V delivered 8.5 V under load and the one rated at 9V also delivered 8.5 V under load.

2) Your suggestion that there was an over-current situation that caused the erratic behavior and shut down sounds right. After I bypassed the voltage regulator that I soldered in, the TOTAL current that the project drew was 0.75 A (750 mA). Collective gasp. Yes, I know, the Arduino Micro max current is 50 mA. However, let's keep in mind that this current is not going through the Arduino (see schematic). The LEDs are driven by two constant current open drain TLC5947 breakout boards each with 24 channels. Adafruit makes the breakout and they installed resistors so that each channel delivers 15 mA of current. So, let's see, 24 channels X 15 mA X 2 boards = 720 mA. The next question is how much current is flowing through the Arduino itself, and I don't have that measurement yet.

3) The 7805 voltage regulator is described here: addicore.com/Addicore-L7805CV-5V-Voltage-Regulators-5-pieces-p/114.htm

Oh, dear, In the process of measuring the current and voltage, I bypassed the voltage regulator in order to reproduce the erratic behavior that I described previously. I, um, fried my Arduino Micro. This is a first...a rite of passage ! The LEDs are full on, no change, the reset in unresponsive and the IDE does not recognize the board's serial port. Anyway, no worries, I will buy another Arduino Micro, solder it in and all should be OK as long as I have the voltage regulator in place.

--THinRTP

THinRTP: The Jameco ReliaPro UL Listed DC converter (rated at 6VDC, 800 mA) delivered 8.0 V. The DC converter rated at 12V delivered 8.5 V under load and the one rated at 9V also delivered 8.5 V under load.

Assuming your 750mA load, nothing here matches up with what is expected. Except that I would expect a 800mA rated supply to be in regulation, not wildly out of it.

THinRTP: 3) The 7805 voltage regulator is described here: addicore.com/Addicore-L7805CV-5V-Voltage-Regulators-5-pieces-p/114.htm

A TO-220 package can handle SIGNIFICANTLY more current than the tiny sot-23 package used on the Arduino board....

THinRTP: Oh, dear, In the process of measuring the current and voltage, I bypassed the voltage regulator in order to reproduce the erratic behavior that I described previously. I, um, fried my Arduino Micro.

Changing connections (including measurements) should not be done while the board is powered. If you're moving probes around while powered, your attenshion (your eyes) should 110% be focused on where the leads are going.

THinRTP: Yes, I know, the Arduino Micro max current is 50 mA.

I'm not sure where you got that from. The max current that the 3.3V rail is rated for is (erroneously) 50mA. (It's actually 150mA).

The 5V rail is rated for whatever you can draw until it shuts down, and that depends on Vin.

THinRTP: However, let's keep in mind that this current is not going through the Arduino (see schematic).

I'm confused. Has this been your setup the entire time, or just after adding your own regulator?

I’m confused. Has this been your setup the entire time, or just after adding your own regulator?

This has been my setup. The only change was adding the regulator.
To allay any confusion…
Please see the uploaded schematics.

Version 1 (no regulator) caused the erratic behavior/shutdown when I used a DC converter plugged into the DC jack, but there was no problem when power was supplied using the USB cable attached to the micro USB jack on the Arduino.
Version 2 (with regulator) functioned perfectly with all available DC converters and also worked perfectly with the USB cable (as expected).

To reiterate, the question is, “why is the voltage regulator needed?”
It seems like all of the components, including the Arduino Micro, should be happy with the power that is supplied by the DC converter directly, but they aren’t.

THinRTP: To reiterate, the question is, “why is the voltage regulator needed?”

What value decoupling caps are you using when you do (and don't) have the regulator?

My guess is that your LEDs are causing enough ripple that it giving the on-board regulator a hard time. When you use your off board regulator, it becomes the "decoupling" device.

The whole point of using the TLC5947 is that you should never connect the LEDs (anodes) themselves to your regulated supply; the chip is deliberately rated to 30V on this part and its purpose is to control the current. The 5V supply to the TLC5947 breakout should of course come from the Vcc pin - or the same connection if you are supplying this via another regulator. If you power the system from the coaxial (“barrel”) jack, you can probably take the LED supply conveniently from the “Vin” which is merely a diode drop from the coaxial jack.

Looking at the datasheet, there are some conditions under which the TLC5947 may draw up to 90 mA and you have two such modules, so you may well need a heftier regulator and preferably a switch-mode, to power the Arduino (on Vcc) and the “breakouts”.

James and Paul,

Thanks for your comments.
Re the capacitor question, I have not added any capacitors (as discrete components) to the circuit. The capacitors present are embedded in the voltage L7805 voltage regulator and the Arduino micro board.