SparkFun Forums

[kuro6]

Hello,
How does the power select work on the Arduino Pro Mini 3.3V/8MHz? In other words, if I connect both a RAW input (a 7.4V battery for my system) and FTDI VCC input (3.3V programming header input), what will the mini run off? I ask because my system only works when both these power sources are connected at the same time, i.e. when the battery is plugged into RAW and the programming headers are plugged in (in order to program on-the-fly and communicate through UART).

Additional background/clarification: My system consists of the arduino pro mini and a drv8833 carrier board, a 7.4V LiPo, and two motors. When powered from from the LiPo, my motors spasm/lock/stall. Motors work normally when Arduino pro mini board receives power from 3.3V VCC programming header pin and 7.4V RAW battery input.

[Mee_n_Mac]

According to the schematic, the RAW voltage goes into a 3.3v regulator and it's output goes to the red 'power' LED and then through solder jumper SJ1 to become Vcc. Vcc goes to the 328 MCU and to the FTDI header.
https://cdn.sparkfun.com/datasheets/Dev ... ni-v14.pdf
FWIW you shouldn't connect a RAW voltage and an external Vcc at the same time. The regulator won't like that.

Perhaps SJ1 is 'blown' open ? Does the power LED come on w/just external 3.3v applied through the FTDI header ? Does the power LED come on w/just the RAW voltage applied ?

In any case just the external 3.3v should be sufficient to run the board.

[Valen]

How is does that driverboard and the motors sit between the Lipo and the Pro Mini? Please show all the connections.

[EDIT]I'm guessing the two ground pins are not fully connected on the driverboard. See the image on this page:

http://www.pololu.com/product/2130

[kuro6]

Thank you for your replies. While trying to debug I noticed that the A1/A2/B1/B2 DRV8833 board outputs jump all over the place when measuring with my mm if only the battery is plugged in. If both the battery and the programming header are plugged in at the same time, the voltage settles to zero until I send the command to move, at which point i read a steady 7.4V-ish. I wonder if this is an issue with the Arduino Pro Mini, the DRV8833 carrier board, or both.

I should also mention that I accidentally blew the FET (shorted power and ground while probing) so I had jump Vin to VMM on the drv8833 board. In other words, FET Q1 is no longer part of the circuit. Circuit behavior is the same before I blew the FET and after. I've made sure all other components work as they should after blowing the FET. Nevertheless, I have two new drv8833 boards on the way.

@Mee_n_Mac: Doesn't seem as if SJ1 is blown open. The red power LED comes one when i plug in the 7.4V LiPo into RAW only, when I plug in the programming header only (3.3 V Vcc), and when I plug in both the battery and the header.

@Valen: Please see below. It seems to me that both ground pins are connected on the carrier board as I have it connected, but I may be missing something.

Wiring diagram.PNG

[Valen]

Yes, the ground pins seem to be connected in that image. But I would measure the resistance between these point to be sure.

The outputs xOUT1 and xOUT2 should go all over the place if the xIN1 and xIN2 is both 0. The outputs are then in high impedance mode, floating, and not delivering any current to the motor. The difference in voltage between them is what is going to drive the motors once the current is allowed to flow.

Did you blow that mosfet before or after you started this thread? Maybe more got damaged by that short.

Does your code do anything with the Ain1,Ain2, Bin1,Bin2, nSleep and nFault pins? How are those pins used by the Pro Mini. Perhaps you can connect leds to it to give an indication what happens there.

[Valen]

If you want to test if it is the Pro Mini or the driver board then you should leave the Mini out of the circuit, and manually change the state of the driverboard inputs. Connect the ground of the programming cable to the driverboard ground, and connect the inputs to either ground or 3.3 volt from the programming cable. If it operates according to the truth-table as in the datasheet then it is probably ok. Make sure the nSleep pin is high, or else the driver does nothing. The datasheet suggests a 20k pull-up resistor as minimum.

I'm not sure how you can test the Pro Mini. Aside from just probing all the traces and endpoints of all the parts.

[kuro6]

I figured out what was wrong. I neglected to include the joystick controller I built on a breadboard in the schematic I posted. It turns out that the joystick was powered by a USB cable and the rest of my circuit was powered by my battery. I had two separate grounds and that was causing my lines to chatter. Thanks for pointing me in the right direction.

As for my original question: How does the power select work on the Arduino Pro Mini 3.3V/8MHz? Since there's no power select, I agree there will be contention if I plug in both a RAW and VCC input so it's best to avoid that situation.

[robertbaruch]

The Pro does not have any power selection at all. Sure, it has a switch for battery vs external, but that only determines which power connector gets to feed the Pro's 3.3v regulator. The problem is that the FTDI 3.3v board ALSO generates 3.3v, and will conflict with the Pro's regulator output. So you can either power the board externally OR plug in the FTDI board, but not both.

All is not lost, though. Assuming your FTDI board is Sparkfun's FTDI Basic 3.3v board, there's a trace on the underside of the board that lets you switch the board's Vcc output between 5v and 3.3v. It comes with the trace connected to 3.3v. If you cut that trace, the board will not output anything at all on Vcc, which means it's safe to connect to the Pro.

I find this highly annoying, because it's a permanent change. That's why I got two FTDI boards -- one to connect to the Pro when I power the Pro externally, and one to power the Pro standalone -- because I can't be bothered to solder/desolder that trace whenever I want to debug standalone.