2017-08-24 08:05 AM
I'm trying to use a Nucleo F411RE with some Arduino shields, and having very little luck.
The behavior I get is that the Nucleo either doesn't power up at all, or are slow and unreliable about enumerating the USB devices it provides, and I get errors about device addition failing in the system log.
Removing JP1 - which limits current draw and forces the device enumeration - makes things a little better. Sometimes. But sometimes it just means that the errors in the system log repeat until I get tired of it.
Providing external power also helps. These two things make me think I've missing something power related.
This happens with multiple Nucleo boards and all my Arduino spare shields (just 3). They are all pre-R3 boards, but one of them is a prototype board with nothing but parallel connections for the reset pin and LED.
Other people don't seem to be having these issues, and I've not turned up anything in the manual.
So the questions are: am I wrong in expecting generic Arduino shields to work with the Nucleo boards? If not, is there some setup I need to make or check in order to get them to work?
Thanks
#shield #arduino #nucleof411re2017-08-24 08:16 AM
Not having any of these issues here, and have loads of boards with reset buttons on them.
Guess I'd be wary of the 3V, 5V and IOREF rails, and if you have regulators what they are doing with VIN, etc.
The Nucleo boards have schematics, review if these have any conflicts with the shields you use.
The Nucleo-64 boards have some use conflicts on the D0/D1 UART pins, in that they connect by default to the ST-LINK VCP, so you can't drive them from the shield and the F411, and there isn't the flexibility in the RX/TX designations.
Current firmware on the ST-LINK is strongly recommended, get a current copy of the ST-LINK Utilities, it should have current drivers, and firmware with the package.
2017-08-24 11:09 AM
I updated the ST-LINK firmware when I started the project; that's simply SOP for me. I'm using Linux, so there are no driver issues. I'm all to aware of the D0/D1 issues. Since they also get used for monitor comms on arduino's, it's really rare to find a shield that uses them for anything.
I'm not sure what I'd be looking for on the schematics. Any hints would be appreicated. Or was 3V/5V/IOREV a hint?
I checked the Adafruit motor shield before I started (schematics are
https://cdn-learn.adafruit.com/assets/assets/000/009/769/original/mshieldv1-schem.png
if you want to look), and everything seemed fine to me: it doesn't output anything back to the board except maybe power, and I made sure that jumper was off. I did check on driving 5V logic with 3.3V circuitry, and the indications are that it should be fine. For this shield, not -0 the 74HCT595N needs ~3.2 volts for a logic high, so things tend to be flaky. But I don't see how that would keep it from powering up before I wrote anything!I'm thinking I missed something (highly likely) and the power system on the board (both of them, actually) is malfunctioning such that any extra load causes the boot sequence to fail. But that's just a wild guess.
Thanks,
Mike
2017-08-24 11:58 AM
The hint was to look in the context of what you have attached, I don't have schematics for those, or a lot of time to commit to raking over them.
The power supply via USB and the on-board regulators is going to be pretty limited. The Nucleo should provide the 5V and 3.3V, the latter likely has a total budget of 150-200mA. This isn't going to be sufficient for motors or cellular modems, etc. Should be quite sufficient for a few buttons and LEDs
2017-08-24 05:24 PM
Knowing where to look doesn't help much if you don't know what you're looking for. Of course I don't expect anyone to look over the schematics, though I was hoping someone might as I had the one handy. Sorry if you gave you a different impression.
The F411RE can provide up to 300ma. Of course I didn't expect to drive motors with that. I've never seen a motor driver that didn't have separate power inputs for the logic and motors. There's usually an option to power the board off the motor driver, even. That doing that tends to improve things is why I think it might be a power issue. But the evidence just leaves me confused:
1) All three boards work fine on UNO's.
2) They all fail on different Nucleo's.
3) The Nucleo's work fine without the board, including driving a similar motor driver breakout.
I tested a newer shield today and the Nucleo started fine. The board is junk as shipped (multiple failed connections and shorts, including motor power to ground) so I'm returning it. Possibly the lesson here is 'Don't use pre-R3 shields.'
Thanks.
2017-08-24 07:48 PM
No offense meant, just trying to scope expectations, I'm unpaid.
Check the supplies aren't drooping. Check the state of NRST
I believe you're going to have to resolve the pin conflicts on D0-D7, especially D1. Consider bending up groups of pins to isolate the ones causing grief.
2017-08-24 08:10 PM
No offense taken, and I had no intention of making you think I was. I appreciate the help. I like to think that a few decades of professional software development has given me good problem solving skills, but this electronics stuff is new to me, and you're helping me get a handle on it.
Given that I'm now waiting for a new motor shield from England, I'll put together an R3 proto shield and leave just leave off the D0-D7 header. But I'm confused as to how D1 could cause problems. I thought it wasn't connected to the mcu, as the pin that should be on is instead connected back to the STM32F103 on the ST-LINK.
2017-08-24 08:21 PM
maybe this table from MBED web site could help: