Core32 upload problem --> SOLVED

[ilmenator]

Hi,

I have a strange problem that I just can’t figure out myself - my STM32_V1 board does not accept any uploads anymore. Here is the story:

After a year or so in storage, I recently got out my STM32_V1 board. This board used to work fine, and I was able to upload apps to it before. What was a little strange when I re-activated it was that the MIDI ports did not show any sign of life. So, when switching the board on, no message would be sent from neither of the two MIDI outputs. I am not sure if this is normal, but with the old Core8 you would get some message, so I assume that is a problem.

Anyways, I connected the STM32-V1 board via USB, and everything seemed to work fine. I was able to upload a few applications using the new MIOS Studio 2.0, until I decided to try the upload of the SEQ_v4 application. That one went fine as well, but since then, the Core behaves in a really strange way. Whenever I power up the board for the first time, re-scan the MIDI-ports, set MIDI input and output ports to MIDIbox SEQ v4, the status of the board seems fine, see screenshot “04.jpg”. If I press the START button in MIOS Studio, I end up with what you can see in “05.jpg”. After that, no more communication with the core is possible, and I have to unpower the core, re-scan MIDI ports, power up the core again, and re-scan the MIDI ports again to get a status message from the core. And so on…

This is really frustrating. I thought I would master the whole system fairly well, but I have no clue what is going on here :wacko:

ilmenator

[philetaylor]

What operating system are you using?

There is a known issue with Windows 7 x64 where the 4 virtual USB midi ports that MB Seq v4 uses don’t work!

I found that changing #define MIOS32_USB_MIDI_NUM_PORTS 1 in mios32_config.h fixes the problem (but you only get 1 USB MIDI port!)

If you do this then you will need to close the bootloader jumper on the CORE32 to be able to upload over USB.

When we tested this, the problem only appeared to effect Windows 7 x64 but it may effect other versions of windows?

This is mentioned in “known issues” section of http://www.ucapps.de/mios_studio.html

Phil

[ilmenator]

What operating system are you using?

Windows 7 32bit, so I assumed that the “known issue” entry is not relevant for me.

If you do this then you will need to close the bootloader jumper on the CORE32 to be able to upload over USB.

Well, this is a V1 board, so I only have J23 available, there is no J27. Does it suffice to close J23?

I’ll try on a Win XP machine later and report back.

Thanks, ilmenator

[philetaylor]

Well, this is a V1 board, so I only have J23 available, there is no J27. Does it suffice to close J23?

I’m pretty sure that J23 won’t work. You could always short the RW pin of either of the J15 ports to ground (that is all J27 does!)

Phil

[ilmenator]

This is what I get when connecting my core32 board to a Windows XP machine using MIOS Studio 2.0. Doesn’t look more successful than on Win 7, so it may not be an OS-related issue.

[T.K]

Do you remember if the latest bootloader release (v1.1) has been installed? (it’s available under http://www.ucapps.de/mios32_download.html)

With older versions the USB MIDI communication via Juce/MIOS Studio 2 will fail (at least under MacOS, I’m not sure if the same effect happens under Windows)

Workaround: upload the new bootloader via UART based MIDI.

As Phil mentioned, the J15:RW pin shout be connected to ground (thereafter power-cycle the core) to force bootloader mode.

On the old core module, there was a resistor between J15A/B where this signal is available (see picture)

Best Regards, Thorsten.

[ilmenator]

It was probably the older bootloader, it’s been a while since I last played with it.

Whatever happened, after programming the bootloader again with the STM Flashloader demo tool, my core32 now behaves as it should :frantics: .

Maybe I should have done that straight away, but I was curious as to why that thing suddenly misbehaved.

Thanks, ilmenator