Posted by diabolo38 on 2015-12-19 20:23

I have use variuos usb devcie (cdc/hid/msc) with various stm32f4 but always via CubeMX and always worked almost of the box
So it”s worth ginving iy a try.

iti s quite easy to miss one init and break everything by copy pastign only part of exmaple it is wize to use the exampel as start point
What about clok ? Are your sure about your 48MHz? if that clcok is wrong usb will not work.
Do you have usb ip clk enable ?
do you have the all the PCD and USB LL file right ?
Not that “fs id” is needed only if you want to suport “otg”, the minimal requirement for devcie only (self pwoered) is DP/DM all other are optional.

if the base usb is ok at least you should get a first Irq (on cable plug ) after that much more need to happen to compelte the enumeration .

Yes usb code from L4689 eval shalll work on dsicovery (same mcu)
note that debuging usb emueration and transsfer comonly lead to enumeration or xfer failure due to short time out on the host side but its still help udedrtanding how things going on device.

Posted by mattes on 2015-12-23 03:11

Thanks for the feedback ‘Diabolo38’. I followed your advice and build a CDC usb project using STMCube.
Was actually straight forward, less complicated than I thought. Some minor Macro error in the stmCube build code but easy to solve and everything builds. Thats the good part.

Runtime is actually still no success. I do not get any USB interrupts when plugging it into a host.
One thing I noticed os that the USB led in active without having it hooked up to the host.

I have the feeling it might be something hardware related. Anybody ever done sucessfully USB on a L469 discovery board?

Unfortunately the pins for USB are not available on the outside connector. Maybe a trace could shed some more light on the issue.

Mat

Posted by mattes on 2015-12-23 03:37

Just did the same exercise with a F072 discovery board. and ‘Voila’ that’s what I was looking for. It enumerates the first time I plug it in.
Not sure how that works with Win7, Does it need a special driver?
No problem though on Linux. I have now a new /dev/TTYACM0 device.

That broings uo the question: what is wrong with the L469 discovery board?

Mat

Posted by mattes on 2015-12-24 20:40

got my hand on a L0 discovert board. Created the USB cdc code with STM Cube. buolds and works like a charm. Enumerates once you plug it in.

Still raises the question what’s wrong with the L4 discovery?

I checked the DP and DM line seems to be hooked up fine from the connector to the CPU pins. Ran a LA trace. Onc ethe connecytor is plugged in. I dont see any edges on the DP and DM lines. the lines stay low all the times.

Mat

Posted by bhaskarm on 2016-01-25 16:58

Were you able to solve the USB issue with the STM32L4 discovery board? I am facing the same problem now. I added a 8mhz crystal just to make sure there were no clock issues involved in the USB connection. I cannot get the USB device to work, DP and DM stay low. I also noticed that the Vbus stays low too unless the onboard regulator is turned on.

Thanks,
Bhaskar.

Posted by bhaskarm on 2016-01-26 03:57

I made sure the USB clock was 48Mhz in the stm32cube settings. I had my suspicions when I did not find a USB example for this board, I have to agree with your conclusion that something is wrong for the time being.
Thanks for replying Mat. I will post an update in case I get it working.

Bhaskar.

Posted by rsaez on 2016-02-22 12:34

Hi,

Last week I had the same problem with 32L476GDISCOVERY. I resolved it following the instructions from UM1879 document:

In order to connect the OTG_FS_VBUS and OTG_FS_ID signals from the connector CN7 to the OTG FS hardware IP of STM32L476VGT6, please remove the LCD from its socket U5, and close SB24 and SB25.

Then, I built a project with STMCubeMX configuring USB_OTG_FS as device_only with activate_vbus and resolving clock issues automatically.

I hope this can help you.

Best regards,
Raúl.