Page 6 of 60

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 00:02
by arkadiusz.makarenko
terriblefire wrote: 18 Apr 2020 23:53 Worth noting...
Clock, reset and supply management
– 1.7 V to 3.6 V application supply and I/Os
– POR, PDR, PVD and BOR
– Dedicated USB power
– 4-to-26 MHz crystal oscillator
Internal 16 MHz factory-trimmed RC (1%
accuracy)

– 32 kHz oscillator for RTC with calibration
– Internal 32 kHz RC with calibration
Can you use the internal 16Mhz?
I don't think it is precise enough for USB.

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 00:10
by terriblefire
arkadiusz.makarenko wrote: 19 Apr 2020 00:02 I don't think it is precise enough for USB.
Should be enough to prove the clock is the issue.

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 00:20
by arkadiusz.makarenko
terriblefire wrote: 19 Apr 2020 00:10
arkadiusz.makarenko wrote: 19 Apr 2020 00:02 I don't think it is precise enough for USB.
Should be enough to prove the clock is the issue.

Clock is not the issue.
I am almost certain that it is config of the chip.

I have just checked HID2AMI (which did work 100% ok with f105). I swapped chip to f722 and it exactly the same story.

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 00:31
by terriblefire
arkadiusz.makarenko wrote: 19 Apr 2020 00:20 Clock is not the issue.
I am almost certain that it is config of the chip.

I have just checked HID2AMI (which did work 100% ok with f105). I swapped chip to f722 and it exactly the same story.
There is the STM32 Utility

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 10:05
by arkadiusz.makarenko
I have ordered Nucleo f722 dev board to understand this chip better.

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 10:12
by terriblefire
arkadiusz.makarenko wrote: 19 Apr 2020 10:05 I have ordered Nucleo f722 dev board to understand this chip better.
Good call. Is that dev board schema available?

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 11:48
by arkadiusz.makarenko
I have implemented simple blinky and what I did find.
When I clock it from internal clock it is nice steady blinking. But when I set up external clock crystal it does work, but it is disco - visible difference between led toggle.

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 11:58
by terriblefire
arkadiusz.makarenko wrote: 19 Apr 2020 11:48 I have implemented simple blinky and what I did find.
When I clock it from internal clock it is nice steady blinking. But when I set up external clock crystal it does work, but it is disco - visible difference between led toggle.
That is really weird. And the same on HID2AMI with the F722?

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 13:41
by arkadiusz.makarenko
terriblefire wrote: 19 Apr 2020 11:58
arkadiusz.makarenko wrote: 19 Apr 2020 11:48 I have implemented simple blinky and what I did find.
When I clock it from internal clock it is nice steady blinking. But when I set up external clock crystal it does work, but it is disco - visible difference between led toggle.
That is really weird. And the same on HID2AMI with the F722?
Yes ...
but Clock Resolver is not doing right job, there must be a bug there or on code generation.
I have slowed down core to 48Mhz manually, (automatically it goes to randomly either 92Mhz or 216) and I managed to get device Ready, and so far no crashes and LED toggling nicely and steady. I need to check USB HS (I read that code generator is buggy as well enabling wrong clocks on this section)

EDIT
And I have now both USB HS and FS responding up to Ready state. :)

Re: TF CD32 Riser Revision 2 Design Complete

Posted: 19 Apr 2020 13:54
by terriblefire
So clock generator code is a pile of crap? Damn if you cant trust this then....