[Q] STMbluepill wifi & usb connection control #bluepill


universe24
 

I have recently built the STM bluepill v1.8.
The initial flash of OnStep into the bluepill look good. Yellow LED light up and RA motor start autotrack.

I got a D1 mini from a friend who previously built similar system. Without any further flashing this D1 mini, the wifi up with ip which not default use by OnStep. However, I was unable to connect via OnStep mobile apps.

I flash the latest WSW to the D1 mini and the wifi not show up anymore. I've tried many times and still fail.

After that, I test the OnStep ASCOM connection. The connection fail too.

But the RA motor will start autotrack whenever the system power up.

May I know what happen?

Thanks.


universe24
 

This picture show those light that up when power on the device.
The LED on CP2102 only always ON if not connected to computer USB port. If connected, light will go OFF after a short while.
The blue light on the D1 mini blink several time then stop about a second then blinking again.
Both LED on STM32 Bluepill are always ON. No blinking.
RA motor is rotating as set to Autostart in Config.h



Dave Schwartz
 

The Wifi module blinking 3 times followed by a pause, repeating forever indicates it is not establishing it's serial connection with the main processor. This board is designed so that SWAP is not necessary in the SWS configuration so leave it off. The baud rates on both sides should be set to the same value - I use 115200.

If tracking is enabled from the start, the green LED on the STM32 should be blinking since that is the tracking LED.


universe24
 

That's weird.
The board tracking LED never blink since the first time the OnStep flash into it. The LED just light up and stay after it being flashed.

The bluepill seem to be working fine after I use the utility tools in the file section. And OnStep can be flash into it.
The D1 mini also seem working fine after several flash of different example sketch in the arduini IDE.

I starting wondering if the stm32cubeprog version has something related to the board refuse of communication issue. Will try with older version.


universe24
 

After setting all to 115200, the D1 mini blinking 3 times followed by a pause, as described by Dave Schwartz.
An it never establish the connection.
The board tracking LED no blinking. T_T


Mike Ahner
 

On Sat, Jun 25, 2022 at 08:58 PM, universe24 wrote:
I starting wondering if the stm32cubeprog version has something related to the board refuse of communication issue. Will try with older version.
I used STM32cubeprog v2.4.0 for RobotDyn F303 and the required STM Board Manager v1.9.0, this worked for me. I was upgrading to the OnStep 4.24r to correct for the guiding bug.

-Mike


Mike Ahner
 

On Sun, Jun 26, 2022 at 09:13 AM, universe24 wrote:
After setting all to 115200, the D1 mini blinking 3 times followed by a pause, as described by Dave Schwartz.
Doing this speeds up the connection process, but I have never changed mine. It still defaults 9600 and eventually connects at a faster bau. Connection time is 3~5 seconds.
-Mike


universe24
 

It seem that I got a clone bluepill after some read on the internet.


Mike Ahner
 

On Mon, Jun 27, 2022 at 09:27 AM, universe24 wrote:
It seem that I got a clone bluepill after some read on the internet.

Hi, Do you mean a clone BluePill board or a clone STM32F103C8T6 chip? It looks like a STMicro brand mpu chip and I believe all of the BluePill boards are just built from open-source schematics, so I don't think this is your problem. If you mount starts tracking, then OnStep is running. You can use the serial channel from with the Ardunio IDE to issue commands to OnStep and verify it's working.

This command will return the name of the OnStep controller and let you know the OnStep is active.
Get firmware name: GVP#    Reply: On-Step#

Here is a list of all OnStep commands:
https://onstep.groups.io/g/main/wiki/23755

More likely is a problem with a solder joint or wiring connection to the D1 Mini, or sometimes the D1 Mini doesn't flash on the first try. You can read in the Wiki (https://onstep.groups.io/g/main/wiki/26881) about flashing trouble but mostly it seems some D1 Mini boards require flashing 3-5 times for them to work.

I would check all the solder connections from the D1 Mini RX/TX pins to the Blue Pill RX/TX pins with a continuity tester to be certain you have good connections. Be sure and test the soldered connections on both the D1 Mini and the Blue Pill board. You D1 Mini is not new, but the Blue Pill mpu and the OnStep controller are new so there is always a possible bad solder connection.

-Mike


universe24
 

Hi Mike,

I mean the STM32F103C8T6 chip.
But after some further readings, I found that the chip in the picture of my previous post is actually one of a good variant of STM32F103C8T6 chip.

I have bought another two BluePill from local store instead of order them directly from China.
These two BluePill manage to flash OnStep and both wifi & serial communication is in good shape.
I didn't do any other further checking on the solder connection except re-wired the ground (-) of the 12V, 2.5A power source.

But here is the weird part, after I successfully flash the two BluePill mentioned above, the BluePill that gave me problem also being flashed successfully with tracking LED blinking while tracking.
Both the wifi & serial connection success too.
This make me wondering if all this while the board not getting enough power due to the loosen ground point of the power source.

By the way, inside the OnStep.ino, the firmware name command is ":GV?#". No wonder the board not replying "On-Step#"

The D1 Mini seem to be fine. Connection established quite fast.

So now I have three working BluePill. XD
With one of the BluePill found to be High-Density with 512KB of storage.

William


George Cushing
 

No, that IC is apparently a fake. The link contains a comment:

"The important part is that markings with:
991KA 93
MYS 807
means those chips are fake."

I keep getting ads for the "Longan Nano RISC-V GD32VF103CBT6 MCU." It is (was?) an attempt to duplicate the ST product with a non-ARM core. After 3 years the effort seems to have fizzled. Nuclei Systems doesn't appear to be developing tools for it at this time. 


universe24
 

There seem no where to get genuine chip.
Even the "established" store here also selling the fake one. See the picture attached.