- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hello,
I'm working on moving to a new STM32 platform. We currently have prototyped on the SN8205 platform from murata with success. Now we are moving towards a SN8000 with a STM32F429 MCU. I have been looking at SN8000 WiFi Module with STM Discovery Board and with some help I have managed to get the basic platform up and running, so I can run ex. the GPIO snip and the UART snip.
The next step is to have the SN8000 working, my hope was to test it using the SCAN snip, but I have yet to get it to run.
It seems to die on the "wiced_wifi_scan_networks" call, looking into it it seems to be the "result = wiced_rtos_get_semaphore( &scan_semaphore, SCAN_LONGEST_WAIT_TIME );" line which is, I assume, crashing the STM32.
The connection to the SN8000 module has been checked, double checked and seems to be OK. Has anybody which has connected the SN8000 to a STM32F4xx MCU seen the same? and even better found a solution?
Regards
-Renejos
Solved! Go to Solution.
- Labels:
-
SDK 3.x
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I just received my custom PCB, it works like a charm. The PCB does not have pull up resistors and STM32F4 MCU and the WiFi module is very close to each other.
My conclusion is that many of the problems experienced, was caused by the use of prototype setup.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have enabled the WWD error messages and I get a "Could not turn on APSTA & Could not set AMPDU parameters" in the log.
This problem is discussed in this old thread Re: wiced_ioctl_sleep semaphore timeout
Has anybody found that answer to this problem?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi
Any update on this issue ?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Nope I'm trying to get SPI working as a fix to get moving, but I'm keep getting a
"Read 0, instead of 0xFEEDBEAD from the WLAN chip
Could not initialize bus"
So I will have to look closer into it, but I also need to get SDIO working at some point.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Actaully I had make some modifications , Use LSI insted of LSE in the platform_mcu_powersave_init( void ) function,
1. Did you faced that issue ?
2. Or Is it because of that we're using wired setup for communication..?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
And these modifications is to get the SPI interface going or just in general?
I am going tI dig a little in the SPI and hopefully get it going, but the custom PCB which we are developing use the SDIO so at some point I need to get that working. Did you get the SPI to work so that the scan snip can run?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
1. The modification of LSI insted of LSE in the platform_mcu_powersave_init( void ) was general , cz if you check the schematic of stm32f429 discovery , the LSE 32khz osc is not fitted , so I had to enable LSI(internal) to get it going ..
2 SPI is also still showing some issues
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Update: Did some experiment , I have a SN800x EVB module, so I tried disconnecting the WICED_GPIO cable completely . So it is throwing the same error we are getting while trying with the STM32F429 discovery.
Starting WICED v3.1.0
Platform SN8000x initialised
Started FreeRTOS v7.5.2
Initialising LwIP v1.4.0.rc1
Could not initialize bus
Waiting for scan results...
# Type BSSID RSSI Rate Chan Security SSID
----------------------------------------------------------------------------------------------
Starting WICED v3.1.0
Platform SN8000x initialised
Started FreeRTOS v7.5.2
Initialising LwIP v1.4.0.rc1
Could not initialize bus
Waiting for scan results...
# Type BSSID RSSI Rate Chan Security SSID
----------------------------------------------------------------------------------------------
So I strongly suspect this is an issue because of improper handling/connection of WICED_GPIO_CABLE
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
With the WICED_GPIO_CABLE, do you mean the additional ribbon on the SN8000 module?
Currently I also get the "Could not initialize bus" when using SPI, and the bootstrap GPIO pins seems to be at correct state to indicate SPI to SN8000 module.
From what I can see the NSS (CS) pin of the SPI interface, does not match the GPIO pin which is set in the platform files I am using, I have been wondering if I see the same problem on the SPI as I had on the USART2 where I only managed to get one of the two possible pin configuration working.
Also the WWD_PIN_POWER I have had connected to VDD, but that does not seem to be the cause of the problems.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Finally now able to do a scan operation....
Had to do some tweaks and , there are some unexplained phenomenons also , assuming this is because
of the signal integrity issue with our wired setup on SDIO interface.
1. Put 47k pullup's in CMD,D0,D1,D2,D3
2. Change the value of sdio_init_structure.SDIO_ClockDiv from 0 to 6 in host_platform_enable_high_speed_sdio( void ) in wwd_SDIO.c, which reduces the SDIO clock frequency to 6MHz.
3. Connected end of the DSO(Oscilloscope) probes to CMD and D1 holes of the SN8000x module (MYSTERY ) .(Hope it's because of signal integrity issue).
Anyway I can scan and see the Available networks.. But after some scan doesn't show any results
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
It sounds wonderful that you have managed to get it scanning. I'm wondering if the STM32F4xx should have internal pull ups for the SDIO pins ? Looking at the schematic for the broadcom evaluation boards, there does not seem to be any pull up resistors. With that said, the SN8000 EVK, does have 47k on the lines.
When looking at the datasheet of the SN8000 module, it states that the module does have internal weak pull-ups.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Our custom hardware bring up also looks good..
That must be the issues related to our wired setup.
We got the reports that custom hardware is able to do scanning properly even @ 26Mhz of SDIO operation.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I'm battling a "Could not initialize bus" error after having wired the SN8000 module.
Was hoping to get atleast to the "Could not turn on APSTA & Could not set AMPDU parameters" again before trying to add the pull ups.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Ok, after adding the 47k pull ups and changed the value from 0 to 6, I'm now back at "Could not set AMPDU parameters".
Any ideas?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Actually when we digged deeper we found that our firmware download was getting timeout ,there's a semaphore waiting for the firmware download using SDIO to complete and that gave us the error message "Could not initialize SDIO bus".
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This is excatly what I experienced when trying to change the clock devider, before adding any pull up to the SDIO.
After having added the the 47k pull ups, I'm stuck with a "Could not set AMPDU parameters".
Edit: Super, the first was sent from my iPhone and did not seems to have been accepted, but after creating the second post the first also showed up. So sorry for repeating my self
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
This was excatly what I experienced whan changing the clock devider value and not having added the 47k pull ups.
You got it fixed by performing the steps you describe?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Yes, that's most likely a hardware problem(wired setup), it's working even in our custom h/w without these sdio frequency change
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Just to clearify, you have it running on both a prototype setup and a custom hardware?
All the connections on my prototype setup seems to be OK, and reads 0-1ohm, so mabey the length of the wires are not matching or too long?
The custom HW has also been fitted with the 47k pull ups ?
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi,
To setup STM32F429 Discovery board with BCM43362 module please take a look at this blog, SN8000 WiFi Module with STM32F429 Discovery Board. For STM32F429 Discovery board there are no external pull up needed.
Seyhan
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I have made some progress with the scan snip. I noticed that the guides stated that the SN8000 module should be plugged into the EVK board. As soon as I did this, it started scanning and seems to be running fine.
I'm trying to get it to work when it is not plugged into the EVK board, and simply use the SN8000 as a stand-alone module. For what I can see on the SN8000 schematic, the only thing I can see is 47k pull up resistors on the data lines (except CLK) and the VDD & GND.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
I just received my custom PCB, it works like a charm. The PCB does not have pull up resistors and STM32F4 MCU and the WiFi module is very close to each other.
My conclusion is that many of the problems experienced, was caused by the use of prototype setup.
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Glad to know your module is working ..