- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Hi there
I'm using an AP6212A that contains an BCM43430 on the RedBear Duo and on the RedBear IoT pHAT with a breakout board on my Mac. The BCM43430 on the Duo has a 26 Mhz crystal, while the pHAT uses a 37.4 Mhz crystal.
Problem: after power-up, the patch ram from WICED Studio 4.1.1 is sent on the Duo (26 MHz), then H5 is enabled. Everything seems to work fine, but when HCI LE Encrypt is sent via H5, I don't get any further messages from the BCM43430. The Host is repeating the HCI LE Encrypt command over an over, but no nothing is received.
Interestingly:
- there's no problem when using H4 on either platform
- there's no problem on the Mac using the pHAT with the 37.4 MHz crystal
I'm not convinced that the different clock configuration could cause this and I don't have a breakout board with 26 MHz to test on the Mac.
My workaround for now: calculate AES128 in software instead of using the Bluetooth Controller.
Any ideas, comments, suggestion?
- Labels:
-
Wifi+Bluetooth
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
Some ideas,
1. Some different uart speed. If long packet posible errors.
2. Different firmwares for 26mhz and 37,4mhz
3. Uart wired connection bad. Wires must be gnd rx gnd rts gnd cts gnd tx...
4. Maybe in pHAT board ap6212 without a letter?
Darius
- Mark as New
- Bookmark
- Subscribe
- Mute
- Subscribe to RSS Feed
- Permalink
- Report Inappropriate Content
hi darius, thanks for your suggestions!
My main hope is that Cypress verifies this report and checks if it's a bug in the firmware and if no, would have any ideas what's causing this specific problem (only h5, only this command), or, if if yes, releases an update/fix for it. The firmware for both setups was from the latest WICED Studio 4.1.1, one from the 43xxx/../43438/26Mhz and one from the .../ 37_4Mhz folder.
UART speed was kept at default 115200 in all cases. UART wiring should be fine since it works e.g. correct with h4, and it always hangs at exact this command, independent of the time when it was sent (early in the boot sequence, after full boot up of the stack).
Both boards have the same AP6212A and I've verified their local name. RedBear used the AP6212 for development, but shipped the RedBear Duo with the AP6212A. The IoT pHAT came later and they used the same module they already had experience with.
Cheers, matthias