I recently picked up a couple BlueTooth v1.2 Stamp Modules (neat little gadgets by the way) and am trying is use one to receive bluetooth serial information from a remote device. Unfortunately, I seem to be having a lot of problems consistently discovering and connecting to the bluestamp when it is configured as a discoverable slave.
When I’m using a PC to connect to the bluestamp, having connection difficulty isn’t too big of a deal since I can simply hit “Search for bluetooth services” 5 to 10 times until the bluestamp is discovered. Unfortunately, this becomes a much bigger problem when connecting to the bluestamp with any remote device where the connection process is automated.
Is there anyway to improve the chances of the bluestamp being discovered? I was reading the BlueRadios AT Command document, and it looks like the “adjust page scan interval and window” command on p.14 might have something to do with the discoverability of the device, but I’m not quite sure how to use it.
Anyway, if anyone knows anything about this command or any other way to improve the discoverability of the bluestamp, I’d appreciate any help I could get.
The most likely cause for discovery issues is the placement and/or location of the Stamp relative to the final application design. If you were selling them SMT modules their RF PCB layout most certainly would be the first place to start. Second is the VCC needs to be 3.3Vdc +/-10mVPP max ripple both are obvious but over looked. Very high ripple voltage will start to affect the radios RX sensitivity.
To verify the above: remove the stamp completely from the design and connect an external regulated 3.3Vdc lab bench power supply. After you turn on the power supply clip on the power to the radio power pin to make sure you have a clean power boot up. Since the device defaults as slave discoverable you only need power and ground attached to verify the RF performance. You can use any third party Bluetooth device like a USB Class1 Dongle and you should be able to discover and connect to the Slave unit under test 10 for 10 times at 50 meters line of site (LOS). Do not set the unit flat on the ground and expect great results since most of the RF is absorbed and not radiated when the antenna is sitting flat on the ground. Just rest the unit on a card board box. Make sure the power and ground leads are away from the antenna.
Also make sure the Page Scan Window and Interval is set to factory default since this provides the best discovery performance since the RX is listening for a Mater connect request at a 50% duty cycle.
I just tried adding a larger decoupling capacitor and (I guess contrary to BlueRadio’s advice) lowering the Page scan interval while keeping the window the same such that the new duty cycle is 66%. Now I seem to be able to connect about 9 times out of 10!
Oh, one last thing in case anyone else if having the same problem I was…
Even though the AT command reference says that the default trasmit power level is it’s highest value (15 dBm), it might not actually be set to this. If you send the ATSI,14 command it’s supposed to return the current transmit power level, but if you type it in without setting the transmit power level first it simply returns “default” instead of a value. If you set the transmit power level to 15 dBm, (ATSPF,15,+) then use ATSI,14 it will return 0F.
I was having a few more problems discovering the bluestamp with various other devices, but once I tried explicitly setting the trasmit level to 15dBm, it started working much better.