Getting a voltage scope of the power-supply pin at the MCU of the board will give you quite a bit of information on whether the power supply is stable enough for the board running.
You could try putting a very small inductor inline from the battery to the board power-supply to see if that improves performance? (330 Ohm or less)
You shouldn't need to change code to allow for battery versus USB power. The battery-powered device would ideally want to have code to utilize less power consumption and maximize battery life, but otherwise the only difference I could see would be ECO Cap Trim settings, and BLE Component settings.
The BLE component has a "low power" option, toggling that might change performance.
The ECO Cap Trim settings also may affect the RF performance (located under Clocks->ECO->"Configure"
The place I am connecting the battery is the VDD pin of the module which according to the schematic already has a 330 ohm inductor supplying the SoC VDDD power plane. Connecting a scope to the VDDD pin of the module, scope AC coupled, 50 mV/div vertical, 1 us/div horizontal, I can see about 25mV peaks of noise on the Pioneer board module and significantly less on the battery powered module.
Was curious if in PSoC Creator Design Wide Resources/System/(VDDA.VDDD,VDDR) settings made a big difference. The default voltage appears to be 3.3V whereas the battery would be 3.0V.
The code is identical for the Pioneer board module and stand alone module versions. ECO cap trim and BLE Component settings good things to check, none the less.
Still really odd there should be such a significant performance difference, makes me wonder if i am missing something obvious.
Thanks for giving me things to look at, anything else comes to mind please let me know.
You're welcome. iirc, I looked up what that voltage setting under the System is for, and it was more or less a documentation/numbering for the compiler to use?
The development kit should have a power supply pin somewhere on it, perhaps connecting the AAs to the CY8CKIT with the module on it to see if it is a difference between the module being plugged into the dev kit board?
(Obviously, this would burn more batttery power, but might help for determining differences between the two modules)
Okay good, does not sound like those voltage settings In System are critical then.
I did a similar test to what you suggest and used the stand alone module in the development kit. It appeared to exhibit the same erratic behavior as it did stand alone. Thinking it might just be a bad module and have ordered another.
Unfortunately can’t risk taking the working development kit module and trying it stand alone as we need that working for client demo. Maybe the power switch bounce from the battery pack somehow damaged the stand alone module?
Sent from Mail for Windows 10
6A697AD0735B4B59831F24D4F54B1DC3.png 148 bytes