We use the CC2650MODA in our product, and have samples with good (>15m) and and bad (<5m) wireless range.
We have the module connected to a Cortex A9 via SPI and are monitoring the RSSI and PER from the module.
We see that the RSSI of a good and bad module are identical but PER is around 0.3 to 5% for a good and 5 to 30% for a bad module
We have extensivly checked the "good" and "bad" samples and have looked at the modules supply,firmware and layout of the board it is placed on.
We know that we do not 100% follow the deisgn rules for the layout. We have no traces in the antenne area under the module.
There is however a 3V3 plane which also runs up to the side of the module in the antenna area. We will change the layout but are unsure if this will fix all the range issues.
Is the incorrect layout a very likely root cause for the bad range of some samples?
Does anyone have suggestion of other root causes of test to find the root cause?
