I would like your opinions/advice regarding an unexplained observation between a U.FL-terminated antenna and an SMA-terminated antenna. Both antennas are from the same manufacturer and normally tuned to the 868/915 MHz band for IOT / ISM use, the only difference is the connector termination, either U.FL or SMA.
So the measurements made on a custom made radio operating in the 868 MHz band using the U.FL-terminated dipole antenna shows some behaviour that is different from the same antenna but terminated with SMA.
The transmitter RF power is about 500mW and it comprises of regular and continuous 40ms Tx pulses spaced about 100ms apart. The RF power amplifier supply is monitored with a CRO and its nominal at 3.6V. The transmitter board source current is monitored and supplied from a 3A linear power supply.
So when the transmitter is operational with a standard antenna, the source current drain peaks at about 800mA. This comprises of the peak current drawn by the RF PA (a CC1190 from TI) and then the rest of the steady drain by the other circuitry on the prototype.
The U.FL antenna is tested first (with an SMT U.FL connector on the PCB) and the expected peak current of about 800mA is observed. However when the antenna is touched by hand, the current peaks increase to > 1.6 A! The power supply also dips from 3.6V to about 2.1V. When it's released and floating in air, the current returns to the normal 800mA peaks. When touching the antenna and rotating the U.FL termination, the effect seemed to be moving in & out, in an unpredictable way, almost like it could be a connection problem...
Now one would say, sure this is caused by loading of the antenna by an RF absorbing mass or a similar effect, but we are not observing the same effect with the SMA antenna.
So to compare with the SMA antenna, we replaced the U.FL with an SMA edge-mount connector and used an SMA-terminated antenna of the same type. The tests were repeated, hand-on, hand-off and there was no increase in peak current drains as observed before!
So now you know why one's scratching their head. If the "loading" effect would be assumed it should be the same effect for both, no? Well it's NOT the case and we are wondering if there's something else at play, more along the U.FL vs SMA termination, bad cable or connection ?!?
So generally speaking the U.FL-terminated antennas I've seen are for Bluetooth and/or WiFi use, and then perhaps for more recent IOT applications. And SMA have been in use for decades, well known for their reliability!
Has anybody had similar experiences and is/are aware of limitations of the U.FL-style termination, connection issues such as grounding etc. that may case this unexpected behaviour? Sure thing, SMA is well known for being a very robust, threaded, consistent connector termination widely used, but I was not expecting U.FL to cause such apparent problems to do with grounding or bad contacts somewhere...