Bluetooth RF performance testing has been a key aspect of Bluetooth Qualification since its inception back in 1998. The first RF test requirements were released in 2003 which included a reference to 300-328 which stated the following procedure for extreme temperatures:-
For tests at extreme temperatures, measurements shall be made in accordance with the procedures specified in clause 184.108.40.206, at the upper and lower temperatures of the range as follows:
– Temperature: -20°C to +55°C; Where the manufacturers stated operating range does not include the range of -20°C to +55°C, the equipment shall be tested over the following temperature ranges:
a) 0°C to +35°C for equipment intended for indoor use only, or intended for use in areas where the temperature is controlled within this range;
b) over the extremes of the operating temperature range(s) of the stated combination(s) or host equipment(s) in case of plug-in radio devices. The frequency range as in clause 220.127.116.11 and the e.i.r.p. limit in clause 18.104.22.168 shall not be exceeded. The temperature range used during testing shall be recorded and shall be stated in the user manual.
But recently 300-328 has now been updated to state:-
Where tests at extreme temperatures are required, measurements shall be made over the extremes of the operating temperature range as declared by the manufacturer.
Basically, temperature and voltage extreme testing have been significantly reduced to only output power and sensitivity tests. Majority of the receiver tests that requires testing at extreme conditions have now been removed.
This change does introduce a number of positive and negative attributes for manufacturers. The obvious positive factor is reduced Bluetooth RF testing, hence it is now cheaper and easier for manufactures to bring fully Bluetooth Qualified products to market. The negative aspect is assuring if their device will work at extreme temperature and voltages.
Bluetooth is an established standard which means there are assumptions that the RF specification is mature hence there are no need for temperature and voltage extremes. Being a BQE for over 5 years and leading the Bluetooth service at UL for several years, one of the most common failures I have seen in the technology is that the RF would fail at both minimum and maximum temperature extremes for multiple test cases. Two test cases that commonly failed were carrier frequency drift and modulation characteristics, this is a clear indication that there would be a lot of retransmission affecting the quality of a connection. Some may also argue that Bluetooth devices can cope with the high drift but there is a lack of evidence to fully back this up. Normally it is not the chip set that is at fault but the peripheral components that don’t match the temperature range of the chip set.
At UL, we believe there are still significant benefits to performing testing at temperature extremes and will continue to offer this service on an optional basis. This helps our customers to have confidence that their products will perform as expected at extreme conditions after it has been released to the market.
Temperature extremes used to be tested at a minimum range of 0-30 which are not temperatures that are unlikely to happen in everyday life so it really comes down to manufacturer choice and where that product will be used. As a simple example, in the automotive industry, the Bluetooth module could be placed deep within the engine, where temperatures can change dramatically.