Conflict of Interest Notice: My employer makes component SMPS I do not understand why the PFC section is being considered for temperature measurements; pay attention to insulation materials and components that provide basic or reinforced insulation. The important component is the isolation transformer and optos.
If you have rated your unit within the temperature and power ratings of the SMPS, and can demonstarate that there are no single fault conditions that will exceed corresponding test conditions listed in the CB report, then I do not understand why you are repeating Heating tests. If your "normal" operating conditions exceed the rating of the SMPS, find another supplier.... It is important that the original test conditions, as specified by the manufacture or the examing agency, be preserved. A component P.S. that was tested with an enclosure may be dependent on that enclosure to provide proper air flow over critical components. On occasion, I have supplied a "test only" unit to customers using a special chassis, where the thermocouple has been embedded in critical magnetics during manufacture, and that allows t/c wires proper egress, without disturbing desired airflow characteristics. The only acceptable thermocouple wire, for measuring magnetic component temperatures, is type "T". The *worst* possible t/c for mags is type "J"; type "K" can be used if you are careful with placement and routing. Also, the thermocouple instrument must have extremely hi-Z and fully differential inputs, or noise can be coupled to some units. In general, a decent SMPS design should not be susceptible to noise induced by a t/c for most components. Most likely, your instrument is providing a leakage path, or is cross-coupling noise from primary to secondary that is affecting the feedback. luck, Brian -----Original Message----- From: Van Compernolle, Eric [ mailto:[email protected]] Sent: Monday, January 26, 2004 8:03 AM To: '[email protected]' Subject: heating (temperature) test on SMPS with CB report EN 60950 Dear all, More and more we are using purchased Switch Mode Power Supplies for our equipment. The SMPS manufacturer delivers us then a CB report of the SMPS in order to make the test report of the equipment for the safety approval ( CEBEC and UL). Up to now we always performed the heating test of the purchased SMPS: we measured the temperature of the critical components on the smps itself to see if they remain under the safe limit. We are now having a problem with a particular power supply. The 600 W power supply is a fully closed metal box with cooling fans on the front. When we open it and add thermocouples on the PFC part of the power supply, the PFC fails. This is probably due to the fact that we are coupling in some wrong signals on a feedback circuit. We tried it several times with the same result: failure of the PFC part of the SMPS. My question is: is the heating test really necessary? The CB report states that the Power Supply was investigated in a maximum ambient of 50°C. Should it not be sufficiant to measure the temperature of the environment and if this temperature remains down the 50°C in all circumstances, can't we conclude that the heating test is not necessary and that we don't have to open the smps in order to fix the themocouples? best regards, Van Compernolle Eric
