Mr Hughes and Mr Tarver Thanks for your replies. I probably "spammed" the members with poor wording. When a device (component power supply in this case) requires (as per conditions of acceptability) a certain air flow, for specified combination of output power and/or ambient temperature, I interpret the relevant clause as 1.4.12.2; which requires measurements "...made at the least favorable ambient temperature..."
So, assuming I measure a max temperature of less than 120 deg ( - measurement uncertainty) in my (Class B) isolation transformers, using embedded thermocouples, for all rated combinations of rated air-flow and power outputs, then the xfmrs pass the heating test. But, using UL60950 3d ed and EN60950:2000, for 50 deg ambient, the max temp is greater. Please advise if any errors in my interpretation. thanx. luck, Brian -----Original Messages----- Brian - Read 1.4.12.2 for an explanation of what is meant by a "temperature dependent design." Per 1.4.12.1, Tma is your product's maximum rated operating ambient, Tamb is the ambient temperature the testing is performed in. If your product is only intended for use in a 25C ambient and the testing is done in a 25C ambient, you're right, the terms cancel out. But, if you product is intended for use in a 40C ambient ad tested in a 25C ambient, the term is nonzero and adjusts Tmax down by the difference of the two ambient temperatures. Regards, Peter L. Tarver, PE From: [email protected] [ mailto:[email protected]] Sent: Monday, June 02, 2003 12:55 PM To: [email protected]; [email protected] Subject: re: 60950-1 temperatures Brian, I think that my Email of 30/4/2003 should answer your questions. In regard to your second substantive para, you will see that it is not always required to test at the highest ambient temperature. An extract is below... Richard Hughes Are you REQUIRED to conduct heating tests at the highest rated ambient temperature - in your example 30 degC? Well, that depends on whether your equipment has a temperature controlled fan or similar. If your equipment has no such fan or control mechanism then it is permitted to run the test at lab ambient and extrapolate, see 1.4.12.3. For such equipment therefore you will be able to run the tests just the way you seem to have done in the past. However, if your equipment does have such a such fan or control mechanism then it could/would be unfair to run the test at (say) 20 degC and then add 10 degC (or 10 K if you prefer) to your results before comparing the 'corrected' results against the limit values: after all, in such equipment the fan could possibly run twice as fast at 30 degC as at 20 degC. Hence the requirement was added in part to make the test fairer. It was also added because in the future it is expected that there will be a Part 22 that will cover Equipment Installed Outdoors - and such equipment will obviously work at temperatures in excess of 25 degC in the summer (well, in many parts of the world). For manufacturers running their own tests there can also be the. practical advantage of combining product safety tests with product functionality tests (i.e. making sure the equipment doesn't fail to function when run at its max/min rated temperatures). Why change from limits based on a temperature RISE to ABSOLUTE temperature? As indicated above, there are reasons why you may need to run tests at the maximum temperature specified by the manufacturer rather than at lab ambient and then convert. When non-safety engineers used IEC 60950 etc. it was always previously necessary to explain to them that the figures given were RISES above an ambient contained elsewhere in the standard rather than absolute figures. In addition, if the equipment was not designed for a maximum temperature of 25 degC but 32 degC or 40 deg C (to pick a couple of common examples) then it was also necessary to adjust the RISE figures by subtracting 7 K or 15 K (respectively for the examples given). What a complicated way of doing things! In addition, why specify the limits for Normal conditions as a RISE but those for Fault conditions (e.g. see Table B.1) as an ABSOLUTE temperature? All the best, Richard Hughes
