In a message dated 97-03-13, Jim Wiese writes the following with reference to amendment 4 of IEC 950:
<< There also seems to be some confusion amongst test labs as to what "working voltage" to apply to the interface. Some say 240 VAC regardless of how it powered, even though Japan uses 100 VAC as I remember. Others say it is the output voltage of the transformer (assuming it meets the appropriate requirements to make the output SELV), and yet others have told me it is the maximum operating voltage on the telco interface. Any Thoughts? >> Jim: I will venture an opinion on this one. Under clause 6.3.3 of IEC 950 with A1, A2, and A3 (but not A4), the "working voltage" would be the AC mains voltage. This is because clause 6.3.3 is specifically targeted at a fault condition (mis-wired plug) that inadvertantly connects the chassis ground to the AC mains voltage. If the product was exclusively for Japan, the working voltage would be 100 volts RMS. The insulation barrier would have to meet the requirements for supplementary insulation with a working voltage of 100 volts, which invokes creepage, clearance, distance-through insulation, and an electric strength test). Clause 6.4.2.2 imposes an additional dielectric strength test of either 1000 or 1500 volts RMS. In Amendment 4, the test in clause 6.3.3 has been changed so that it simply points to clause 6.4.2.2. Thus, under Amendment 4, clause 6.3.3 has been reduced to a simple dielectric strength test of 1000 or 1500 volts RMS (except for Sweden, which retains the requirement for supplementary insulation at the rated AC mains voltage). Creepage, clearance, and distance through insulation are no longer invoked by clause 6.3.3, except for Sweden. However, Amendment 4 has expanded clause 6.2.1.2 to address the three types of TNV circuits. In some cases, basic insulation is required. In the event that basic insulation is required, it becomes necessary to identify the working voltage in order to determine the required creepage, clearance, and dielectric strength (distance through insulation is not specified for basic insulation). In this case, the working voltage is the voltage on the TNV circuit, NOT the AC mains. However, in the specific case of the S/T interface, not even basic insulation is required. The S/T interface is a "TNV-1" circuit, and clause 6.2.1.2 only calls out clause 6.4.1 for separation of SELV and TNV-1. Clause 6.4.1 leads back to the same 1000 and 1500 volt RMS dielectric strength tests described in clause 6.4.2.2. So, my conclusion is that under amendment 4, for your specific application (S/T interface), there is no need to concern yourself with the "working voltage" at all. You only have to deal with the dielectric strength tests in clause 6.4.2.2. For separation of TNV-2 or TNV-3 circuits from SELV, amendment 4 calls for basic insulation, and the "working voltage" for this insulation is the steady-state voltage on the TNV circuit (ringing signals are not considered in the determination of working voltage). Well, that is my interpretation of how amendment 4 changes the situation for your product. To be honest, I have not spent a lot of time with this because, to my knowledge, amendment 4 has not yet been incorporated into EN 60950. Having it in IEC 950 is nice, but in Europe, EN 60950 is the one that counts. This raises a cautionary note. When someone tell you that Japan uses "IEC 950," make sure that they mean exactly that. In particular, make sure that Japan formally recognizes amendment 4 to IEC 950. In some cases (such as in Europe), there is an administrative delay between the time that the IEC adopts an amendment, and the time that the amendment becomes officially recognized. IEC 950, by itself, is just a reference standard. To have the force of law in a given country, it must be formally adopted or recognized in some way. Let me close this with a question. Does ANYONE on treg know when amendment 4 will be formally adopted as part of EN 60950? I believe that the critical step is for amendment 4 to be referenced in the Official Journal of the EC. I have heard that amendment 4 went out for voting last fall, but I have heard nothing since. I sure would like to know when I can begin using amendment 4 in Europe. Joe Randolph Telecom Design Consultant Randolph Telecom, Inc.
