Doug, I've been in the same situation. Here is basically what I believe to be correct based on multiple conversations with ITS and UL regarding UL 1950 3rd Edition and EN60950 amd 4.
1.) At first glance, it would appear that UL 1950 3rd Edition would require basic insulation between your T1/E1 and SELV. This is pointed out in Table 0.1 under basic insulation. However, if your SELV is ungrounded, note 4 states that basic insulation only applies if your E1/T1 voltage exceeds SELV limits under normal operating conditions (i.e. 60 VDC). Of course, only operational insulation is required between TNV and SELV if SELV is reliably grounded per the operational section of Table 0.1. Section 6.2.1.2 states that basic insulation or better must be used between TNV and unearthed SELV, however UL and ITS have agreed that note 4 of Table 0.1 superseded this requirement and the basic insulation is only required if the TNV voltage exceeds SELV limits. 2.) EN60950 amd 4, IEC950 amd 4, and the proposed incorporation of amendment 4 into UL 1950 3rd Edition all support the above conclusions. Amendment 4 creates 3 types of TNV circuits. An T1/E1 circuit that is dry (no DC) would be a TNV 1 circuit. As such Table 19 shows that all that must be accomplished between TNV 1 and SELV is meeting clause 6.4.1 which is basically just a dielectric test of 1000Vrms although you could also do the impulse test. There are also provisions for connecting TNV 1 to SELV directly in certain instances. Norway will not accept this if the product has a ground connection. By the way, UL accepts amd 4 revisions even though it is not officially published. 3.) The tricky part is that Table 0.1 requires Reinforced or double insulation between TNV and primary circuitry. This is where calculating creepage and clearance can be confusing. In the united states, most ITE is 120VAC Therefore creepage with reinforced insulation between TNV and a 120V primary would be 3.0mm for a pollution degree 2 material group IIIa or b based on Table 6. If your product only had basic insulation between SELV and primary (1.5mm) you would still need 1.5mm of creepage between TNV and SELV (basic insulation at 120V) to achieve the reinforced insulation. The voltage on your Telco is insignificant compared to the primary voltage. In the U.S. it is acceptable to obtain double insulation by using basic insulation between primary and SELV and then reliably grounding your SELV circuit which acts as a earth screen. In Europe however this is not always acceptable since some countries either do not use ground or have very unreliable grounds. Since Europe uses 230/240VAC, you can extend the above example to that and get a creepage number of 5.0mm. 4.) The easy way to do this is to make sure you have reinforced between SELV and primary, this eliminates the need for creepage and clearance between your telco circuit and SELV since reinforced exists already between SELV and primary. Of course this is overkill in some countries and the U.S. where you could ground your SELV and have the basic insulation between SELV and primary to create the double insulation required between TNV and primary. 5.) Another easy way is to power your product with a reinforced wall pack transformer. Hope this helps, Jim James Wiese Regulatory Compliance Engineer ADTRAN, Inc. 205-963-8431 205-963-8250 FAX jim.wi...@adtran.com >---------- >From: Doug McKean[SMTP:dmck...@paragon-networks.com] >Sent: Tuesday, January 27, 1998 6:20 PM >To: Treg Discussion Group; nebs@world.std.com >Subject: Re: telco spacing, listings, etc. > >Dear People, > >Well, I'll be the first to say it - >I must be the dumbest person around. > >After having several very patient souls email me >off-line with suggestions I'm still in a quandary. >That is absolutely no reflection on these good people. > >If my product connects to a dry T1 line, the signals >on the T1 line are 3 volt signals. I have a very >technical report detailing results from the >signal mask. In some cases, the report even specifies >milli-volts. All digital. No analog. > >Someone did mention (off-line) that the US views the >T1 line as SELV with the capability of hitting 150 volts. >So, you go to Table 6, UL-1950 and lo n' behold you get >1.6 mm for Basic. That's (TIP or RING) -> GND and >(TIP or RING) -> the other side of the line isolation >transformer. Imposing SELV Basic insulation requirements >thus causes testing to be done at 1,000 volts (Table 18). > >He went on to say that Europe considers the E1 line >as NOT SELV with the capability of hitting 250 volts. >They impose primary voltage standards. Thus, you go >over to Table 6, EN-60950 and lo n' behold you get 2.5 mm. >BUT, that's only for That's (TIP or RING) -> GND. >Spacing to traces on the other side of the line >isolation transformer are doubled for reinforced >insulation and that is 5.0mm. Imposing primary >voltage requirements thus causes testing to be done >at 3,000 volts (also Table 18). > >Both standards agree of 0.4mm for inner layers. > >Now, standards and requirements for primary and secondary >traces outside of telco? No problem. > >But, I'll be damned if I can get someone to tell me (no insult >intended to anyone - maybe it's my stupidity) the rationalization >to "working voltage" of the T1 or E1 line that gets me some >resolution to Table 6 in either standard. And I'll be damned >again if I can find anything in my UL, FCC or CTR reports >or standards that mention anything close to 150 volts for >domestic stuff or 250 volts for European. > >Didn't mean to offend anyone that has helped so far. >The discussion both on and off line has been an eye opener. >Just very frustrated. I'm trying to define and write up >some agency design criteria for engineering. > >And don't be afraid to email me off-line, either. > >mailto:dmck...@paragon-networks.com > >Regards, Doug > >
