An error is your confusion of limited current circuits, IEC clause 2.4, with limited power circuits, clause 2.11. A Class 2 (not to be confused with Class II) transformer equates roughly to the limited power circuits of clause 2.11. Note also that a Class 2 transformer is not necessarily compliant with the two levels of protection required of SELV circuits.
Reviewing the circuits involved, you have - ISDN-U interface, a TNV-1 circuit (exposed plant, low voltage) - RS232 circuit, an SELV circuit (not exposed plant, low voltage) - POTS ports, a TNV-2 circuit (not exposed plant, TNV voltages due to the ring circuit voltages) The ring circuit is not limited current since it is not below 0.7 mA peak or 2mA d.c. None of the others are limited current circuits either. Note that the TNV-2 circuit must meet the voltage limits of clause 6.2.1.1 b) and appendix M (should not be difficult). Since the product is powered by a low power circuit in compliance with clause 2.11, components do not require a fire enclosure (per clause 4.4.5.2). That's about all you gain for the limited power circuit. You still have an electrical enclosure (of TNV circuits) to deal with. Your next decision is the protection between the TNV-2 circuit and the SELV circuit (see clause 6.2.1.2). You have two choices. First, you can provide an isolating barrier of basic insulation, which involves isolation in the ring generator, isolation in the signal paths between the SELV circuits and the SLIC chip with its associated TNV-2 circuitry, and isolation of the dc power to the TNV-2 chips. The second method (a condition to table 19) is to ground the SELV circuit so that faults in the TNV-2 circuit will not affect it. This is usually a reasonable choice only for commercial equipment since about half the residential environments have no ground available. Note that this ground must be a permanent connection, which means other than through the power plug. Dropping the ring voltage to SELV limits would change the TNV-2 circuit to SELV status, but the best you could do would be a 42V square wave which might drive a specifically provided phone but probably not the general unknown product such as fax and answering machines. The reason for the isolation needed between TNV2 and SELV is that the phone line, which is not particularly well protected, might get grounded somewhere. The other side of that phone circuit, which has the ring voltage on it and is tied to SELV, is connecting that ring voltage to accessable SELV circuits, making the ring voltage appear on any chain of stuff connected to your RS232 port. Peter Tarver wrote: > > Jim - > > I offer the following. I hope you'll take the mild chiding as my own > asininity, rather than anything serious. > > Regards, > > Peter L. Tarver > Nortel > [email protected] > > >---------- > >From: JIM WIESE[SMTP:[email protected]] > >Sent: Friday, February 27, 1998 6:32 AM > > > >Dear Colleagues, > > > >I have a question regarding safety of a customer premise device that is > >typically owned by a residential customer. It is an ISDN device with a > >U-interface, a RS-232 data port, and two independent POTS ports. The > >POTS ports are used to provide analog service to a standard phone via > >the ISDN line. The POTS ports are not intended to leave the customer > >premise or hook up to outside plant connections. Assume the device is > >powered by a Listed Class 2, 9-12 VDC wall transformer. > > > >I would assume that most products of this type are Listed to UL 1950 2nd > >or 3rd edition or UL 1459. My question specifically relates to UL-1950 > >3rd edition with the assumption that UL now recognizes clauses from > >amendment 4 of IEC 950 (TNV 1-TNV3 specifically), which they do. > > Unless using UL1459, CSA950/UL1950, 3rd ed. would apply. And you're > right, they do use the Amendment 4 changes, in a semiunofficial way, of > course. > > > >Assume that an engineer wanted to take the 9-12 volt input, and step it > >up to 65 or 70 volts DC, then feed the 65 or 70 Volts DC into an off the > >shelf SLIC (subscriber loop interface circuit) IC. The SLIC then turns > >that into 65 or 70 Vrms for ringing, which will go out on the POTS > >interface. > > > >Questions: > >1.) Since the unit is powered by a class 2 wall transformer and > >therefore everything inside the unit would be considered a limited > >current circuit (except the ISDN port) would these voltages that exceed > >SELV limits be allowed without isolation between the 65-70 VDC circuits > >and the user accessible circuits (RS-232)? > > Everything inside would _not_ be considered a "limited current circuit." > I suspect you mean "limited power circuit," per Subclause 2.11, but a > Class 2 input wouldn't necessarily mean the 65-70Vdc would meet the > requirements for limited power circuits either. > > For the circuit voltages that are converted at other than SELV voltage > limits, the parts of those circuits not at SELV must remain dc (</= > 150Vdc) and comply with Table 8 or 9 with regard to power and current > limits to be considered limited power circuits. As soon as the voltage > is modulated beyond the +/-10% limits accepted for a circuit to be > considered dc (interruption by, say, a relay is not addressed in > Subclause 2.11), the circuits are not and cannot be treated as SELV or > as limited power circuits. As a side reference for the last sentence, > see Note 4 of Tables 11(a) and 11(b) of the 1996 US NEC. > > After the 65-70Vdc is modulated out of dc, TNV-2 would seem to fit the > circuit/application definition. > > > >2.) Excluding the ISDN port which would be TNV 1, are the POTS ports; > >TNV 2, SELV, limited current circuits, something else? > > This fits the definition of TNV-2. > > > >3.) Is there a limit on the ringing voltage if everything in the unit > >is considered a limited current circuit? > > The ringing voltage/current will still have to comply with Clause 6 and > Annex M. Although it may be somewhat novel to try to treat it as a > limited current circuit (again, I don't think that's what you really > meant). [If it did comply with Subclause 2.4, would there be enough > energy available to drive the ringer of a 500 or 2500 set (an electronic > set might be able to respond)?] > > [If the output can meet the requirements for a limited current circuit, > would it be difficult to comply with Subclause 2.2.6, Clause 6 and Annex > M, etc.? I suspect not and it may be easier, since there aren't the > capacitive circuit impedance limitations as there are for limited > current circuits. A limited power circuit may need additional > treatment.] > > Is your concern that you can't or don't comply with the ringing > cadencing of Annex M, that you want to apply limited current (power) > circuit criteria? If so, another SLIC or tweaking of the supporting > circuitry is called for. > > > >4.) Are creepage and clearance an issue if the unit is comprised of > >limited current circuitry, SELV, and TNV 1? > > There are specific cases in Amendment 4 where creepages and clearances > may not be a difficult issue to comply with. Read Subclauses 2.2.6 and > 6.2, if you use TNV-1 and TNV-2. I'll leave the limited current circuit > aspect as an exercise for others. > > > >Any information would be helpful, > > > >Thanks, > > > >Jim > > > > > >James Wiese > >Regulatory Compliance Engineer > >ADTRAN, Inc. > >205-963-8431 > >205-963-8250 FAX > >[email protected]
