OK, I love a challenge. I went to the lab and started doing dielectric tests on champ connectors with pad to pad spacings of .020" (.5mm). They consistently passed at 1800VAC for 60 seconds and broke down at between 1800 and 1900 VAC. Of course that will change with the CTI of the board and RH.
Secondly, I did some evaluations of spacings between adjacent channel pairs in some common channel banks that regularly are used to mix TNV1 (ISDN, DDS), TNV 2 (DSO-DP's, DSU-DP's,), TNV 3 (FXO's, FXS's), and Hazardous voltage secondary circuits TNV "?" typically DC from 120 up to 200 VDC (Fractional T1 FT1-DP's, span powering UBR1TE's ISDN, span powering OCU-DP's DDS, and other span powering channel cards). Here are the results: AT&T SLC 200 J1C265AE-1,L1, SDM76177-01, SAMANSOAMA .5mm spacing on the amp champs between adjacent channels. AT&T SLC Series 5 J1C182AE-1,L3, SD7C117-02, SMRA43B .5mm spacing between adjacent channels. AT&T DDM+ J98725DB-1,L1,A CLEI T1MRUJUX UL Listed 538E .4mm telco to telco, or telco to SELV wire wraps appear to reduce some of the spacing even more Alcatel D448 Channel Bank 628011-000-005 Part 68 Registered HLUSA 20862-XD-N .6mm pad to pad on the champ connectors between adjacent channels. AT&T D4 Channel Bank 00031769L1 .3mm between channels, 2.0mm to ground, .3mm to SELV and -48V. Fujitsu FDLC Channel Bank FC9601SFC1 1989 ser # 00392 50 pin champ connectors .35mm (.014") channel to channel. AT&T D5 Channel Bank J98743AA-2, L2 SD7C361-01 .5mm spacing channel to channel, wire wraps appear to reduce this spacing even more. I could go on but I think points out the spacing deficiency issue. In addition, the types of channel cards mentioned in the second paragraph above are mixed in channel banks all over the country in almost every central office. That is the beauty of channel banks. As long as the information can be divided into 64Kbps chunks, it can be assembled into a 1.544 Mbps T1 data stream and cross connected into a digital switch, be shipped across the country and unassembled at the far end with TNV 1, TNV 2, TNV 3, and HVSC TNV? intermingled there until it reaches the end customer. Jim Jim Wiese ADTRAN, INC. 901 Explorer Blvd. P.O. Box 140000 Huntsville, AL 35814-4000 256-963-8431 256-963-8250 fax [email protected] >---------- >From: Jan Purwin[SMTP:[email protected]] >Sent: Wednesday, April 29, 1998 4:30 PM >To: [email protected] >Subject: Re: UL 1950 3rd, and IEC 950 > >Jim > >Since you are refering to a Champ conneector or RJ21X connector that >assumes that ALL circuits are from Same TNV voltage. >I personally have not seen mixing T1 and PSTN in the same bundle >When we do FCC part 68 all circuits are treated same and tied together >than isolation is between TNV1 and SELV. >Unfortunatly T1-CSU, BRI-U is treated as a TNV circuits even there is no >ring >voltages present. > >I am not sure where you got an idea that .3 mm will pass 1000VAC test. >The two bear conductors at the sea level will arc across at the rate of 1mm/ >1KV >of potential > >JIM WIESE wrote: > >> I appreciate your comments, and I did not imply telco's do not provide >> protection. In fact they do a great job. However the spacings on an >> AT&T D4 channel bank (for instance, I could list endless others) are >> less than .3mm from the tip/ring of one channel to the tip/ring of >> another channel (look at the spacings of the "amp champ" pins). Since >> adjacent channels can have cards that are TNV1, TNV2, TNV3, or Hazardous >> voltage secondary circuits (190VDC fractional T1), separation is <.3mm. >> According to UL 1950 3rd edition, the creepage at 190VDC is 2.0mm (1.5mm >> at 120VDC). .3mm is substantially deficient with regard to UL 1950 3rd >> edition. In other words the network has <.3mm between different TNV's >> at the C.O., yet UL 1950 3rd Edition would require 2.0mm at the >> terminal device. What is the point if the network is deficient in >> spacing. Also network equipment that happens to be evaluated for a >> listing is almost surely done to UL 1459 or a UL category such as >> "DUZO". The only thing required is a dielectric test, not spacings. By >> the way, .3mm will pass a 1000VAC dielectric in most cases. >> >> I would also be very interested in information regarding any Bellcore >> document that addresses spacings between adjacent circuits. The only >> thing I can find is a dielectric test for embedded power supplies that >> requires 1000 VAC between the embedded power supply and the telco >> network. Bellcore GR-1089-CORE does not address this, but only >> requires equipment to pass the Lightning and Power Fault tests in >> section 4. >> >> Jim >> >> Jim Wiese >> ADTRAN, INC. >> 901 Explorer Blvd. >> P.O. Box 140000 >> Huntsville, AL 35814-4000 >> 256-963-8431 >> 256-963-8250 fax >> [email protected] >> >> >---------- >> >From: Jan Purwin[SMTP:[email protected]] >> >Sent: Wednesday, April 29, 1998 10:13 AM >> >To: [email protected] >> >Subject: Re: UL 1950 3rd, and IEC 950 >> > >> >Jim you are not quite right. >> >There is a lot of protection and separation that are defined in NEBs and >> >NEDS standards. >> >Those are incorporated into the equipment that is installed in a central >> >office. >> > >> >If you have a chance to go to typical central office and look how RBOC >> >installes equipment you would >> >had better idea about protection separation and equipment construction >>that >> >goes into CO equipment. >> >Jan Purwin >> > >> >JIM WIESE wrote: >> > >> >> I am looking for some guidance in understanding the separation >> >> (creepage/clearance) requirements with regard to TNV circuits in UL 1950 >> >> 3rd and IEC 950 and its derivatives. >> >> >> >> Why is there a requirement that TNV circuits have basic insulation >> >> between TNV1 and TNV3 circuits? >> >> >> >> I am asking the question for the following reasons: >> >> >> >> The telco network provider is generally exempt from listing requirements >> >> altogether. They have no restrictions on separation of circuits and >> >> have TNV 1, TNV 3 and span powered circuits (200 VDC HDSL, T1 etc.) >> >> intermingled at the central office in channel banks, cross connects etc. >> >> They provide these services via cables that are spliced, cross >> >> connected and intermingled in the Outside Plant. They are also >> >> intermingled without regard to spacings at the distribution and >> >> demarcation points. Since it is a fact that these services are not >> >> separated by "basic" insulation, why would it be important to separate >> >> TNV1 and TNV 3 in the terminating equipment. To me it seems like a >> >> chain made of paper, and at the end of the chain the standards are >> >> requiring a steel link. >> >> >> >> Secondly, do the standards assume that faults of TNV 1, TNV 3, and other >> >> high voltage telco services could be faulted by the telco provider >> >> inadvertently and therefore this type of fault is accounted for by the >> >> dielectric test (and other restrictions) between SELV/chassis and TNV 1, >> >> and TNV 3. >> >> >> >> Any input would be appreciated and helpful. >> >> >> >> Thanks, >> >> >> >> Jim >> >> >> >> Jim Wiese >> >> ADTRAN, INC. >> >> 901 Explorer Blvd. >> >> P.O. Box 140000 >> >> Huntsville, AL 35814-4000 >> >> 256-963-8431 >> >> 256-963-8250 fax >> >> [email protected] >> > >> > >> > >> > > > > >
