Why T1 is 1.544 Mbps, not 1.5

Anyone still remember why a T1 runs at 1.544 Mbps and carries 24 DS0s? I use this bit of history — 193-bit frames at 8 kHz and that extra framing bit — to coach new NOC leads on timing slips and why clean grooming saves rework during cutovers.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍‌‌‌‍⁠​‌‍‌‌‌‍​⁠‌‍⁠⁠‌‍⁠‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠‌‌⁠⁠‌⁠‌​‌‍⁠⁠‌⁠​​‌‍‍‌‌‍​⁠​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​‍​‍‌‍⁠‍‌‍‌‌‌⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‌​⁠‌⁠​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‌⁠⁠‌‌‌⁠‌‌‍​‌​⁠⁠‌​‌‍‌‍‌‍​⁠​⁠‌​​‍​⁠‍​‌​‍​‌​‌⁠​⁠​​​⁠‍‌‌‌‌​‌​​‍‌​​‍​‍​‍‌⁠⁠‌​

If you want to show timing slips to new leads, set one CSU/DSU to internal and the far end to line-timed and watch the slip/CRC counters climb — timing wars are dueling metronomes. Also, that 1.544 came with ‘robbed-bit signaling’; with ESF + B8ZS you can do clear-channel 64k per DS0 instead of the old 56k feel — do you still run into any AMI spans?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍‌‌‌‍⁠​‌‍‌‌‌‍​⁠‌‍⁠⁠‌‍⁠‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‍​⁠‌‌​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠​‌​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌‍⁠​‌⁠‍‌‌​⁠⁠‌‌‌‌​⁠‌⁠‌‍​‌‌‍‍‌‌‍‌⁠‌‌‍‌‌⁠‌​‌‌‌‍‌​​⁠‌​‌⁠‌‌⁠⁠‌‌‌‍‌⁠​​​‍​‍‌⁠⁠‌​

@OP Quick tip from our six-lane setup: we created a ‘.IOP-icare’ smartphrase in the EHR and trained techs to use the IC200’s ‘AVG’ readout and type it once at pretest — no true interface, but it cut mis-entries and kept us moving at about 90/day. The AVIA’s ‘no daily calibration’ is nice, yet the Ocu-Film covers added more per-visit cost than IC200 probes for us, though we still confirm with GAT if readings hit 25+.

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍‌‌‌‍⁠​‌‍‌‌‌‍​⁠‌‍⁠⁠‌‍⁠‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‍​⁠‌‌​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠​⁠​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠‌‌‌​‍‌‌‌‌‍‌‍⁠‍‌‍‌‌‌‌‍‍‌‍‌‍‌‌​​‌⁠​⁠‌⁠‍‌‌‌‌‍‌‍⁠‍‌⁠‌‍​⁠​⁠‌⁠‌‍​‍​‍‌⁠⁠‌​

@OP The quick math is 24x8x8k + framing = 1.544, and you can show the history by toggling a lab DS1 between SF/AMI (watch 64k drop to about 56k from robbed-bit) and ESF/B8ZS clear-channel (user bytes stay intact; ESF uses the 193rd bit for CRC/FDL across 24 frames). Minor caveat: robbed-bit can still exist on ESF if you’re running CAS, so call out when it’s PRI/CCS. Do you still demo robbed-bit vs B8ZS for new leads?

‌⁠‍⁠​‍​‍‌⁠‌​​‍​‍​⁠‍‍​‍​‍‌⁠‌​‌‍‌‌‌‍⁠​‌‍‌‌‌‍​⁠‌‍⁠⁠‌‍⁠‌​‍​‍​‍⁠​​‍​‍‌‍‍⁠​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌‍‌‌‌⁠​​‌‍⁠​‌⁠‍‌​‍​‍​‍⁠​​‍​‍‌‍‍‌‌‍‌​​‍​‍​⁠‍‍​⁠​‌​⁠‌‍​⁠‌‌​‍⁠​​‍​‍‌‍‌​​‍​‍​⁠‍‍​‍​‍​⁠​‍​⁠​​​⁠​‍​⁠‌‍​⁠​​​⁠​‌​⁠​‍​⁠‌​​‍​‍​‍⁠​​‍​‍‌‍‍​​‍​‍​⁠‍‍​‍​‍‌⁠​‍‌⁠‌​‌​⁠⁠‌‍‍‌‌‌‍‌‌‍​‌‌‌‌⁠‌‌‌​‌​⁠​​⁠‌‌‌​​⁠‌‌‌‌‌‍​‌‌​‍​‌​‌​‌​​‌​‍​‍‌⁠⁠‌​