Hacker News new | ask | show | jobs
by russell_sears 742 days ago
My grandfather used to work on telephone field network boxes (e.g., the big boxes full of messes of wires in residential areas).

One job was to look for outliers in the network, and they spent time studying areas with an unusually large number of issues and ones with an unusually low number of issues.

There was a part of the phone network that was decades overdue for an overhaul but had no issues, so they inspected it. (This was decades ago, so the replacements for this antique could be modern day museum pieces).

When they took a look, everything was corroded beyond reason, as expected. However, the connections were still low noise / low resistance.

The old boxes used some sort of post connector and a crimp. It had something like two or four points of contact for redundancy (all contact points would need to corrode before it failed).

In the boxes with no failures, the (long gone) technician simply stuck the end of the wire into the post crimp hole, then wrapped slack wire around it a dozen times.

This gave it 100’s of contact points, and (after reverse engineering the technique) it took something like 1/10th as long per connection.

Sadly, we’ll never know if the installer was a genius, lazy or both.

Anyway, the crimp connectors in fig 19-25 and 19-28 of the nasa article look like the same concept but turned inside out.

2 comments

I'm not sure why more of those didn't use wire-wrap, it's super fast and is really reliable. I've seen some bell terminals that were wirewrap but punchdown was more common in my experience.
As a young guy who was (un)lucky enough to be exposed to punchdown terminals as a field technician for AT&T…

Fuck punchdowns and everyone who said “good enough for me” when deciding to adopt them.

On a brand new install? Sure it works great!

In a box that has been sitting in the rain for a little over a decade? No.

Laziness tends to be a great catalyst for genius work