Why -48V for telephone lines, and not positive?

Don't think pinhole leak. You are assuming modern insulation. Think cotton insulation that may get damp - that's all they had when this stuff was invented.

If negative was grounded, the electroplating effect would take copper away from the wires and plate it on the grounding rods/stakes. With positive grounded, the electroplating effect takes copper away from the grounding rods and plates it on the wires. Grounding rods are made to be much thicker and easier to replace.

Extra credit question: why do most electronics have negative ground? Hint: think about the physics of vacuum tubes.

BTW, in the old manual switchboards, tip was chosen to be positive/ground and ring to be negative/power because the tip is more likely to contact some exterior grounded surface.

It is indeed to reduce galvanic corrosion in all hardware contacting ground. In most cases, this will include chassis and grounding hardware - those components which by-and-large inevitably make this contact.

This is stated emphatically in NEBS GR1089 sect 72 and 73.

It is also a control method used, along with other types of cathodic protection, in marine, civil and large industrial installations. The intention is to polarize metalic structural hardware with respect to its immediate environment, be it concrete, dirt, salt water or even some kinds of air.

RL

Yup. Stroger Switches were invented in 1884--by an undertaker.

Do you think they used -48V back then?

A lot of choices in this world are simply convienience--or just arbitrary. It always seemed logical that the battery level was raised as soon as distances became significant. I have long believed--with no proof-- that the polarity was selected arbitrarily way back when. That the galvanic thing worked out well seemed like serendipity.

It was less standardized. Short loops were as little as -12V, long loops could be more than -48V. It depended on who was selling the switch and equipment too.

Large parts of the civilized world use -24V as the default on telco loops.

Tim.

I would imagine that the requirements for long(ish) distance communication had been worked out long before, by the Morse telegraph companies (you know, the clerk heroically tapping away with the arrow stuck in his back, dies as the cavalry rides over the hill), though IIRC they used single wire, earth return circuits. The polarisation to avoid corrosion of the wires would have come out of that.

Paul Burke

"Winfield Hill" skrev i en meddelelse news: snipped-for-privacy@drn.newsguy.com...

I would not be surprised if the -48V was what they arrived at by stringing batteries together until "it worked" ;-)

I ran across a web page that says the negative polarity (positive side grounded, IOW) only became standard in the US after WWII.

Best regards, Spehro Pefhany

I doubt that there was any standard in the days before electrical amplification. My guess is that for a long line the voltage might be a good deal more than 48 volts.

I looked in my old college text on Telephony (Albert, Fundamentals of Telephony, Mc-Graw-Hill, 1943) and found no discussion on the polarity issue, but he did offer that magneto (crank) telephone systems used 3 dry cells per set, which would of course work out to 4.5 volts. However, these sets coupled to the line through a transformer so the line carried purely voice current an ringing current with no DC potential. He then goes on to say that most central battery systems used 24 volts, except for dial and toll offices, which used 48. So, I guess that as late as 1943 24 volts was the standard.

He makes no mention of the polarity issue, but I picked through a lot of the schematics of both dial and manual systems and they all show the common-positive system.

I really doubt that. I would suspect that the telephone system was one of the first real examples of engineered systems, definitely engineered electrical systems. Of course, they could have "tweaked to suit" during the developmental era, but I would guess that every nuance and implication of the voltage chosen was carefully considered once they got to the implementation stage.

Indeed. Take a look here to see how much thought they were putting into the system back in 1872...

Here is the section on voltage ang grounding. (note that this was written before there was such a thing as a "volt"!)

BTW, here is an interview with an early radio pioneer who says "normally in an arc transmitter the positive was grounded to simplify cooling problems"...

Guy Macon

Some parts of the world use(d) -60V instead of -48V. That's 5 car batteries instead of 4.

If anyone is designing equipment to run from '48V' they should be aware that there are well defined tolerances and surge amplitudes, etc. See, e.g. ETSI EN 300 132-2 V2.1.2 (2003-09)

Regards, Allan

"Guy Macon" ...

I even think paper isolation, most older PTT cabling here in the Netherlands.

Thanks, that's what I guestimated too. It came up as a car-pool discussion and this was my hypothesis.

Don't have to "think of the physics". It soaked into my reflexes by now ;-) I started at 10 with 'modifying' my first tube radio, I'm now 52 and still designing. And I still "ground the commons" of circuits (even if portable).

Makes sense - and DC adapters often prove it.

Regards, Arie de Muynck

"legg" ...

Thanks, this is the best proof for my hypothesis.

I knew, but was still speculating if it was also the reason for the negative voltage on telephone lines. The dampness on the old telegraph isolators would cause a small current - and galvanic corrosion. After an industry switches to a certain polarity, they would stick to it.

regards, Arie de Muynck

Most telegraphs had a different system: a battery on each end. This resulted in negative ground at one end and positive ground at the other end, and ground halfway between positive and negative somewhere in-between.

The line would be the voltage of the battery at the recieving end. This wouldn't matter because telegraphs are current-operated.

"Guy Macon" ...

Only during signals - a low percentage of the time, I assume. And a floating line in the meantime.

Regards, Arie de Muynck

"Guy Macon" ...

Got it - on unidirectional systems. I thought they were all half duplex to save on wiring.

Regards, Arie de Muynck