I found that if you had 2 adjacent pay phones you could make free calls by inverting the 2 handsets and placing the call on one phone, but putting your money in the other so the sound of the money dropping would make the operator think that you had paid for your call. Your money would be returned to you when you hung up the second phone.
Was it literally just the sound of money dropping that validated the call? Presumably this was before automatic switching and a live operator had to connect you?
I once made a call from a pay phone and after I was done it rang. I picked up. A very nice lady told me I had not paid enough and whether I could please put in another quarter or whatever. That was an interesting experience.
In my mind I thought that pay phones were fully automated, especially the count the money part.
Someone across the street could have been messing with you.
This sounds like it was from a time where you could ask to borrow a bars phone to make a free local call and it could be worth someone's while crossing the street to collect a quarter.
At one point it was. You could even record the sound on a handheld recorder and play it back. But it hasn’t worked in a long while. Pay phones mitigated this in a low tech way - by software muting the handset.
Edit: oh yeah, and after that, you’d just call the operator and tell them the keys were sticky and to dial the number for you, then you’d “insert the coins” by playing the tones.
Well, or you just third party billed the call to someone you didn’t know. That worked too.
> Pay phones mitigated this in a low tech way - by software muting the handset.
So does this mean that there was a dedicated second microphone hidden somewhere within the payphone body, that would continue to record the sound the coins made?
The coins weren't recorded, it was just a short tone that indicated 5¢. A quarter was five quick tones. A dime two.
I used to mess with the proctor test set (dialing 117 in the US at least on GTE) and only ever convinced it I inserted a nickel even with many tries using a digital recorder or a computer speaker.
After this mitigation was put into place enterprising phreaks would take a lighter to the microphone part of the handset, twist it open, cut the yellow wire, and redbox away.
It was automated. The coin drop sound (or at least what was generated for the wire by the phone) was what drove the switch. All signals on POTS lines were sounds (thus the tones for pressing buttons or the numbers of clicks from dial phones).
There’s a (very old, by now) Henry Morgan sketch I remember about an expensive long distance call: after an interminable series of dings and bongs, the operator admits that she lost count and has to return the coins so he can try again.
(Once they straighten out the funds, there’s a series of about six different operators needed to get the call across the country; all talking to, over, and past each other. “Hello, Hyannis, this is Truro. I have a call for Los Angeles! Westwood 5689.“ — “Hello Wellfleet, this is Hyannis. I have a call for Los Angeles. Westwood 5689.” — “Hello Boston, this is Wellfleet. I have a call for Los Angeles, Westwood 5698.” — ”No, Westwood 5689.” — “Who was that, Wellfleet?” — “That was Truro, Boston.” — “No, I’m Truro. That was Hyannis.” Meanwhile, the caller: “Are we still in Massachusetts??”)
No, there were tones that the phone sent for each coin value. IIRC they were fairly brief and somewhat "blippy", unlike the flat DTMF tones. Back in the day you could build a Red Box (https://en.wikipedia.org/wiki/Red_box_(phreaking)) that would simulate these tones.
Automatic switching came decades before automated billing. See "Operator Assisted Toll Dialing" (1949).[1] The operators are tone-dialing calls, but the billing system is entirely paper based.
Billing automation came in the 1960s, but didn't involve computers yet. Special purpose hardware, paper tape, and punched cards were involved.
