I did a quick search on 1670 schematics and came up with very little. Note however that several sources identify the USR101 chip as a ROM - that's clearly not the case, as a schematic shows no address or data bus, and you can see a line from the coupling transformer to an RC network and several pins on the device. It's almost certainly the analog front end and the modulator/demodulator.
If you could find a datasheet for that device you'd probably be able to figure out how the whole thing works, but I'm guessing those were never available to the public and have been lost to the mists of time.
The issue with the USB modems might actually be a bit different... more frequent use of single-pair digital business telephone systems in the '90s lead to problems with people connecting them to conventional modems. Some of these digital systems used unusually high battery voltages (sort of a PoE-esque method of providing meaningful current to the desk phones for their lights and displays), and there was a concern that they could damage the ADC on modems. So a lot of newer modems actually check the voltage on the line before they connect the audio electronics to it, as a safety precaution. There's an AT response that indicates that the modem is unhappy with the line voltage and I tenuously remember that at least some modems have an AT command to defeat this protection and make them connect anyway. Digital Line Detect is what the feature is usually called, but you might more generally call it a line voltage check.
However, recently a friend of mine tried to use the same trick with two USB modems from Trendnet and it failed. It seems they either don't have oomph enough or they rely on 9vdc off-hook carrier voltage to work.
(hmm, actually by that time it's likely that all the Codex tech had been killed, and your modem came from the Motorola UDS folks in Huntsville. Makes sense - Motorola Codex couldn't build a $500 modem to save their lives, never mind a $100 one, and a lot of the folks interested in building consumer modems left to found Zoom Telephonics)
> 9vdc off-hook carrier voltage to work
In the US it's nominally 48v, although it drops to a small fraction of that with any significant (or even fairly insignificant) load.
I guess enough signal makes it through T1 to let it work on dry lines. That sort of thing isn't unheard of [1]. I used to mess around with obsolete dial up modems as a kid and I could definitely get them to connect with just wires - sometimes. It didn't always work, and some modems would not work with other ones, etc.
The weird thing here is the 1670's second long pulses. Maybe it would do something else if connected to a real phone line? I'd surely try it myself but I only have a 1660 modem on hand.
0 - https://www.zimmers.net/anonftp/pub/cbm/schematics/misc/1670...
https://www.atlasobscura.com/articles/barbed-wire-telephone-...
Or RC “ringing”. That click stopping the sound makes me curious to see a plot of bytes per second sent and received. Is the modem resetting itself every time it clicks and introducing a small pause or data loss? Or is it completely divorced from the sound.
These modems don’t have fans built in right? If you isolate the speaker does it still make the noise?
There is no data being sent or received yet when it makes this sound; as soon as it dials and starts handshaking it goes away. You can hear that here: https://www.youtube.com/watch?v=S-ChICic2TA
EDIT: actually that's only partially true -- in the video above it's still making the noise when the dialling is happening. It disappears as soon as it detects a carrier from the other modem. This is weird because the click really sounds like a relay and the function of the relay should be to hang up the connection. Obviously, hanging up while in the middle of dialling is not a good idea.
I'd totally believe a modem from the 80's was lacking a sophisticated Control System for the idle state.
0 - https://www.zimmers.net/anonftp/pub/cbm/schematics/misc/1670...
