Help understanding the amarican POTS standard

[dusterthefirst]

Hello, I am fairly new here so apologies if there's anything wrong with my post, be it formatting or just understandable. With that out of the way let me just get straight to the point.

I am a very beginner electronics hobbyist so I currently don't understand much of the terminology and what have you. I understand the basics of DC and digital electronics but am not so well versed in AC and analog electronics. having said that, I have been trying to create a "POTS phone base station", or in other words, just emulate a POTS phone line that I can connect any old (touch tone/rotary) phone up to. Ive found 2 helpful resources that document the standard quite well, but I am still having trouble understanding it fully. Here are the 2 pages I have been looking at:

• https://www.electroboom.com/?p=1034
• http://www.rane.com/note150.html

And heres what ive gotten out of it so far

• To ring: 60V RMS 20HZ AC signal across the tip and ring (I have a working prototype for this)
• To detect if the phone is on hook: 48V DC across tip and ring. Drops to ~6-8 VDC

The main thing I do not understand is how the phone drops the voltage. Does the power supply have to be current limited somehow? Cause the way i see it shouldn't the lines always have 48V if that's what I'm supplying? I tried to test this by connecting an old phone up to my power supply at 48V DC but when i picked up the phone the current shot way up which I presume was from a short, so I disconnected it from power.

If anyone could help explain in more layman terms the standard that would be very welcomed. If not the whole standard someone could just explain this hook detection that would be splendid.

Attached below is also my ringer circuit if anyone has any suggestions to improve it.

Info on the image:

• On the left, the 4 ICs make up a SSR H-Bridge to create the AC.
• The power rail labelled HV is powered at 55V, the highest my DC/DC boost converter will do.
• Between the DIP ICs is a NPN BJT that inverts the opposite side of the H-bridge so that they will never be on simultaniously
• The two DIP ICs to the right of the BJT are isolation solid state relays, they disconnect the H-bridge when its not ringing
• Last but not least, at the right is a 555 timer set to run at 19.998HZ with a duty cycle of 49% or something around that
• The button the the right just enables the isolation SSRs and the 555 timer

[ve7xen]

The exchange generally has an output resistance of 1000-2000 ohms. In effect, this creates a current limit of ~20mA at -48V. The handset signals off-hook by increasing its current draw, reducing the voltage at the end of the line, and increasing the voltage at the exchange across the output resistor. On the handset side, in a modern device, this is typically done with a zener clamp.

This type of interface is called an 'FXS' and is implemented by a 'ringing SLIC'. You can get modules and ICs that perform all of these tasks, and their datasheets might provide additional insight into the implementation.

There is also some good information here: http://www.ti.com/pdfs/bcg/building_residential_gateways.pdf

[AndyC_772]

I spent far more of my professional career designing telephone equipment than I care to admit.

It's a world of rather bizarre acronyms, which make it hard to find what you're looking for unless you already know them. An "FXS" port supplies power and can drive the phone. Its counterpart is an "FXO" port, which looks like a phone, ie. it plugs into a phone line.

"SLIC" is "Subscriber Line Interface Circuit", and there are various options depending on how long a cable you need to support and which countries' line interfaces you need to be compatible with. Fortunately a US line uses an impedance of 600 ohms, which most devices will support.

The one potential issue you've not yet mentioned is ring-trip detection. When the phone rings, its ringing circuit draws a modest current, but this can potentially increase greatly when the phone is picked up (this is called "looping the line"). Your line interface needs to detect the increase in current which occurs when the handset is lifted and must shut off the ringer very quickly, to avoid blasting the user with a very loud noise and potentially damaging something. There are generally two methods, called 'ac' and 'dc' ring-trip detection, and you may want to choose between them based on whether or not your interface maintains 48v dc even while it's ringing.

[dusterthefirst]

There is also some good information here: http://www.ti.com/pdfs/bcg/building_residential_gateways.pdf

Thank you so much for that document, thats what ive been looking for this whole time! thanks so much.

[dusterthefirst]

The one potential issue you've not yet mentioned is ring-trip detection. When the phone rings, its ringing circuit draws a modest current, but this can potentially increase greatly when the phone is picked up (this is called "looping the line"). Your line interface needs to detect the increase in current which occurs when the handset is lifted and must shut off the ringer very quickly, to avoid blasting the user with a very loud noise and potentially damaging something. There are generally two methods, called 'ac' and 'dc' ring-trip detection, and you may want to choose between them based on whether or not your interface maintains 48v dc even while it's ringing.

I had not thought of that before, but i definantly will look into that, I have a hall effect current sensing breakout board i could probbably put to good use with this

[rstofer]

Measure the polarity of the line.  Lift the receiver and note that the polarity has reversed.  Call somebody and, when they answer, note that the polarity reverses again back to the original state.  "Answer Supervision" is the term.  At least that's the way it worked when I was messing around with telephony back in the early '80s.  Haven't checked lately.

[AndyC_772]

IIRC that polarity reversal is how an answering machine knows when to stop recording.

[mindcrime]

I was going to build something similar once, inspired by my (mostly brief) foray into the world of phone phreaking back in the day. But I got distracted and never wound up doing much. So all I can really contribute to this discussion is to suggest that you might find this book useful:

https://www.amazon.com/Understanding-Telephone-Electronics-Joseph-Carr/dp/0750671750

Note that there is a copy available online, at classicrotaryphones.com, but I'm not sure if it's legit or not. Use your own discretion in deciding if you want to find/download it or not.