POTS Phone Line Power Capability

That is not correct on both points. The current doesn't flow on the surfac e, it flows on the area of the conductor furthest from the center. This is the entire surface only on a round shape, tube or rod. So by flattening y our conductor you are shoving the current into the corners or at best into the short edges of a flat, thin conductor wasting the vast majority of it.

Check with any of the ham radio groups to see how they make small loop ante nnas. They know to use round tubing so the current is spread evenly around the circumference without wasting the center bulk. Copper is not cheap.

There are antennas made of copper strap, but that is about keeping the cost low, not getting the most performance.

--

  Rick C. 

  ----- Get 1,000 miles of free Supercharging 
  ----- Tesla referral code - https://ts.la/richard11209

e and roads and electric lines were down for over a week. Anyone stupid en ough to try to deal with that combination themselves deserves to go to jail .

roadway the tree rolled, freed a branch and got hit in the head. He will never be the same.

ning to deal with it.

a friend who uses it to cut wood for heat. I think that's a bit silly mys elf, but it is what his family has always done.

If we did that, we would have been cut of from food and fuel for over tw o months. Even then, the FEMA contractor simply dropped the bucket on a fro nt end loader to drag the cut up branches to the other end of the street, w hile destroying the blacktop. It ended up costing the homeowners around $9,

000 each to repair the private roads.

Hi !

About 10 years ago we designed a specific autonomous emergency audio alarm transmitter with -48V on-hook PSTN line voltage (we've used the good old TAS-151 phone line emulator) to qualify the design.

On on-hook condition, 150mW (say 3mA@48V) is nearly admissible. We even experimented this on Silvertel modules (Ag1170/Ag2120).

I remember EU telecom standard TBR21 ... now supersed by ETSI (ETSI ES

203 021) standards or so.

Best regards, H

sources, but I was wondering if it might be possible these days to power a phone/answering machine entirely off the phone line.

Are you saying this was permitted by the standard? 3 mA is a lot more than 10 uA... 300 times more. But it won't be 48 volts anymore. They have res istance at the central office to limit the current and the line has resista nce. I would estimate the voltage to be in the 35 volt ball park depending on line resistance.

But even at 10 uA on hook, I think a phone that also uses off hook and ring ing power is very practical. If the phone doesn't ring and you don't make many calls, then the power requirements will be very low. Off hook conditi on gives plenty of power for calls. Ringing power is the cherry on top!

--

  Rick C. 

  ----+ Get 1,000 miles of free Supercharging 
  ----+ Tesla referral code - https://ts.la/richard11209

48/0.003=16KOhm very far from off-hook loop current with 600Ohm min. I guess nothing is really specified in EU standards for that point ... may be, but it works. Give it a try and see ...

About 15+ years ago I designed a capacitive charge pump down converter exactly for extracting on-hook power. Basically charge a bunch of caps in series and then switch them to be parallel - a bit like a marx generator backwards. If 100% efficient it would have converted 10uA at

50V to 100uA at 5V but efficiency never approached 50% and it was ridiculously complicated so it was easier to make do with 10uA.

piglet

sources, but I was wondering if it might be possible these days to power a phone/answering machine entirely off the phone line. So many require AC po wer and fail to operate, might it be possible to vampire enough power from the phone line so a cordless phone/answering machine might work from the ph one line alone?

at power level draw is allowed on a phone line. I did find a Remote Subscr iber Terminal POTS equipment spec from Tellabs that specified it would dete ct on hook at 9 kohms. It also specifies a 25 mA loop current at 600 ohms which I take to mean in combination with the internal current limiting resi stance in line with the 48 volt PSU. Doing the math I figure the in line r esistance is 1,320 ohms which gives a 10 kohm load up to 4 mA at around 42 volts and 180 mW.

not in real time operation.

I was thinking the on hook power would charge a cap to the full 48 V max at 10 uA max. When power is needed this cap could then be run through an eff icient buck. The idle current would need to be very low, but the ground po wer used when active would be relative to the end use, not the charging spe ed, so more latitude.

Switched caps are good, but as you say with such a large number of switches a bit busy. Done inside a chip it would have more potential for optimizat ion. I remember a TI part from 20 years ago that was a dropping regulator using switched cap design. It was intelligent in that it would provide the output voltage you needed while using the cap ratio that was as close as p ossible to the right voltage. The remainder of the voltage drop was achiev ed by using the switches as LDOs in essence. The efficiency was better tha n straight LDOs, no inductor and much lower noise. I believe it was a Euro pean design, German maybe? Shows what can be done with the switches if you bring it all on chip.

--

  Rick C. 

  ---+- Get 1,000 miles of free Supercharging 
  ---+- Tesla referral code - https://ts.la/richard11209

r sources, but I was wondering if it might be possible these days to power a phone/answering machine entirely off the phone line.

n

than 10 uA... 300 times more. But it won't be 48 volts anymore. They have resistance at the central office to limit the current and the line has res istance. I would estimate the voltage to be in the 35 volt ball park depen ding on line resistance.

ringing power is very practical. If the phone doesn't ring and you don't m ake many calls, then the power requirements will be very low. Off hook con dition gives plenty of power for calls. Ringing power is the cherry on top !

I'm not certain what you are suggesting. Are you saying if you draw 3 mA f rom a phone line it remains in the on hook state indefinitely and no one kn ocks at your door about the excess current? I thought someone had an accou nt where he did get a visit from the phone company repair guy?

--

  Rick C. 

  ---++ Get 1,000 miles of free Supercharging 
  ---++ Tesla referral code - https://ts.la/richard11209

Yes, thats's close but for PV solar they find that Vmp voltage of max power and then regulate to that voltage until they do another sweep. But I would not imagine that Phone line would change after the first measurement.

I thought you wanted to do tracking of both off-hook and on hook. Off hook would be much easier and more useful.

If you or someoe here had a real landline, you could measure and make an MPPT curve of power vs. voltage which should just depend on the CO to home loop distance.

Don't you watch General Hospital ?

e:

e

at

ng clear is that the maximum power drawn from the phone line when off hook requires an MPPT circuit. But that's not terribly important. I did some s imulations using a circuit that emulated a fixed resistor of about 7 kohms. It would charge up 1 mF to 5 volts in about two ring times.

off hook operating voltage. So clearly that was a bad design.

give you a phone line via a VOIP box which plugs into your home? That may be. But they can't give you that if you aren't on a fiber. Where my one h ouse is they would need to run fiber for some 12 miles or more and they ain 't gonna do that. My other house could never use a 56 kbps modem because i t's on a pair gain amplifier. Know what that is? When you figure it out y ou may understand that the phone company doesn't update old equipment if th ey don't have to.

ikipedia they don't call it an amplifier. That was the term the phone guy used at the time. Seems it was installed in the 70s when people were getti ng second lines and they didn't have the copper for it. I was told it was a digital mux using three phone lines to provide some dozen or so connectio ns. I would guess that's one pair each for data in each direction and one for clocking and sync or just clocking and the sync was in the data.

so this is what we have until something drastic happens like hover boards (not the ebay kind, the Back to the Future kind).

be

hough.

ook and/or ringing circuit to account for the line resistance. Actually, I 'm not certain either one is needed. The ultimate current is limited by sp ec, so the MPPT would have limits on the operating envelope.

to

y

rivial exercise. The "time" you are talking about is on the order of singl e digit milliseconds or faster. As I've pointed out, the issue is the limi ts imposed by the regulations. I don't know the details, but drawing curre nt in excess of the 10 uA allowed but not enough to register as off hook wi ll likely be flagged as a "fault" resulting in notices from your provider. So MPPT is probably without value for off hook operation. Ringing again h as limits on the current you can draw, so I'm willing to bet MPPT would vio late regulations.

t gives maximum power. Are you suggesting someone is measuring imaginary p ower instead of real power?

I don't know why it would be done with sweeps. Dithering will keep you ver y close to MPPT at all times and track changes as they happen. Using a swi tcher circuit makes it very easy to add the dither to the switching timing.

On hook and off hook are using the same power source. On hook MPPT is not needed. You are limited by regulations. Off hook is likely also limited b y requiring a minimum current to pull in the off hook detection which also keeps you away from the MP point (half the battery voltage.

It is the ringing circuit where MPPT might be useful, but I'm not even sure that will work. There is a REN limit. I designed a circuit in simulation to emulate a resistive load over the ring voltage. I expect this will be the MP point like the other cases because of regulatory limits.

If you really want to use MPPT, it is just easier to measure current and vo ltage and dither your set point to see which of the two is better. Then ad just your set point and dither around that. Why try to calibrate anything to conditions that won't be repeated at another installation?

But none of this is useful. My previous thoughts did not take into account the reality of the regulations.

alue that can't be used without mucking the phone state. I'm not talking a bout uA. The off hook state is over 10 mA and well less than 24 volts. On ce the voltage drops below around 24 volts you have passed the MPPT. Regar dless of what the phone company is checking for, it is not allowed to draw that sort of current unless you are off hook. You can't stay off hook fo r long unless you are making a call. That is checked and the phone line wi ll go into a state where it ignores your on hook condition for some time.

Why do you keep going on about using MPPT? It's not practical to use in vi ew of the regulations it seems. Also, it's not really needed.

It was an interesting thought experiment. I came up with a rather simple c ircuit to emulate a resistive load while charging a capacitor. It's very m uch like the current regulator I did last month but the reference point is a voltage tap off the incoming voltage. A base-emitter junction of a PNP c ompares the current sense resistor to the voltage divider with a diode in t he current sense path to balance out the BE voltage. Works rather nicely. The diode is also the LED for the ring indicator optoisolator. The circui t charges 10 mF to over 5 volts in four rings.

--

  Rick C. 

  --+-- Get 1,000 miles of free Supercharging 
  --+-- Tesla referral code - https://ts.la/richard11209

I think it was you that brought up MPPT. I am agreeing that it might be a good idea.

When the line rings, I would just full wave rectify that AC voltage and store it in a capacitor maybe. The frequency is around 20 Hz or so.

One sweep to find the Vmp and then just use that voltage from then on as it should not change and there shouldn't be any reason to dither after that. But you have to find the one voltage that gives you the Vmp max power point voltage first. Then it should just sstay there.

rote:

the

or

s e

that

thing clear is that the maximum power drawn from the phone line when off ho ok requires an MPPT circuit. But that's not terribly important. I did som e simulations using a circuit that emulated a fixed resistor of about 7 koh ms. It would charge up 1 mF to 5 volts in about two ring times.

of off hook operating voltage. So clearly that was a bad design.

e

ey give you a phone line via a VOIP box which plugs into your home? That m ay be. But they can't give you that if you aren't on a fiber. Where my on e house is they would need to run fiber for some 12 miles or more and they ain't gonna do that. My other house could never use a 56 kbps modem becaus e it's on a pair gain amplifier. Know what that is? When you figure it ou t you may understand that the phone company doesn't update old equipment if they don't have to.

n Wikipedia they don't call it an amplifier. That was the term the phone g uy used at the time. Seems it was installed in the 70s when people were ge tting second lines and they didn't have the copper for it. I was told it w as a digital mux using three phone lines to provide some dozen or so connec tions. I would guess that's one pair each for data in each direction and o ne for clocking and sync or just clocking and the sync was in the data.

wn, so this is what we have until something drastic happens like hover boar ds (not the ebay kind, the Back to the Future kind).

to be

s though.

f hook and/or ringing circuit to account for the line resistance. Actually , I'm not certain either one is needed. The ultimate current is limited by spec, so the MPPT would have limits on the operating envelope.

ant to

very

a trivial exercise. The "time" you are talking about is on the order of si ngle digit milliseconds or faster. As I've pointed out, the issue is the l imits imposed by the regulations. I don't know the details, but drawing cu rrent in excess of the 10 uA allowed but not enough to register as off hook will likely be flagged as a "fault" resulting in notices from your provide r. So MPPT is probably without value for off hook operation. Ringing agai n has limits on the current you can draw, so I'm willing to bet MPPT would violate regulations.

aw

at.

that gives maximum power. Are you suggesting someone is measuring imaginar y power instead of real power?

very close to MPPT at all times and track changes as they happen. Using a switcher circuit makes it very easy to add the dither to the switching timi ng.

ot needed. You are limited by regulations. Off hook is likely also limite d by requiring a minimum current to pull in the off hook detection which al so keeps you away from the MP point (half the battery voltage.

ure that will work. There is a REN limit. I designed a circuit in simulat ion to emulate a resistive load over the ring voltage. I expect this will be the MP point like the other cases because of regulatory limits.

voltage and dither your set point to see which of the two is better. Then adjust your set point and dither around that. Why try to calibrate anythi ng to conditions that won't be repeated at another installation?

unt the reality of the regulations.

a value that can't be used without mucking the phone state. I'm not talkin g about uA. The off hook state is over 10 mA and well less than 24 volts. Once the voltage drops below around 24 volts you have passed the MPPT. Re gardless of what the phone company is checking for, it is not allowed to dr aw that sort of current unless you are off hook. You can't stay off hook for long unless you are making a call. That is checked and the phone line will go into a state where it ignores your on hook condition for some time .

view of the regulations it seems. Also, it's not really needed.

e circuit to emulate a resistive load while charging a capacitor. It's ver y much like the current regulator I did last month but the reference point is a voltage tap off the incoming voltage. A base-emitter junction of a PN P compares the current sense resistor to the voltage divider with a diode i n the current sense path to balance out the BE voltage. Works rather nicel y. The diode is also the LED for the ring indicator optoisolator. The cir cuit charges 10 mF to over 5 volts in four rings.

I was only trying to set a reasonable current but someone mentioned MPPT an d without giving it much thought I went with the idea. But once I looked a t the issues there is no opportunity to use it I'm pretty sure. Even with the ringing signal you are limited to what power level you can draw. The b est you can do is to achieve some set REN that you know will work with what ever other loads are on the line.

Yes, that is what my circuit does. But the current has to be limited and t he spec is a resistive load. So I use two transistors to set the current a ccording to the voltage like a resistor would do.

--

  Rick C. 

  --+-+ Get 1,000 miles of free Supercharging 
  --+-+ Tesla referral code - https://ts.la/richard11209

rote:

the

or

s e

that

thing clear is that the maximum power drawn from the phone line when off ho ok requires an MPPT circuit. But that's not terribly important. I did som e simulations using a circuit that emulated a fixed resistor of about 7 koh ms. It would charge up 1 mF to 5 volts in about two ring times.

of off hook operating voltage. So clearly that was a bad design.

e

ey give you a phone line via a VOIP box which plugs into your home? That m ay be. But they can't give you that if you aren't on a fiber. Where my on e house is they would need to run fiber for some 12 miles or more and they ain't gonna do that. My other house could never use a 56 kbps modem becaus e it's on a pair gain amplifier. Know what that is? When you figure it ou t you may understand that the phone company doesn't update old equipment if they don't have to.

n Wikipedia they don't call it an amplifier. That was the term the phone g uy used at the time. Seems it was installed in the 70s when people were ge tting second lines and they didn't have the copper for it. I was told it w as a digital mux using three phone lines to provide some dozen or so connec tions. I would guess that's one pair each for data in each direction and o ne for clocking and sync or just clocking and the sync was in the data.

wn, so this is what we have until something drastic happens like hover boar ds (not the ebay kind, the Back to the Future kind).

to be

s though.

f hook and/or ringing circuit to account for the line resistance. Actually , I'm not certain either one is needed. The ultimate current is limited by spec, so the MPPT would have limits on the operating envelope.

ant to

very

a trivial exercise. The "time" you are talking about is on the order of si ngle digit milliseconds or faster. As I've pointed out, the issue is the l imits imposed by the regulations. I don't know the details, but drawing cu rrent in excess of the 10 uA allowed but not enough to register as off hook will likely be flagged as a "fault" resulting in notices from your provide r. So MPPT is probably without value for off hook operation. Ringing agai n has limits on the current you can draw, so I'm willing to bet MPPT would violate regulations.

aw

at.

that gives maximum power. Are you suggesting someone is measuring imaginar y power instead of real power?

very close to MPPT at all times and track changes as they happen. Using a switcher circuit makes it very easy to add the dither to the switching timi ng.

That idea is more complicated than the dither circuit. No modes, just a sm all modulation on the feedback controlling the feedback set point. Very si mple. Why screw with modes and holding a set point value? How long before the set point holding cap drifts?

--

  Rick C. 

  --++- Get 1,000 miles of free Supercharging 
  --++- Tesla referral code - https://ts.la/richard11209

The point is that 3mA shall be the max current when Public Central Office Equipment delivers the ring signal (48VDC with a sine signal

90Vac 20/30Hz superimposed). That's what we did at the time ... successfully.

POTS is an old topic, here in Europe, Analog Central Office equipment shall is being obsolete and will be abandoned from 2023.

I think WLL (Wireless Local Loop) will certainly yet exists for years (e.g. Analog I/F between ADSL Boxes to DECT base stations), Please consider the Analog Slic modules (FXO, FXS), Silvertel provides yet good modules for that that purpose.

H

The point is that 3mA have had to be the max current when Public Central Office Equipment delivers the ring signal (48VDC with a sine signal

90Vac 20/30Hz superimposed). That's what we did at the time ... successfully. BTW I don't see how 100uA would be monitored by the public CO ... quite sure that leak current (diff or common mode) in POTS could be x10 or so in worst case.

POTS is an old topic, here in Europe (France), Analog Central Office equipment shall are now obsolete and will be abandoned from 2023.

I think WLL (Wireless Local Loop) will certainly yet exists for years (e.g. Analog I/F between ADSL Boxes to DECT base stations), Please consider the Analog Slic modules (FXO, FXS), Silvertel provides yet good modules for that that purpose.

H

PS. Please see pp.9 for Leakage current and elsewhere Impedance limit during Ring :