murray@jumbo.dec.com (Hal Murray) (12/05/89)
Suppose you have some remote instrumentation, like a seismometer, or weather station, and you are transmitting that data back to a central collection point via a phone line. Can you scrounge any power from the phone line? I'm not thinking of many watts, just enough to run a few slow CMOS chips. Is there an obvious book I should know about that answers things like this?
psfales@cbnewsc.ATT.COM (Peter Fales) (12/06/89)
In article <14230@jumbo.dec.com>, murray@jumbo.dec.com (Hal Murray) writes: > Can you scrounge any power from the phone line? I'm not thinking of > many watts, just enough to run a few slow CMOS chips. > > Is there an obvious book I should know about that answers things like > this? Check "Understanding Telephone Electronics" from your nearest Radio Shack. It is an interesting book, very high level in some areas, very detailed in others, but as I recall it has some good discussions of your question. -- Peter Fales AT&T, Room 5B-420 N9IYJ 2000 N. Naperville Rd. UUCP: ...att!ihlpb!psfales Naperville, IL 60566 Domain: psfales@ihlpb.att.com work: (708) 979-8031
larry@kitty.UUCP (Larry Lippman) (12/07/89)
In article <14230@jumbo.dec.com>, murray@jumbo.dec.com (Hal Murray) writes: > Suppose you have some remote instrumentation, like a seismometer, > or weather station, and you are transmitting that data back to a > central collection point via a phone line. > > Can you scrounge any power from the phone line? I'm not thinking of > many watts, just enough to run a few slow CMOS chips. The answer is yes. Since you did not specifify where you are referring to a dial subscriber line or to a leased line, I will briefly cover both issues. If you are referring to a conventional subscriber telephone line at the remote site which you dial using the public switched telephone network, then you can obtain between 1 and 2 mA from the line when it is on-hook. Some simple Ohm's law says this is about 100 mW. Using a carefully designed DC-DC converter, one can convert this voltage to say, 7.0 volts to trickle-charge a battery. The battery can then store this energy to provide a much higher current to power a modem, microprocessor and related circuitry during the intermittent intervals when the telephone line is automatically answered and data is exchanged. The precedent for the above circuit design using a DC-DC converter is in telephone subscriber line carrier circuits, such as the Continental AML. If you are referring to a leased line, then you can superimpose a certain amount of DC power provided that you use dual-winding transformers or inductors to provide DC/signal separation. In this day and age, telephone companies will generally not provide leased line circuits having DC continuity beyond the serving central office. If you are able to obtain a leased line circuit, you are generally limited to 100 mA of current and no more than +/- 135 volts from either conductor to ground (i.e., 270 volts conductor-to-conductor). This type of circuit will give you LOTS of power - even with a long loop resistance of say, 3,000 ohms. > Is there an obvious book I should know about that answers things like > this? Not that I am aware of. I'm afraid that experience is the only teacher here. <> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. <> UUCP {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry <> TEL 716/688-1231 | 716/773-1700 {hplabs|utzoo|uunet}!/ \uniquex!larry <> FAX 716/741-9635 | 716/773-2488 "Have you hugged your cat today?"
sorka@ucscb.UCSC.EDU (Alan Waterman) (12/07/89)
One thing you forgot to mention is that you get a lot more power out of the line when it is ringing. I got the worst shock of my life when I held the red and green wires(one in each hand) while the line sent a ring signal. It almost knocked me out. It mush have been high voltage and low amps because it went through me real easy.
ems@Apple.COM (Mike Smith) (12/07/89)
In article <6300@lindy.Stanford.EDU> sorka@ucscb.UCSC.EDU (Alan Waterman) writes: > >One thing you forgot to mention is that you get a lot more power out >of the line when it is ringing. I got the worst shock of my life when >I held the red and green wires(one in each hand) while the line sent >a ring signal. It almost knocked me out. It mush have been high voltage >and low amps because it went through me real easy. Vague memory from too many decades ago... I think the ring signal is pulsed DC at about 50 V and the current is 'all the inductor can carry'... Most lethal is a LOW current modest voltage signal that ocilates at a particularly sensitive frequency to the heart muscle. I've forgotten the frequency and if it was AC or pulsed DC. -- E. Michael Smith ems@apple.COM 'Whatever you can do, or dream you can, begin it. Boldness has genius, power and magic in it.' - Goethe I am not responsible nor is anyone else. Everything is disclaimed.
brian@ucsd.Edu (Brian Kantor) (12/07/89)
Typical phone lines in the USA have around 48 to 52 volts DC on them when the phone is on-hook (hung up, eh?). That drops to "talking battery" of about 4 to 10 volts when the phone is off-hook. During ringing, about 90 to 120 volts of 20 Hz alternating current is applied to the line. To power something off the phone line, you have to determine whether the device needs constant power or only needs power when the phone is off-hook. You see, if you draw any significant amount of current (a few milliamps or so), the central office equipment senses that as a phone being off-hook. Devices that require constant power generally derive it from the phone line by using extremely low-power circuitry (almost always CMOS stuff), and feeding the incoming line through some high-value resistors, rectifiers, regulators, and filtering components to keep the ringing voltage and lightning surges from frying the device. Devices that only need power when the phone is to be off-hook can grab more power, but they need to do it through a coupling circuit to allow the audio interface to the phone line to survive. The same sort of protective devices and isolators are needed. You can buy complete modules to interface to the phone line, but I've not seen any marketed which also provide device power. Probably you could twist one into doing that, though. A few minutes spent with a voltmeter and a telephone line is a rewarding experience, and educational too. A scope will let you see some of the timing and frequencies used as well. Try not to zap your ass on the ringing voltage. - Brian
richm@amc-gw.amc.com (Rich Moran) (12/08/89)
} } If you are referring to a conventional subscriber telephone line } at the remote site which you dial using the public switched telephone } network, then you can obtain between 1 and 2 mA from the line when it } is on-hook. Some simple Ohm's law says this is about 100 mW. Using a } carefully designed DC-DC converter, one can convert this voltage to say, } 7.0 volts to trickle-charge a battery. The battery can then store this } energy to provide a much higher current to power a modem, microprocessor } and related circuitry during the intermittent intervals when the telephone } line is automatically answered and data is exchanged. } This is fine if you aren't worried about FCC regs. Part 68 specifies a minimum 5M-ohm input impedance for DC on tip-ring at voltages below 100V. Even with a 56V battery at the CO, that's less than a milliwatt. rich -- ============================================= Rich Moran richm@amc.com =============================================
johnd@sco.COM (John Dubois) (12/08/89)
In article <3542@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes: > If you are referring to a conventional subscriber telephone line >at the remote site which you dial using the public switched telephone >network, then you can obtain between 1 and 2 mA from the line when it >is on-hook. ... ><> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. ><> UUCP {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry ><> TEL 716/688-1231 | 716/773-1700 {hplabs|utzoo|uunet}!/ \uniquex!larry ><> FAX 716/741-9635 | 716/773-2488 "Have you hugged your cat today?" How much current can you be sure you can draw before the phone company decides the line is bad? For example, if I was going to design a commercial product, how much could I use? Is there a specific minimum for on-hook line impedance? John DuBois johnd@sco.com spcecdt@ucscb.ucsc.edu
larry@kitty.UUCP (Larry Lippman) (12/08/89)
In article <5651@internal.Apple.COM>, ems@Apple.COM (Mike Smith) writes: > Vague memory from too many decades ago... I think the ring signal is > pulsed DC at about 50 V and the current is 'all the inductor can carry'... Ringing voltage is typically 20 Hz AC @ 105 volt RMS upon which is superimposed -50 volts DC (with respect to ground). The DC component is present to facilitate operation og a "ring-trip" relay or solid-state circuit. The telephone ringing supply is traditionally current limited to about 100 mA by a resistance lamp (i.e., ballast lamp). Resistance lamps are still used in much ESS apparatus due to their inherent simplicity and reliability. Newer ESS apparatus uses solid-state circuits for ringing supply current limiting. Ringing voltage is not applied to a telephone line through an inductor, so there is no current limiting by that means. <> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. <> UUCP {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry <> TEL 716/688-1231 | 716/773-1700 {hplabs|utzoo|uunet}!/ \uniquex!larry <> FAX 716/741-9635 | 716/773-2488 "Have you hugged your cat today?"
reynhout@wpi.wpi.edu (D Andrew Reynhout) (12/08/89)
In article <5651@internal.Apple.COM> ems@Apple.COM (Mike Smith) writes: >In article <6300@lindy.Stanford.EDU> sorka@ucscb.UCSC.EDU (Alan Waterman) writes: >> >>One thing you forgot to mention is that you get a lot more power out >>of the line when it is ringing. I got the worst shock of my life when >>I held the red and green wires(one in each hand) while the line sent >>a ring signal. It almost knocked me out. It mush have been high voltage >>and low amps because it went through me real easy. >Vague memory from too many decades ago... I think the ring signal is >pulsed DC at about 50 V and the current is 'all the inductor can carry'... Correct me if I'm wrong, but my memory from altogether too recently is that the ring signal is a 90+VAC signal, at 17hz. Current, I have no information on. The shock was never particularly bothersome to me, and I've been hit more times than I can recall. Of course, electric shock affects different people differently. As an aside, a local paper once printed an article originally appearing in the NYTimes, explaining how to wire your own extension telephones. They didn't bother to mention the ringing signal. I wrote in, noting that the application of such a jolt to a person wearing a pacemaker, for example, could be more than irritating...possibly deadly. I suggested taking one extension off-hook (48VDC) before attempting anything. There were a number of other glaring errors in the article, like the completely incorrect explanation of the two-four wire possibilities. They printed it, and forwarded it to the NYT, who is rumoured to have also printed it... my journalistic debut.. :-) Andrew -- Andrew Reynhout (Internet: reynhout@wpi.wpi.edu) "Maybe if we pretend this never happened, they'll all just...go away." - Laurie Anderson
jeffp@phred.UUCP (Jeff Parke) (12/09/89)
In article <5651@internal.Apple.COM> ems@Apple.COM (Mike Smith) writes: >... >Most lethal is a LOW current modest voltage signal that ocilates at >a particularly sensitive frequency to the heart muscle. I've forgotten >the frequency and if it was AC or pulsed DC. It turns out that the most lethal frequencies for putting out your heart are in the 40-100 Hz range. 60 Hz isn't a good one. Most deadly would be a small (threshold) current, since larger currents can actually get the heart going again - all depending on duration of contact and factors about the subject heart's own susceptibility. Early defibrillators (devices that deliver large current/voltage to start up a fibrillating heart) actually used 60 Hz, but were not especially effective. Modern, effective defibrillators are all DC; they deliver a 5 to 15 ms long single monophasic or biphasic waveform. -- ...dataio!pilchuck!phred!jeffp {Jeff Parke} Genie : JEFFP | DELPHI : JEFFPARKE | CIS : 71511,1512
larry@kitty.UUCP (Larry Lippman) (12/09/89)
In article <548@scorn.sco.COM>, johnd@sco.COM (John Dubois) writes: > > If you are referring to a conventional subscriber telephone line > >at the remote site which you dial using the public switched telephone > >network, then you can obtain between 1 and 2 mA from the line when it > >is on-hook. > > How much current can you be sure you can draw before the phone > company decides the line is bad? For example, if I was going to design a > commercial product, how much could I use? Is there a specific minimum for > on-hook line impedance? The maximum on-hook current draw depends upon the type of central office apparatus and the set point parameters for the Automatic Line Insulation Test (ALIT) apparatus. Most ALIT apparatus, however, checks for leakage from tip & ring to ground, and NOT leakage from tip to ring. Under most circumstances, 1 mA is a "reasonably safe" amount of leakage (corresponding to a 50,000 ohm leak) which should not interfere with dialing, ringing or supervisory signaling. ALIT apparatus is generally programmed for a leak threshold ranging between 50,000 and 150,000 ohms, depending upon the telephone company practice and condition of the cable plant. The primary purpose of ALIT is to give early warning of cable failures, and not subscriber line apparatus troubles. If you are going to design a commercial product, then you are in trouble because Part 68 will not allow ANY usable amount of leakage from tip to ring. In article <1052@amc-gw.amc.com> richm@amc-gw.amc.com (Rich Moran) writes: > This is fine if you aren't worried about FCC regs. Part 68 specifies a min. > 5M-ohm input impedance for DC on tip-ring at voltages below 100V. Even with > a 56V battery at the CO, that's less than a milliwatt. If the remote monitoring line is an off-premises extension (OPX) of a customer-owned PABX, then it is exempt from any FCC regulations. We do not know exactly what the original poster had in mind, so we'll give him the benefit of the doubt. :-) <> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. <> UUCP {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry <> TEL 716/688-1231 | 716/773-1700 {hplabs|utzoo|uunet}!/ \uniquex!larry <> FAX 716/741-9635 | 716/773-2488 "Have you hugged your cat today?"
roy@phri.nyu.edu (Roy Smith) (12/10/89)
In article <3547@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes: > If you are going to design a commercial product, then you are in trouble > because Part 68 will not allow ANY usable amount of leakage from tip to ring. Is this a new rule? Universal Data Systems (now part of Motorola, I think) used to make a nice little 1200 bps modem called the 212LP, where the LP meant Line Powered. We had a couple of them. The power budget must have been really tight because this was a real bare-bones unit. No dialer or auto-answer. To call out, you had to dial the number yourself on a phone then flip the talk/data switch on the front of the modem. But it definately didn't have any power supply. I suppose it's possible that it drew power from the RS-232 connection, but my impression was that it was driven from the phone line. -- Roy Smith, Public Health Research Institute 455 First Avenue, New York, NY 10016 roy@alanine.phri.nyu.edu -OR- {att,philabs,cmcl2,rutgers,hombre}!phri!roy "My karma ran over my dogma"
wtm@neoucom.UUCP (Bill Mayhew) (12/10/89)
The on-hook allowable leakage for a telephone instument is in the low microamp region for obvious reasons. I know that in the past, and maybe the present, that line sharing in rural areas that gives private service to two parties on the same line used leakage current. One party got regular baseband service. The second party operated on a carrier system AM modulated somewhere around 50KHz - 100KHz. The carrier system interface at the subscriber's drop contained a nicad battery to supply ringing power, as baseband ringing is impossible (without a gas tube frequency-selective ringer in this case). The nicad also supplied talk power for the local loop. When the set was on-hook a small leakage current was used to recharge the nicad pack. A couple years ago, I read about a neat system developed in Italy. They used a fiberoptic connection from the CO to the subscriber residence. The subscriber's set was recharged from the laser power received from the CO. Obviously, this system wasn't real effective for subscribers more than a few Km distant from the CO. One method of getting free power from the phone that I've yet to see exploited is to save the power from the ring signal. I've seen a number of instruments that draw 0.25B or more power when ringing. It would be a farily simple matter to attach an isolation transformer to the line for squirreling the ring power into a nicad. As long as the remote seismograph were called on a regular basis, the ringing could keep the battery fresh, if the device permitted the line to ring a number of times before going off hook. I have an old UDS manually operated modem that entirely phone line powered. It contains 60 some 4000 series CMOS ICs, and even generates RS-232C compliant output voltage for the computer interface by using it's off-hook loading to power itself. Pretty neat piece of work. Too bad that it couldn't go off hook or dial automatically. All of the modem's voltages were derived from clever resistor divider networks; only the -6 volts for the RS232 output was derived from a switched capacitor. At least, that's my recollection. Bill
larry@kitty.UUCP (Larry Lippman) (12/11/89)
In article <1989Dec9.220311.15928@phri.nyu.edu>, roy@phri.nyu.edu (Roy Smith) writes: >> If you are going to design a commercial product, then you are in trouble >> because Part 68 will not allow ANY usable amount of leakage from tip to ring > > Is this a new rule? Universal Data Systems (now part of Motorola, > I think) used to make a nice little 1200 bps modem called the 212LP, where Perhaps I did not make myself clear. I was referring to "leakage", which means the amount of current flow in an ON-hook (i.e., telephone line idle) state. In an OFF-hook state, one can draw as much current as the subscriber loop will allow, typically between 20 and 100 mA. It appears that the orignal poster was referring to designing a telemetering device for data acquisition which would require power on a *continuous* basis, and therefore would draw power from the telephone line while it was ON-hook. <> Larry Lippman @ Recognition Research Corp. - Uniquex Corp. - Viatran Corp. <> UUCP {allegra|boulder|decvax|rutgers|watmath}!sunybcs!kitty!larry <> TEL 716/688-1231 | 716/773-1700 {hplabs|utzoo|uunet}!/ \uniquex!larry <> FAX 716/741-9635 | 716/773-2488 "Have you hugged your cat today?"
sheriffp@spock (Peter Sheriff) (12/11/89)
Ringing in North America is a 90 Vac 20 Hz voltage. This is enough to make your eyes water (I speak from experience). Line voltage (when not ringing) is a nominal 48 Vdc. Be careful if you do use this as a source for any sort of equipment other than a 'phone. If you pull more than a few milliamps, the equipment in the phone office will respond as though the phone is off-hook and after a short time (if no dialing takes place) it will bring up an alarm. This is a low priority alarm so it may take them some time to get round to testing it but when they do, they will detect the current drain and could do something like a) cut off your service (actually, when the alarm activates in the central office, the voltage is cut off at that time). b)Send a repairman around to your home (unlikely) c)Just leave it as it is (you have no service anyway until you remove the cause of the problem). Pete
john@zygot.ati.com (John Higdon) (12/12/89)
In article <1989Dec9.220311.15928@phri.nyu.edu> roy@alanine.UUCP (Roy Smith) writes: >Universal Data Systems (now part of Motorola, >I think) used to make a nice little 1200 bps modem called the 212LP, where >the LP meant Line Powered. We had a couple of them. The power budget must >have been really tight because this was a real bare-bones unit. No dialer >or auto-answer. We used to use these on phone systems for remote programming and diagnostics. The 300 bps version (at least) would do auto-answer just fine. If the 1200 bps won't then one would have to assume that it takes more power to run the 212 protocol, no? -- John Higdon | P. O. Box 7648 | +1 408 723 1395 john@zygot.ati.com | San Jose, CA 95150 | M o o !
jones@hood (Clark Jones) (12/15/89)
In article <3550@kitty.UUCP> larry@kitty.UUCP (Larry Lippman) writes: > Perhaps I did not make myself clear. I was referring to "leakage", >which means the amount of current flow in an ON-hook (i.e., telephone line >idle) state. In an OFF-hook state, one can draw as much current as the >subscriber loop will allow, typically between 20 and 100 mA. > > It appears that the orignal poster was referring to designing a >telemetering device for data acquisition which would require power on a >*continuous* basis, and therefore would draw power from the telephone line >while it was ON-hook. Hmm... maybe he could have a capacitor array that could support his system for a couple of minutes, and then have his box dial its own number every minute or so, and then "listen" to the busy signal for a while, recharging the caps... or maybe if the "time" number in his area is free, it could ring that up once in a while. ;-) ;-) ;-) ;-) Disclaimer: The opinions expressed above are mine and not those of Schlumbrger because they are NOT covered by the patent agreement!
wb8foz@mthvax.cs.miami.edu (David Lesher) (12/15/89)
Discussion on difference in power available {on hook, offhook, ringing}: The box does not need to call anybody. Just have it go off hook regularly (say every 30 seconds) for only as long as it take the CO to complain, then go back on hook for a while, etc. I suspect the CO would start griping to the CO-humans soon, but maybe not. It WOULD improve your power budget! -- A host is a host & from coast to coast...wb8foz@mthvax.cs.miami.edu no one will talk to a host that's close..............(305) 255-RTFM Unless the host (that isn't close)......................pob 570-335 is busy, hung or dead....................................33257-0335
emmo@moncam.co.uk (Dave Emmerson) (12/20/89)
In article <14230@jumbo.dec.com>, murray@jumbo.dec.com (Hal Murray) writes: >[] > Can you scrounge any power from the phone line? I'm not thinking of > many watts, just enough to run a few slow CMOS chips. > Theoretically, yes, a little. Legally NO, apart from anything else, it IS theft! Dave E.
Nagle@cup.portal.com (John - Nagle) (12/22/89)
You can indeed get a little power from the phone line, and there are standards on how much you can draw. When off-hook, you can draw over 100 MA, in fact. All sorts of telephone-related devices run off phone line power nowadays. Bear in mind that all sorts of interesting surges and spikes appear on phone lines. Ring is nominally 88VAC, 20Hz. Much larger spikes can occur during electrical storms, and are considered part of the normal operating environment. On the other hand, nothing particularly bad will happen if you short the phone line for brief periods. Equipment that attaches to the phone line is supposed to be FCC registered. Automatic testing from the central office will detect as faults lines which present unreasonable loads. The telco will probably think that they have a cable fault, and if they spend time and effort chasing one down and discover that your equipment is causing a false fault indication, you may be billed for the repair costs. This is especially true for something that presents a DC load during on-hook periods insufficient to take the circuit off-hook. That, to the central office, looks like water in a cable, and tends to be investigated. John Nagle
irwin@m.cs.uiuc.edu (12/28/89)
/* Written 10:21 pm Dec 4, 1989 by murray@jumbo.dec.com in m.cs.uiuc.edu:sci.electronics */ /* ---------- "power from phone line" ---------- */ >Suppose you have some remote instrumentation, like a seismometer, >or weather station, and you are transmitting that data back to a >central collection point via a phone line. >Can you scrounge any power from the phone line? I'm not thinking of >many watts, just enough to run a few slow CMOS chips. >Is there an obvious book I should know about that answers things like >this? /* End of text from m.cs.uiuc.edu:sci.electronics */ A few years a go, the University of Illinois was doing some research on tracking wild animals, in Northern U.S. and in Canada. They would locate them by aircraft, fly low over them, shoot them with a stun gun dart, then land in a clearing, hike back and tab the animal with an ear tag. They also mounted a collar with a radio transmitter on it. They had phone pairs ran up to the top of high ground, spaced some 50 miles apart. Connected to the phone pair was an RCA 1802 CMOS microprocessor, with ROM, RAM and a few CMOS radio receivers. They put the transmitters on various frequencies, based on the type of animal. A Bear was on one frequency, Elk on another, Moose on another, Deer on another, etc. During idle time, the RAM and radio receivers were powered down, just a minimum amount of circutry was active. This would allow the phone pair to recharge the nicads. When they called the phone pair, it would activate the 1802, with the small amount of ROM and stack memory, they would tell the 1802 to output to an I/O port, which would enable the main RAM and they would then download software to it. They could then with the software in RAM, activate various I/O port configurations, which would enable a given receiver. If the receiver was receiving a signal, they would know that Mr. Bear was within radio distance, if he wasn't, they would disable that receiver and go on to the next one. With a grid of phone/microprocessor stations, they could track the animals. By finding which stations were receiving a given frequency, they could plot a vector to figure quite closely where they were. Though I was not involved with this, I know the person who was, that built the hardware and wrote the software for the project. He even went into the "wilds" on a few occasions, to help with the stations being set up and bagging the animals to tag/collar them. He still lives locally and I could put anyone in contact with him that might desire to talk with him via phone. Me, I thought it was a very neat project. Al Irwin Univ of Illinois Dept of Comp Sci irwin@m.cs.uiuc.edu