jgb@prism.gatech.EDU (James G. Baker) (01/21/91)
I've seen small 2400 baud modems that *only* have COM and phone connections.
How do these pull power from the telephone or computer? I need to make a
box with similar needs (a few LS parts) and of course they need a +5 power
supply. Is there a +5V line on RS-232C and is it standard with DB25 and DB9
connectors? Or is there a +12,etc that I could feed into a voltage regulator?
Thanks!
--
BAKER,JAMES G - Undergraduate Lab Instructor, School of Electrical Engineering
____ _ _ Georgia Institute of Technology, Atlanta Georgia, 30332
| | _ |_) uucp: ...!{decvax,hplabs,ncar,purdue,rutgers}!gatech!prism!jgb
(_|. |_). |_). Internet: jgb@prism.gatech.edumcovingt@athena.cs.uga.edu (Michael A. Covington) (01/21/91)
In article <19983@hydra.gatech.EDU> jgb@prism.gatech.EDU (James G. Baker) writes: >I've seen small 2400 baud modems that *only* have COM and phone connections. >How do these pull power from the telephone or computer? I need to make a >box with similar needs (a few LS parts) and of course they need a +5 power >supply. Is there a +5V line on RS-232C and is it standard with DB25 and DB9 >connectors? Or is there a +12,etc that I could feed into a voltage regulator? > No, there are no power lines. What you can do is use switching diodes to automatically give you a path to the most positive pin(s) and draw just a _little_ power from them. It has to be micropower (on the order of 1 mA).
squishy@casbah.acns.nwu.edu (Shishin Yamada) (01/21/91)
In article <19983@hydra.gatech.EDU> jgb@prism.gatech.EDU (James G. Baker) writes: >I've seen small 2400 baud modems that *only* have COM and phone connections. >How do these pull power from the telephone or computer? I need to make a >box with similar needs (a few LS parts) and of course they need a +5 power >supply. Is there a +5V line on RS-232C and is it standard with DB25 and DB9 >connectors? Or is there a +12,etc that I could feed into a voltage regulator? The phone line supplies a small amount of DC bias typically when off-hook. I know in certain areas it also supplies a small bias even when on-hook. However, these voltages are NOT meant for use, and should not be used for driving external circuits. On the other hand however, the RS-232 standard has connectors for +5V on its line. It should in serious cases have a +12V line. The modem most likely makes its connection here. Certain computers (notably non-ibms) and possibly portables do not support these lines because it in these machines their power supplies could overheat and breakdown. This is probably no longer a huge problem since IBM keeps building huge wattage power supplies and newer circuit boards make better use of less power. In the MacIntosh world (my niche), the RS-422 interface (a better and compatable system to RS-232) does not support power lines. This is because the Mac Plus was VERY prone to power overload with its small power supply. In the example of the Practical Peripherals Pocket Modem, the Mac version needs a 9 Volt battery to make it work. This might also be necessary with any portable computer (pocket modems chief end-user). Well, I hope that helps :) ---------------------------------------------------------------- Shishin Yamada Northwestern University Electrical Engineering Class of 1991 ----------------------------------------------------------------
henry@zoo.toronto.edu (Henry Spencer) (01/21/91)
In article <19983@hydra.gatech.EDU> jgb@prism.gatech.EDU (James G. Baker) writes: >I've seen small 2400 baud modems that *only* have COM and phone connections. >How do these pull power from the telephone or computer? ... If you're thinking of the really tiny portable ones, there is generally a 9V battery tucked in somewhere, I believe. It is possible to pull a small amount of power from the phone lines, but this has to be done very carefully. There have been line-powered modems, but my impression is that they historically have not been very satisfactory. The available power is quite small and somewhat variable. >...s there a +5V line on RS-232C and is it standard with DB25 and DB9 >connectors? Or is there a +12,etc ... No. There is no dependable source of any power voltage on an RS232 connector with 25 pins, much less one with 9. Some very small leased-line modems draw their power from the RS232 signal and modem-control lines, but this takes quite clever low-power design. The total available is measured in individual milliamps. -- If the Space Shuttle was the answer, | Henry Spencer at U of Toronto Zoology what was the question? | henry@zoo.toronto.edu utzoo!henry
mcovingt@athena.cs.uga.edu (Michael A. Covington) (01/21/91)
In article <2820@casbah.acns.nwu.edu> squishy@casbah.acns.nwu.edu (Shishin Yamada) writes: > >On the other hand however, the RS-232 standard has connectors for +5V on >its line. Er... not as far as I can determine. I've never seen one that did.
grayt@Software.Mitel.COM (Tom Gray) (01/22/91)
In article <1991Jan21.025151.6524@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes: :In article <19983@hydra.gatech.EDU> jgb@prism.gatech.EDU (James G. Baker) writes: :>I've seen small 2400 baud modems that *only* have COM and phone connections. :>How do these pull power from the telephone or computer? ... : :If you're thinking of the really tiny portable ones, there is generally :a 9V battery tucked in somewhere, I believe. : :It is possible to pull a small amount of power from the phone lines, but :this has to be done very carefully. There have been line-powered modems, :but my impression is that they historically have not been very satisfactory. :The available power is quite small and somewhat variable. : The telephone loop supplies power at a nominal voltage of 48V (max 56, min 42). The DC resistance of the loop will vary from 0 to 1800 ohms. During an active call (line off hook) you may draw up to 40mA from the loop at whatever voltage the loop is going to give you (remember the loop resistance). When the line is inactive, you may not draw more than 3 or 4mA from the loop Drawing more than this will falsely start a new call.
shirriff@sprite.berkeley.edu (Ken Shirriff) (01/22/91)
In some article, someone writes: >:It is possible to pull a small amount of power from the phone lines, but >:this has to be done very carefully. "The Art of Electronics" (Horowitz and Hill) describes in detail how much power can be extracted from phone lines, how to do it, and how to design low-power circuits in general. And, to kill two birds with one stone, here's my book review: I just got "The Art of Electronics" (2nd ed.) on the net's recommendation. It's crammed full of useful information and is interesting to read. It covers nearly everything (e.g. transistors, op amps, digital, computer interfacing, amplifiers, high frequency, low power, measurement, high accuracy) It's low on theory and concentrates on practical circuits, advice, and techniques. It moves too quickly to be an introductory book, but would be great for anyone else. I give it a thumbs up, even though it's expensive ($50). Ken Shirriff shirriff@sprite.Berkeley.EDU
ftpam1@acad3.alaska.edu (MUNTS PHILLIP A) (01/22/91)
In article <6128@healey>, grayt@Software.Mitel.COM (Tom Gray) writes... >In article <1991Jan21.025151.6524@zoo.toronto.edu> henry@zoo.toronto.edu (Henry Spencer) writes: >:In article <19983@hydra.gatech.EDU> jgb@prism.gatech.EDU (James G. Baker) writes: >:>I've seen small 2400 baud modems that *only* have COM and phone connections. >:>How do these pull power from the telephone or computer? ... >: >:If you're thinking of the really tiny portable ones, there is generally >:a 9V battery tucked in somewhere, I believe. >: >:It is possible to pull a small amount of power from the phone lines, but >:this has to be done very carefully. There have been line-powered modems, >:but my impression is that they historically have not been very satisfactory. >:The available power is quite small and somewhat variable. >: > >The telephone loop supplies power at a nominal voltage of 48V (max 56, min 42). >The DC resistance of the loop will vary from 0 to 1800 ohms. > >During an active call (line off hook) you may draw up to 40mA from the loop >at whatever voltage the loop is going to give you (remember the loop resistance). >When the line is inactive, you may not draw more than 3 or 4mA from the loop >Drawing more than this will falsely start a new call. FCC Part 68 specifies this for the United States. I believe the on-hook impedance must be greater than 10 megohms. I have designed products that incorporate "line sniffers" that detect loss of the -48V on-hook voltage. The strategy is to slowly charge a low leakage capacitor thru a 10 megohm resistor and discharge it rapidly thru a coupling transformer or optocoupler. The whole cycle takes a few seconds. Interestingly, this all falls apart when the phone line is shared; the off-hook voltage is not enough to detect reliably. So we added special loop current relays to detect the off-hook condition. Philip Munts N7AHL NRA Extremist, etc. University of Alaska, Fairbanks
squishy@casbah.acns.nwu.edu (Shishin Yamada) (01/22/91)
I might be wrong about that RS-232 and power supply. I remember working on the IBM way back when (and frying a homebrew periapheral interface adapter). At that time it was on a big IBm with a 25 pin RS-232 port. I do not think the 9 pin RS-232 port has the same thing. I could very well be wrong (and thinking of the IBM's parallel port since I worked with that too). It is hard because I remember how tough it was to come up with spec sheets for the RS-232 pinouts at the time. Anyways, now I am a dedicated Mac User with its RS-422 ports that are downward compatable with RS-232. In more enlightening (and accurate) news, I just picked up a Practical Peripherals promo sheet about their new modems. In particular the Practical Pocket Modem (2400PPM) says the following: "...Thanks to its remakable design, no-fail power comes directly from a RS232 port and the telephone line; you never have to find a wall outlet or change batteries.... Mac users will need a separate 9V battery adaptor included in the optional Macintosh package." I know for SURE that Mac doesn't support any power lines since the Mac Plus. The Mac Plus and earlier supported +5V on their 9 pin serial lines (very similar to RS-232 nine pin standard, and I believe the same pinout too). More recent macs drop this because of third party use that damaged many weak Mac Plus power supplies. Having thought about storing power on such a small modem poses problems, as a storage device has to be extremely small. Also, if taken from the phone line when "off-hook" (a DC voltage is assumed), you'd have to look out for the 80-100 Volts AC that occur during ringing. If it was drawn from the RS-232 data/hankshake lines, it would need a method of tapping, filtering, and stabilizing the power. It would seem to me that tapping from the DTR line would be the BEST bet (if there's no power line) since it could be turned on/off at will from the computer and is not used in modems without flow control. In this case, the 2400PPM seems to be a standard 2400 bps without data compression/error control, which means it probably doesn't require DTR handshaking. Please note, I am only guessing... Anyways, I do know of a more probable and better source of information. You could try "GO PPIFORUM" on Compuserve or even the "comp.dcom.modems" news group. --------------------------------------------------------------------- Shishin "Squish" Yamada Squishy@Casbah.Acns.Nwu.Edu Northwestern University Electrical Engineering Class of 1991 --------------------------------------------------------------------- PS: If anyone has a text of RS-232 standards (25/9 pin), I would appreciate an e-mail. For sake of conscience, I'd like to see again what was on those specs.
olson@bootsie.uucp (Eric K. Olson) (01/23/91)
The Farallon (sp?) MacRecorder is a little 8-bit 22 KHz digitizer that plugs into the Macintosh serial port and draws its power from the Modem Control lines (DSR/DTR)! I considered doing this, but I wanted the modem control lines for data. -Eric -- Eric K. Olson, Editor, Prepare() NOTE: olson@bootsie.uucp doesn't work Lexington Software Design Internet: olson@endor.harvard.edu 72A Lowell St., Lexington, MA 02173 Uucp: harvard!endor!olson (617) 863-9624 Bitnet: OLSON@HARVARD
adamd@rhi.hi.is (Adam David) (01/23/91)
In <2820@casbah.acns.nwu.edu> squishy@casbah.acns.nwu.edu (Shishin Yamada) writes: >In article <19983@hydra.gatech.EDU> jgb@prism.gatech.EDU (James G. Baker) writes: >>I've seen small 2400 baud modems that *only* have COM and phone connections. >>How do these pull power from the telephone or computer? I need to make a >>box with similar needs (a few LS parts) and of course they need a +5 power >>supply. Is there a +5V line on RS-232C and is it standard with DB25 and DB9 >>connectors? Or is there a +12,etc that I could feed into a voltage >regulator? CMOS parts and clever power circuitry are needed here to draw power from the RS-232C and/or the phone line. >The phone line supplies a small amount of DC bias typically when off-hook. >I know in certain areas it also supplies a small bias even when on-hook. >However, these voltages are NOT meant for use, and should not be used for >driving external circuits. However some products use the phone line for power. The phone company would not be happy with any large load on the wires, and indeed the equipment which attempted to load the wires too much would be unreliable or simply wouldn't work. There are probably recommendations for maximum permissible loading of the phone lines. >On the other hand however, the RS-232 standard has connectors for +5V on >its line. It should in serious cases have a +12V line. The modem most >likely makes its connection here. Certain computers (notably non-ibms) and >possibly portables do not support these lines because it in these machines >their power supplies could overheat and breakdown. This is probably no >longer a huge problem since IBM keeps building huge wattage power supplies >and newer circuit boards make better use of less power. Some "RS-232C" ports have power connections but the vast majority don't. On some newer systems which use RTS/CTS handshaking (as opposed to DTR/CTS), it may be possible to draw power from DTR depending on current drive capabilities of that signal. DTR is in these cases at a steady positive level, either all the time the computer is switched on or whenever the serial port is open. >needs a 9 Volt battery to make it work. This might also be necessary with >any portable computer (pocket modems chief end-user). I have used a portable modem and use portable computers. I don't intend to use the portable modem much on batteries because the modem fails instananeously as soon as the batteries hit a certain voltage. What I'd really like to use is a modem (2400 bps or more) that draws all its power from the phone lines (at least while online), using a high level at DTR to switch the modem on. Does anyone know about such creatures? -- Adam David. adamd@rhi.hi.is
raoul@eplunix.UUCP (Nico Garcia) (01/26/91)
>In article <19983@hydra.gatech.EDU> jgb@prism.gatech.EDU (James G. Baker) writes: >>supply. Is there a +5V line on RS-232C and is it standard with DB25 and DB9 >>connectors? Or is there a +12,etc that I could feed into a voltage >regulator? Ummm, I'm looking at a listing of the EIA signals from a Maxwell modem right here. It says pin 9 is +Testing Voltage, Pin 10 is -Testing Voltage. Now, if the *terminal* is wired appropriately, there could indeed be a voltage source there. I've got one of the old phone line modems, though, 1200 baud and almost indestructible. I kinda like it. Runs off the "off-hook" current, which is supposed to be about 10-15 volts across a 600 ohm telephone load. -- Nico Garcia Designs by Geniuses for use by Idiots eplunix!cirl!raoul@eddie.mit.edu
henry@zoo.toronto.edu (Henry Spencer) (01/27/91)
In article <1010@eplunix.UUCP> raoul@eplunix.UUCP (Nico Garcia) writes: >Ummm, I'm looking at a listing of the EIA signals from a Maxwell modem right >here. It says pin 9 is +Testing Voltage, Pin 10 is -Testing Voltage. Now, if >the *terminal* is wired appropriately, there could indeed be a voltage source >there... Those test voltages are usually run through something like a 1k resistor so you can't accidentally fry things by misconnecting them. This also severely limits their usefulness as power sources, assuming your device generates them at all. (Many [most?] professional-grade modems do, but I doubt that many terminals do. PC-market modems are anyone's guess.) As I said before, but not emphatically enough: there is *no* pin on a standard RS232 connector which is guaranteed to be able to supply a serious amount of power. Some devices do put power pins on RS232 connectors; this is useful if you've got it, but it's not dependably available on a random RS232 connector. The standard RS232 pins are all signal pins except for those two test voltages. -- If the Space Shuttle was the answer, | Henry Spencer at U of Toronto Zoology what was the question? | henry@zoo.toronto.edu utzoo!henry
tell@oscar.cs.unc.edu (Stephen Tell) (01/28/91)
In article <1991Jan21.142804.5726@uhura.neoucom.EDU> wtm@uhura.neoucom.EDU (Bill Mayhew) writes: > >>>On the other hand however, the RS-232 standard has connectors for +5V on >>>its line. >> >>Er... not as far as I can determine. I've never seen one that did. > >There are a few non-standard computers that supply power on their >COM ports, the Amiga series from Commodore, for example. In fact >the Amiga is a bit of a problem because there is enough oomph on >the COM port to toast improperly wired items, though normal modems >usually have Commodore's pin uncommited. Just a minor correction: This was true of the original Amiga-1000, but not of the whole series. The Amiga-2000 has a nice, well-behaved serial port that is electricaly very similar to the one on PCs and other machines. I believe this is true of the 500 and 3000 machines as well. The 1000 did have all sorts of weird signals on pins that were unused on "standard" RS232. These signals are available on some sort of auxiliary pin header inside the 2000, so an adaptor is available to create a port that looks like the one on the 1000. I think there were problems with the nonstandard signals, so thats why they changed on the later models. Incidently, the parallel port was also revised so that the later models can use standard IBM-PC printer cables, while the A1000 could not. -------------------------------------------------------------------- Steve Tell tell@.cs.unc.edu 919 968 1792 #5L Estes Park apts CS Grad Student, UNC Chapel Hill. Carrboro NC 27510