Summary of the Cave Rescue Telephone Project

Fourth Edition (Web Edition)

© Copyright 1989 by John Halleck All Rights Reserved

Originally published as a Wasatch Grotto (National Speleological Society) Project Summary. It was subtitled "Where did all those 'Church of Murphy' telephones come from?"

Drawing of a handset looking into the mouthpiece and earpiece

Table of Contents

  1. Table of Contents
  2. Acknowledgments
  3. Introduction
  4. Project History
  5. Final Version of the Phone
  6. Comments on use of the phones in "One Wire" mode
  7. Comments on Networks of Phones
  8. Use of Telephone Lines as Power Lines
  9. Symbols and Parts
  10. Quick and Easy Cave Rescue Telephones
  11. Advice on Building Rugged Phones
  12. Things to watch out for


There are few projects that are the work of only one person, and this project is no exception. There are many people that deserve a word of thanks for help in the work described here.

Technical Assistance:

Grunt labor and testing, the most thankless part of the job:

Miscellaneous Support:

Special thanks to the anonymous designers at AT&T and Western Electric for designing equipment that was rugged enough to use as the basis for the designs described here.

Any time I sit down to write a list like this, I find I have left someone off, or forgotten them. People always come by afterward and tell me about it. Well, apologies in advance to those people.

Special thanks to the following, who all know what they did, and who can all have fun trying to figure out who the others on this list are: (Yes this list duplicates some of the above)
Alexander, Brent, Carol, Dave, Dave, Don, Frank, Glenn, Jean, Jim, Joe, Joe, John, Keith, Ken, Ken, Ken, Kirsten, Kristen, Kelly, Larry, Lora, Mark, Nahaj, Neil, Paul, Randy, Ricardo, Rick, Rod, Rod, Rose, Steve, Steve, Terry, Tom, Veda


The communications delay between a doctor on the surface and a patient in a typical cave rescue may be several hours, and may even stretch into a day or more in some caves. This delay could be fatal in worst case, and is bad in the best of cases. It affects the medical, logistic, and command functions at a cave rescue more than it does in any other type of rescue.

On the surface radios can be used and even if the radios fail, shouting and hand signals can back them up. In a cave rescue, except for the most trivial cases, these normal methods don't do the job. Twisting cave passages prevent "line of sight" communications, and tons of dirt prevent the reliable use of radios.

In our tests, radios were only good for line of site plus about twenty feet at both ends. Specially designed cave radios exist, but are rare enough that a incident commander can not count on their existence at a rescue without a great deal of prior arrangement. There are caves where radio will work, for whatever reason, but you can't count on it.

Old Army field telephones have become the standard solution to this problem. Despite their drawbacks, they do allow reasonable communications to be maintained.

Unfortunately army field telephones are becoming hard to come by, and their price is now putting them out of the reach of many cave rescue groups that can find them. When they are purchased, their age tends to mean that they are neither as waterproof nor as rugged as they once were. Field telephones tend to be just large enough to cause problems getting them in and out of small cave crawlways.

Another possible, and better, solution is the old Army sound powered phones. They are much smaller, and much more rugged. They can be more easily waterproofed, and have less in them that can go wrong. Another advantage is that they do not require batteries. The disadvantages are that they have a somewhat limited range, and they are not easy to find anymore.

Field telephones are a World War II era solution to communications problems. Since a good deal of progress had been made since then, we felt that a solution could be built that addressed the size, weight, and ruggedness problems. As the problem was attacked, we found that many of the problems could be addressed without complicated electronics we expected and in a manner that other groups could copy.

Project History

The original cave rescue telephones (called Murphy Phones) were nothing more than an earpiece and mouthpiece of a standard telephone handset in series underground, and both in series with a battery on the surface. These cost us about $7.50 each to produce, with most of the cost going into making them waterproof and rugged.

Mark I

Putting the phones receiver, transmitter, and a battery in series

This is a simple, but reliable, arrangement that solved the immediate problem, but had the disadvantage that it was not compatible with the standard field phones that others might bring to a rescue. Standard field phones could also, in theory, damage our phones when they were ringing the line.

What we did not realize at the time were the hidden advantages that the phones really had. The obvious advantages were that they were cheap, rugged, and simple to produce. The main advantage is more subtle. When you talk into a phone you can hear yourself talk. Telephone people call this "side tone" and the easiest way to test for it is to blow into the mouthpiece, if you can hear yourself blowing, you "have side tone". Side tone was to become the most important element of the phone testing and design.

The original Murphy phone was a series circuit that failed whenever there was a break anywhere in the line. When this happened everyone on the line had their phones "go dead" and instantly knew that there was a problem. This had the effect of getting people to instantly want to fix the problem. In multiple-phone circuits, this pinned down the point of the problem very rapidly. Even when local children were recruited to test phones the debugging problem was simple enough for them to handle.

Disadvantages of the design were that the surface phone, having a battery, was different than the underground phones. Confused practices with "Everybody grab a phone and go set up" tended to end up with either no surface phone in the net, or several.

Because these were so easy to "debug" this was always quickly cleared up. While the Murphy phones solved the problem of being a design that any rescue group could produce cheaply, the incompatibility with standard field telephones was still a large issue for any group that would have to attend rescues with other groups.

The first field phone compatible version that we tried was a design suggested by the late Frank Reid.

Mark II A

Battery, transformer, mouthpiece in series, capacitor, center taped transformer in series, earpiece between center tap and one output leg

This has a capacitor to keep any DC current on the line from saturating the transformer, and it also has the side effect of protecting the phone from the 18 Hertz ringing signal that field phones generate. The earpiece uses the transformer as an "autotransformer" and acts to impedance match the line and the earpiece. The capacitor is one microFarad, and the transformer is an eight ohm to 1K ohm center-tapped transformer. This version receives quite well, but does not transmit to the line as well as some other models. It also behaves worse than expected in very noisy environments, becoming unusable when the original phones were still going strong.

The Mark II A version solved problems of compatibility, but introduced several problems of its own. The main thing noticed was that whenever the line connection got bad, more of the current was shunted through the earpiece. This meant that the worse the connection, the louder a user sounds to themselves, and the softer that they talk to make up for it. This is just the opposite of what we want. As the connection gets worse, more and more current is shunted through the earpiece and is unavailable for the line.

This behavior made it difficult to notice when the line had trouble. There was often a long time between a line being broken and anybody noticing the problem. After enough troubles of this sort, it became clear that a phone that was compatible with standard field phones, and had the original Murphy phone's side tone behavior, would be much better.

The Mark II B addressed the problem with a second transformer. None of us were very sophisticated, and it was reasoned that if the direct-connection phone worked well, then stepping the signal down the same amount that we stepped it up would work well also.

Mark II B

Two transformers, capacitor between them, earpiece on 8ohm side of one, mouthpiece and battery on the 8ohm of the other

The Mark II B transmitted well, and restored the behavior that we liked with side tone, but because of signal losses in the transformers it sounded faint. It transmitted to the line well, but could not receive well.

If transformer losses were the problem, we thought that to fix it, we would put all the line power into the earphone...

Mark II C

1K transformer, capacitor, and earpiece in series with line, mouthpiece, battery, and 8ohm of transformer in series.

This Mark II C version does not work well at all. It transmits badly, and it receives worse. The problem is an impedance mismatch. We decided it was time to learn the real theory, and try again.

Final Version of the phone

FINALLY... We come now to the design that we have settled on. The Mark II D Murphy phone transmits well, receives well, and is as easy to debug as the original Murphy phones. It is compatible with standard field telephones, and can also be attached in parallel to a network of Mark I phones. They cost us less than $15.00 each to produce. If you need total compatibility with our original Murphy phones, a 1K ohm resistor can be attached across the binding posts of the Mark II D's.

Mark II D

Earpiece impedence matched with one transformer, mouthpiece circuit as before, transformers and capacitor in series with line

All of the Mark II Murphy phones have one more important difference from the Mark I telephones: they are able to drive the phones through a ground loop. (More on this later)

Currently the only problem the Mark II D phones exhibit is that they can neither ring nor be rung by standard field telephones. This has not proved much of a problem for our group, but caused a number of complaints at the NCRC yearly cave rescue training seminar.

There are several solutions to this problem, and I have produced several versions of ringers. However none of them are compatible with standard field telephones. Several ideas have been suggested, but so far no ringers have been produced that are compatible, simple, rugged, and cheap.

Comments on use of the phones in "One Wire" mode

"One Wire" mode, also called "Ground Return" or "Ground Loop" mode and is where only one wire is strung between the telephones, and the current returns using the ground as the return wire. Both phones have one wire plugged into the ground, and one wire between the phones.

While "One Wire" is not our default mode, we have done a lot of work with the current model in this mode. In every instance so far this has worked quite well, with a transmission of over 600 feet into a cave in one attempt, and a transmission of 1500 feet on the surface in another.

The most important consideration seems to be the contact at the point where the wire enters the ground. The state of the earth between the points seems to make little difference to the connection, but an inch of mud at both ends makes a big difference.

We have found that the best approach is to either stick the wires into mud, or to pour a little bit of water on the ground and make a mud hole to stick the wire into. An inch of bare wire sticking into mud works better than several feet of wire in dry dirt. Salty water works better, but not enough to be worth it in most cases.

We have yet to find a place where this did not work, but we are not sure enough of ourselves to establish this as our default method.

Comments on Networks of Phones

There were some initial misgivings about the use of the original model of the phones when more than two were hooked up. It turned out that the original phones worked well in both serial connections (as expected) and in parallel connections (which was unexpected).

The only real consideration was that we had to add another 1.5 volt battery in series when we went to two underground phones in series.

The final model worked well when added (in parallel) to networks of the original Murphy phones, or when added to networks of standard field telephones. We have had as many as seven phones in a network, including a mixture of original (in two wire mode) and Mark II versions (in both one and two wire modes).

Use of Telephone Lines as Power Lines

The Mark II telephones block direct current, and put the audio on the telephone lines as alternating current. This allows the lines to be used as both telephone lines and Direct Current power lines at the same time.

Power on the line must be Direct Current.. This use will not affect the telephone signal significantly. Recharging underground batteries, lighting underground command centers, and even running small heaters are practical uses. Such uses can save the transport of many of the batteries that would otherwise have to be hauled underground.

Symbols and Parts

Parts used in the phones

Telephone Microphone: Carbon Microphone Symbol This is the standard carbon microphone found under the mouthpiece of a Bell System telephone.

Telephone Earpiece: Standard Speaker Symbol This is the standard 250 ohm telephone earpiece, it comes with a metal oxide varistor between the two terminals. Leave that in place.

Binding post: Symbol for a connection I use two Radio Shack 274-661 (They come four to a package). These have proven reliable under heavy use, and allow you to connect just about anything to them. They do tend to fail when struck, at the point where they screw into the body of the phones.

Capacitor: Capacitor Symbol We use Radio Shack 272-1055. This may have a higher voltage rating than needed, but I think that the extra ruggedness is worth it.

Transformer: Transformer Symbol I use two Radio Shack 273-1380's.

In addition, there are several other parts used that don't affect the diagrams.

Battery holder: I use Radio Shack 270-398, that holds two AAA batteries. It has often been argued that using AA batteries instead would make it easier to find batteries. The Radio Shack AA holders (270-382) have proven unworkable since they fall apart when exposed to extremes of heat and cold with batteries in them. They do hold up if the batteries are stored separately. Dave Ursin's experiments have shown that the KeyStone metal battery holders will do the job.

Internal Wires: I use 20 gauge insulated stranded wire.

Terminal screws: I install these on the battery holder to give an alternate battery connection in cases where the AAA's are too hard to come by. I pull mine from a Radio Shack 274-663, and break them out with a pair of pliers.

Solder: It seems just about anything has worked so far. I use Radio Shack's because I visit them often.

Electrical tape: You could argue that electrical tape on all the connections is not needed since everything is protected, but there have been some problems with a nut coming loose and rattling around on the inside, causing intermittent shorts.

Quick and Easy Cave Rescue Telephones

What is the cheapest method for having cave communications available when you need them? Well, the approach below is a good candidate. This model is equivalent to the original Mark I model, and shares both its incompatibility with standard field telephones and its inability to run in one wire mode.

All you have to do is wire two cords in advance, and beg, borrow, or otherwise obtain two handsets at the time of the rescue.

Start with two telephone handsets with the cords attached. Cut the connectors off the end of the cords away from the handsets.

Phone cord with the four inner wires exposed

Inside the handset cord you will find four insulated wires.

Each of these "wires" has something resembling tinfoil inside. This substance is, at best, hard to deal with, and very hard to solder.

The easiest way to deal with the wire is to use crimp connectors (available at either Radio Shack, or many automotive parts places).

Just take the wire and bend it over.

Diagram of bending the internal wire over the insolated wire

slip a crimp connector over the top:Diagram of connector slipped over wire

and crimp it on... Diagram of connector crimped in place

The resulting connection is solid and will hold up to reasonable use.

You will need to put a connector on each wire, and strip off two or three inches of the outer sheath so that the wires are long enough to work with.

Cord with connectors applied, and wires labeled A-D

For both surface and underground phones, connectors "C" and "D" will go to the telephone lines.

For the underground phones, connect "A" and "B" together.

For the surface phone, connect "A" to one side of the battery, and "B" to the other side.

That's it. You are finished.

You will have phones that work, are cheap, and will run for weeks on a single 1.5 volt battery. If you are using more than one underground phone, you may need to add an additional 1.5 volt battery for each extra underground phone. The easiest way to make the connections is to screw a few wood screws into a block of wood and hold the connectors together with them.

Since these connections do not affect the handset itself, you only need the modified handset cords in the rescue cache.

Those of a technical bent should know that these work by wiring the microphone and earpiece in series. The unmodified handsets wire the microphone to the outer two.wires, and the earpiece to the inner two wires. So if one outer and one inner wire are tied together the handset is now wired in series. If one of each is attached to a battery then the battery, microphone, and earpiece are all wired in series.

Advice on Building Rugged Phones

  1. Use the older Western Electric (AT&T) handsets (Model 500's work well), the new electronic ones are nowhere near as rugged or fit for subterranean use. Test that they work before you buy them. Open them and make sure there has not been soup or anything else spilled into the guts. (We did find one that worked, that we did not open before buying, and when we opened it we really did find dried soup. Well, we hope it was soup.)
  2. Dismantle the handsets. You should see a screw-on mouthpiece, a screw-on earpiece, a receiver (with green and white wires on it), a carbon mike transmitter, and a plastic base for the mike (With red and black wires), and (attached to the base) a modular plug.
  3. Remove the modular plug, grab each of the contact wires and bend them out straight in front of the jack. You should now be able to pull the wires out of the plug. This will leave you with all of the wires accessible from the mouthpiece. They will all have copper wire pins on the ends.
  4. Break off the tabs that held the modular jack in place. (You will need the room if you install the line binding posts at the bottom of the phone)
  5. Drill holes for the binding posts, and install them. It is best to drill a hole that is just slightly too small, since screwing in the binding posts seals them watertight. Don't place them so that they keep you from reinserting the receiver and microphone.
  6. Seal the hole in the end of the phone (that the cord used to plug into) with silicone rubber. DO NOT get the rubber on unprotected metal. (It gives off a very mild acid as it cures.) Ordinary candle wax brushed on the metal will protect it from the rubber if you want to cover it. Cover the hole from both inside and outside so that the seal can't pop out from either direction. Let this cure overnight. (Leaving everything open aids an even and watertight cure.)
  7. When you put the earpiece back in, cover it with a plastic bag and carefully screw the earpiece back on over the plastic bag. Cut off the excess. (Actually we just rip it off, but cutting it off stands less chance of pulling it out.)
  8. Put the microphone base and microphone back in, and cover it with a plastic bag. Screw the mouthpiece back on and cut off the excess plastic as above.

Things to watch out for

  1. Do not over tighten the binding posts. Finger tight only! [It is possible to tighten them enough to twist the binding posts on the body and short things inside]
  2. Do not push the wires so far through the posts that they touch other posts.
  3. After attaching the wires, wrap them around the phone two or three times, and tape in a stress relief loop, so that if you tug the line, the wires take the stress and the binding posts don't.

Go to ...

This page is
© Copyright 1989 by John Halleck, All Rights Reserved
This page was last modified on February 22nd, 2001