If you own an American AC/DC radio using
300mA heater valves, chances are it uses a line cord resistor to drop the
120V mains to the required heater voltage. And, in this day and age, there's
an even better chance that the resistive element has gone open circuit.
The asbestos and fabric insulation has not always stood the test of time,
and the inner rubber insulation has often gone brittle. The most common
problem of course is that the nichrome resistance wire hasn't taken to
continuous flexing over the years and has numerous breaks in continuity.
Line cord resistors have not been made for many years, N.O.S. examples
are rarely seen, and when they are, still suffer the same problems.
Restorers of vintage radios have used a number of methods to get the radio going again, but all unfortunately detract from originality, sticking out like dogs balls, and some have electrical disadvantages. Before reading on, go to this page to learn more about these methods.
Making a new line cord resistor.
Faced with the open line cord resistor problem with all my Meck FM converters, and the Emerson CF255, I had a good think about how to deal with the problem. The Emerson CF255 is so small there is no room for any kind of internal dropper resistor or transformer. While the Mecks allow a small heater transformer to be fitted, or in the case of the version with 150mA heaters, a normal resistor, the ideal repair is still to use a line cord resistor. With such things no longer available it was time to think about making my own.
What is needed is insulated flexible resistance wire, and two ordinary conductors for the live and neutral. Then, the whole lot needs to be enclosed in a flexible sheath.
The latter is the easy part. Having learnt the technique with wiring my Model T Ford, shoelace should work as an outer covering here, and look the part.
The live and neutral can be any normal plastic wire.
The tricky part is flexible resistance wire, and insulated at that! What to use? Nichrome wire out of radiators and jug elements isn't flexible, and isn't insulated. Plus, the resistance is usually too low
Then I had a brainwave. What about the element of an electric blanket? Previous measurements in the past revealed something in the order of hundreds of ohms. The current rating is about right, it's insulated, and very flexible. Designed to have the weight of someone moving about on it all night, it should also be very rugged; in fact considerably more so than the original line cord resistor!
The type of electric blanket used in Australasia consists of a length of plastic covered resistance wire, sandwiched between two layers of wool, which forms the underblanket.
The element may actually consist of two lengths of wire, which are switched in various series and parallel configurations to control the heat level. More expensive types apply the 240V to the one element via a thermostatic control. Some older types used 32V or 17V transformers with either a thermostat or switched tappings. The resistance of this type would be too low for this application, however.
Once the resistance of the entire length had been measured, it would be easy to calculate how much would be required for a particular resistance value. All good in theory, but I had to find an electric blanket to pull apart! As it happened, I saw one in a trailer, waiting to go to the tip, whilst out riding one afternoon.
Pulling it apart & examining
I had thought it would be a simple matter to simply pull the wire out of the blanket, but it turned out it was secured by some kind of adhesive. This meant having to cut the wool alongside the element for its entire length. With about 16m of element per side of a double blanket, it took several hours.
Just over 16m of flexible insulated resistance wire was retrieved from this electric blanket.
The element from this blanket actually
consisted of two windings with a layer of insulation between them. Somewhat
reminiscent of coaxial cable. This allowed only one physical element, but
it could be internally switched for the three heat settings. For line cord
resistor use I could see this being useful. For example, one could use
one or both lengths of resistance wire in series or parallel to get the
required resistance with an appropriate length. An interesting feature
is that the inner winding has less ohms per metre than the outer winding.
Inner and outer conductors can be seen here. Each has a different amount of resistance per metre. Care is required in stripping the outer insulation. It's best to cut down the side and peel off or the inner (blue) insulation is pulled off also.
Much to my delight I also found the wire
was solderable. The next step was to find out the characteristics of this
wire. At a length of 1645mm, the inner resistance was 819R, making
it 0.498 ohms/cm. The outer resistance was 1704R, making this 1.03 ohms/cm.
With inner and outer in parallel, it was 0.34 ohms/cm, and in series we got 1.54 ohms/cm.
Shoelace outer covering.
Using shoelace as an outer covering is a technique I learned from when I was making extra wiring harnesses for my Model T. Being black fabric it looks like the original, and it's easy to slip over modern plastic wire. However, only the flat kind of hollow lace can be used. This is not so commonly available in black, and in long lengths. I had run out of the original which I'd used for the Model T and it was no longer available. It took quite a bit of searching, but eventually found 180cm football boot laces in Big W.
Constructing the new line cord.
The procedure to make the new line cord is quite straightforward, although slipping the shoelace over it requires patience. Knowing the ohms/cm value allows the wire to be measured and cut to get the right resistance for the particular radio. For the Meck FM converter, 358 ohms is required. One could use the inner wire, or the outer wire, or both in series to get this resistance. The inner wire has less resistance per cm than the outer, so the cord would need to be longer than if the outer was used. Alternatively, an even shorter cord could be made by wiring the two in series. Typical appliance cords (including the original on the Mecks) are 1.8m or 6 feet, so ideally I'd like to keep to around that length.
Testing with 300mA flowing through the different combinations showed that the two in series just got too hot. The power dissipation was concentrated in too short of a length resulting in it being too hot to touch comfortably, and the plastic was getting a bit soft and springy. The inner wire, on its own, ran warm, but not unduly and would need 716cm. While the actual cord could be made shorter by folding it back and forth twice, the cable diameter would get too big. Trying the outer wire on its own worked out well. Here, 315cm gave 360R.
The resistance wire only had to be folded back once to get a 1.8m cable, effectively making a four wire cable (along with the two ordinary conductors). And it didn't run too hot.
Obviously, what worked for me is likely to be different to others. Different electric blankets are likely to use resistance wire with different characteristics, so the ohms/cm value I quote is not likely to be the same as what you use. While 300mA of current should be OK for most blankets (equivalent to 75W@240V), the key thing is that you don't dissipate too much power in too short of a length. The line cord will run hotter the shorter you try to make it, and could be a fire hazard.
The example I'm using with the Meck is probably the worst case scenario as the heaters require 12.6V @ 300mA. Most AC/DC radios have a heater string voltage much higher than this which means less dissipation in the dropper. Also, no normal series heater type radio uses valves drawing more than 300mA.
The important points for making a 300mA line cord dropper are:
Dial light tapping.
Prior to 150mA valves, where the rectifier has a dial light tapping on its heater, sets with 300mA valves used a tapping on the line cord. The purpose of this is to shunt the dial lamp, so as to protect it from the switch on surge, when the valve heaters are cold. I discuss this in reference to the Perco FM tuner here.
There is no reason why such a tapping cannot be provided with the homemade line cord. All that has to be done is to cut away the outer insulation, attach the connecting wire, and then cover the join with heatshrink.
Don't use for 240V!
Seeing you're making a new line cord resistor, the thought might occur that you could make it longer and therefore convert the radio to run off 240V without a stepdown transformer. Indeed, I'm aware of this practice being done in the UK where imported US sets were sold. Firstly, these American AC/DC sets have poor, or no isolation between their chassis and the outside world. Secondly, the original aerial and earth isolation capacitors (if they exist at all!) don't have high enough breakdown voltage. Thirdly, isolation between internal parts and outside connections may be inadequate. Finally, the switch contacts may not be suitable for 240V. And remember, the way the power switches are wired in these sets means the chassis may be live when switched off, although at neutral potential when switched on. It appears from the construction of all my American radios, project designs in their magazines, and other examples of electrical wiring, that Americans treat their 120V supply as harmless, and think nothing of exposed live connections. If you're into collecting U.S. electrical stuff you should already have a 240:120V double wound transformer. Use it!
Terminating the resistance wire.
While the solderability of the wire I used meant I could simply run it to the terminations in the radio, it did look a bit fragile like that. Instead, I terminated it with short lengths of normal hookup wire. For mechanical and electrical protection it's necessary to cover the joins in heatshrink tubing. As I found, the heatshrink can still slip off when being threaded into the shoelace, so it's wise to put some contact cement on the wires before heatshrinking.
Only the outer conductor was used. The inner conductor was covered in black heatshrink first to prevent it making contact with the outer conductor. Then the terminating wires were attached and covered in heatshrink. Future line cords will have contact cement under the heatshrink to prevent it being pulled off the resistance wire.
Once the resistance wire has been prepared,
it can be taped, at suitable intervals, to the ordinary conductors - I
used figure eight twin flex.
The resistance wire and twin flex are assembled into a cable. The resistance wire has been folded back to halve the length of the cable.
Then the whole lot is inserted into the
end of the shoelace and gradually fed in. I find that by bunching up the
lace makes it looser and easier to thread over the wires. Once in position,
it can be then pulled to make it tight again, simply by tightly running
your hand down it several times. The ends need to be bound in cotton to
secure it and stop it fraying. More than likely, it will be necessary to
use more than one length of lace. In which case, slip the end of the extra
lace over the bound end of the first length, and again bind with cotton.
Covered in black shoelace, it looks like the real thing. As can be seen, resistance is 360R to suit the Meck CX500.
The cord is now complete and ready for connection to the 120V plug and radio apparatus.
The Meck CX500 now has a proper line cord resistor. Heater voltage is spot on with a 120V supply. The two white wires connect directly to the plug, while the two black wires are the 360R heater dropper.
As you've probably worked out, the first line cord resistor I made went on to one of my Meck FM converters. It was a complete success. The heaters get the right voltage and I haven't had to do any bodgie modifications. This is also the first time I've used a line cord resistor, and from all the stories I've read I was curious about the amount of heat given off. It certainly isn't excessive, and I don't see it being a fire hazard. Having said that I wouldn't leave it coiled up on a piece of paper and run all day unattended, but if laid out I see no problem. I can see why it was such a popular method for voltage dropping. Dissipating 32W over a length of 1.8m outside the cabinet is far more effective than being concentrated inside a small cabinet. Next line cord will be made for my Emerson CF255.