A sample text widget

Etiam pulvinar consectetur dolor sed malesuada. Ut convallis euismod dolor nec pretium. Nunc ut tristique massa.

Nam sodales mi vitae dolor ullamcorper et vulputate enim accumsan. Morbi orci magna, tincidunt vitae molestie nec, molestie at mi. Nulla nulla lorem, suscipit in posuere in, interdum non magna.

Now That was Different–The Kellogg Wave Master

I’ve said before that the early years of a technology produces some pretty strange animals.  Designers are trying new things and companies are rather open to experimentation as they try to catch the consumer market.  At the same time they are saddled with the fluctuating economics of the new tech.  I enjoy collecting such radios from the 1920s since they use unusual circuit designs.   I like to get them working just for the satisfaction of understanding how different they are from other radios.  Sometimes it helps me understand why they fit a niche in their market.  I saw such a radio some time back.  My good friend, Dan, had found a Wave Master, built by the Kellogg Switchboard and Supply Company of Chicago.  I was amazed to see the inside of the radio, but hesitant to buy it since it would be quite a bother to get it home from Chicago.  It also required some veneer and panel restoration that I was afraid might be above my pay-grade.

That was a few years ago.  I had an opportunity to buy a later version of the Wave Master, but it was so compartmentalize by shields that you never could see what made this design unusual and special.  I decided to buy Dan’s radio.

The Wave Master is a tuned radio frequency circuit (TRF).  By the mid 1920s TRF radios were a predominant design.  With no major patents associated with the circuit many TRF makers entered the market.

If you have a radio background, you will recognize that a capacitor and an inductor may be placed in parallel to form a tuned circuit—one that will pass a signal of a specific frequency while rejecting other signals.  The specific signal that is passed depends on the values of the capacitor and inductor.

A mid-1920s TRF radio typically had three of these arrangements each with a companion vacuum tube.  The first two narrowed the radio spectrum to one frequency, then amplified that signal before sending it on to the next stage.  After the first two radio frequency tuning and amplifying stages, a third capacitor/inductor tuner fed a detector tube that converted the signal to simple audio.  Two audio amplifying stages boosted the signal, then fed it to the horn speaker.

This Steward Warner shows off the typical TFR design.  See those three dials?  Each one tunes one of the three tuned circuits.

Look in the back.  You can see that each dial is connected to a brass colored variable capacitor.  In front of each capacitor is a green coil, the corresponding inductor.  Pretty straightforward.

Most TRF radios looked a lot like this.  Did you notice that each coil is oriented 90 degrees different than its neighbor?  That’s how they prevented the magnetic field from one stage from interfering with the next.  Most designers had not yet begun to provide shielding between stages.

Let’s talk for a second about tuning our Steward Warner.  As you may have gathered by this point, you had to use all three of those dials to tune the radio.  If you knew about where on the dials to find your station, it wasn’t so hard.  Turn each dial close to the right spot, then play with each until you had a good signal.  The dials didn’t exactly track with each other and the difference between their settings would not always transfer to a different station at another place in the band.  It was a good idea to write down the settings of each dial for your favorite stations.  If you were just wandering through the ether listening for something new, finding that station may be a bit like manipulating the dial of a combination lock.  The tuning of the first stage is a bit broad—you did’t have to be exactly in the right place to begin to hear that station.  The second and third knobs required more careful tuning—they were more selective.  Why didn’t they put those three capacitors on one shaft and control them with one knob?  The stages of radios of that era didn’t track with each other very well.  Dan, my friend who sold me the Wave Master tells me that later TRF radios were able to use single knob tuning because newer, more sophisticated tubes had enough gain to overcome any non-linearity in the tuned stages.  So until screen-grid tubes showed up, lots of radios had three tuning knobs.  Now you know a bit of what most radio users were putting up with in 1925.

Well after all that, let’s get back to the Wave Master.  If you open it up, it looks a lot different than our Stewart Warner, even though both have very similar circuits.

We see three big variable capacitors, but they are not connected to knobs.  In fact, the warning on the label tells us that they have been calibrated and locked at the factory.

We see the green inductor coils paired with the capacitors to form the tuned circuits.  But they are not normal coils or transformers.  They are variable inductors; in fact they are radio frequency transformers with variable secondaries.  Inside each coil is another coil that can be twisted mechanically.

This movable coil is connected in series with the outer secondary coil of the RF transformer.  In the right position the inner coil enhances the inductance of the outer coil.  Twist that inner coil some and its magnetic field opposes and reduces the inductance of the outer winding.  So instead of using variable capacitors, this radio uses variable inductors.  Now that is a paradyne that didn’t last very long in consumer radios.

Variable inductors were commonly used in radio tuning a few years before, around 1920.  These were called variometers and variocouplers.  Variocouplers are variometers that have tapped coils.

That leads us to one of the reasons that variable inductors were generally abandoned in favor of variable capacitors by the time of our Wave Master.  Most variometers are unable to tune the entire AM broadcast band in one sweep of the dial.  Most radios using inductive tuning had a multi-position switch that controlled those taps I mentioned on the variocouplers.  You actually had to switch between segments of the broadcast band in addition to manipulating the tuning knob.

Inside of the Wave Master we see those tap switches.

To tune in a station the user had to determine in which of nine segments of the broadcast band his station resided, switch to that segment, then use the tuning knob to find it.

All of that bother to tune a station is what amazed me.  The mechanical linkages used to connect the variometer tuning transformers as well as synchronize the tap switches certainly cost money.  Why would Kellogg put themselves at a financial disadvantage to other TRF makers just to make a radio that was rather complex to tune?  Kellogg seemed to have success with this design as they continued the variometer/tuning transformer design in later models.

It wasn’t until I powered up the Wave Master and operated it that I began to understand.  As I explained earlier, a three-knob TRF was not a walk in the park to tune either.  Yes, every station on the dial was available in one 180 degree sweep of the tuning knobs, but you did have to adjust all three to find your music.

In some ways chopping the broadcast band into nine segments was simpler.  The chart inside the Wave Master’s lid told you which segment corresponded to which small segment of the band.

It also told you approximately what numbers on the tuning dial (marked 0 to 180) corresponded to which frequencies within each segment.

Now that sounds a bit complex, but at the time it must have been almost as good as push button presets for your favorite stations.  My other radio expert, Doctor Dave commented that Kellogg may have designed the variometers to only cover a small bandwidth to make them track each other better.

So until single-knob-tuning became standard for radios, the Wave Master may have filled a niche.  I don’t know, but I’m glad to own one—it’s an interesting radio.

For those who like more detail, here is some more information about this unusual radio.

By the mid 1920s most battery powered radios had only two rheostats.  One adjusted the filament voltage for all of the tubes (compensating for wilting batteries).  The second was a volume control that attenuated the signal or just controlled the filament of one tube to limit the volume.  The Wave Master was an over achiever with four filament adjustments!  The radio has a tendency to howl from inter-stage interference setting up oscillations.  Apparently Kellogg decided to address this by allowing the user to play with the gain of each stage, balancing between fire-breathing local stations or weak distant signals.  You have an adjustment for the first RF stage, second RF stage, detector, and the audio amplifiers.

Kellogg felt that once the detector and audio amp tubes were adjusted for your locale and antenna, they could be left alone.  Those adjustments were hidden inside the set.

The two RF amplifier adjustments were on the front panel, labeled Volume and Degree of Power (what does that even mean?).

Anyway, the instructions for all four rheostats, plus tuning are detailed on the label under the lid.  A factory technician measured and labeled four intermediate frequencies (in meters) found in each of the nine “wave zone” settings.

I also notice that the case, chassis frame, and instruction label all bear the serial number of the radio.  The Wave Master also gave me another pleasant surprise.  It’s about the only 1920s radio I’ve purchased that had good audio transformers.  Another just like it sold on EBay recently that was advertised with good transformers too.  Maybe Kellogg Switchboard and Supply used better copper wire than most radio makers.

Kellogg used spring-mounted tube sockets to control noises from microphonic tubes.  The tubes merrily bounce back and forth when you bump the radio.

All components are surface mounted to a phenolic panel with the wiring run underneath.

My Kellogg came to me in very good condition internally, but a bit distressed on the outside.  The top veneer was both burned and broken away from water damage (I’m sure there is a story behind that).  New veneer on the lid was the first step before refinishing.

The panel was rather degraded.  Very little of the lettering was left.

If you have a Wave Master with good lettering, let me know.  I’d like to experiment with making decals from good photos.

Kellogg wasn’t the only maker to use variometer tuning in the mid-1920s.  Lester Jones designed variometer/RF transformers for TRF sets that are electrically similar to Kellogg’s, but mechanically are quite different.  Marketed under the name Telos Vario-Transformers, they used sets of flat coils that rotate in between sets of flat stator coils.  Alan Douglas describes them in his excellent series, Radio Manufacturers of the 1920s (Volume 2, Pages 75-76).  The Telos tuners were used in Jones’s own Melco Supreme radio as well as some other brands as well.  I’d love to experiment with a Telos-equipped radio.  If you have a TRF that has three stacks of red fiber sheets where the tuning capacitors should be, send me an email!


From Dan:

Yes, the three dials were used because it was difficult to track them exactly when ganging them all together.  There was a  transition period where two dials were used with TRF’s. It was found the you could gang and track all tuned circuits other than the first one quite closely. But the first one was affected by the antenna impedance, which could vary with the length and type of antenna used.  So they ganged all circuits after the first one, and used a separate control for the first one. RCA got around this with the Radiola 60 by using what they called an antenna tube. This tube did not have a tuned circuit on the grid but was connected directly to the antenna. This stage didn’t have much gain, but was only for the purpose of isolating the following ganged tuned circuits from the variable antenna impedance.

2 comments to Now That was Different–The Kellogg Wave Master