World's First Transistor Radio

8 minute read

The Regency TR-1

January 2025



I thought that my Invicta 332 was an excellent example of an early transistor radio until a client of mine got this Regency TR-1 as a Christmas present. In terms of size, it is approximately 1.5 times the size of the Invicta ... and there's a good reason ... The Regency TR-1 was the world's first commercially produced transistorised radio in 1955. So, at time of writing this piece, this is a 70 year old transistor radio. And what is quite impressive here is that this one works.

The TR-1 was manufactured by Regency Division of I.D.E.A Inc of Indianapolis, who essentially took the familiar 5-valve domestic radio and 'transistorised' it, to the extent that it does retain some legacy features from its forebears, like the tin-plated metal chassis and the fairly chunky tuning capacitor.

The advent of the transistor, developed at Bell labs back in 1947 heralded a desire to miniaturise, and the portable radio receiver was an obvious choice. The available service sheet carries a couple of quaint phrases when describing the RT-1; One being that it 'conveniently fits into the pocket of a man's shirt'. Try getting away with saying that these days! This one came complete with its own leather carrying case complete with belt-loop. The case had shrunk somewhat, so extracting it from the case took a lot of effort and patience.




I did say that this TR-1 works. However that was not the case when I first extracted it from the case. With the inevitable absence of the required 22.5V battery, I connected it up to a variable power supply and gently 'eased' the voltage up to 20V. Nothing went bang, but neither was there any noise from the loudspeaker ... not even the slightest of hisses.

Time to open it up 'proper'. The tuning knob is retained by a single knurled screw. Removing this reveals three cross-head screws, of which only need the bottom one (with the flat head) needs to be removed.


The TR-1 was the result of a close co-operation between Regency and Texas Instruments, whose transistors were employed. So new and experimental were these transistors that since they were not yet sold commercially, they do not bear a Texas type number. I was hoping that it was not a failed transistor that was preventing the receiver from working. Instead, since there are four electrolytic capacitors in the circuit, my money was on at least one of these failing ... as was the case with my Invicta. In fact, all four of the ancient electrolytics had failed! In the photograph on the right C9, 40uF/3V is second from the top. This was completely open-circuit. Moving down we have C17, 5uF/25V which measured 3nF, then C19, 4uF/3V which measured 200pF, and finally C21, 40uF/3V which measured 1.5uF.

C9 and C21 were replaced with 47uF, while C17 was replaced with 4u7F and C19 with 3u3F. The capacitor at the top is C1, 20nF ceramic which lost a leg during testing. This was replaced with 22nF.
Sadly, replacing all these capacitors had no effect on the silent radio.



The cause of the silent receiver turned out to be more profound than technically complex. The loudspeaker was open-circuit. Initially it even looked like it was part of the chassis, but it turned out to be soldered into a circular hole in the chassis and since it was only soldered at two points on its circumference it was soon removed. Oddly there is only one actual terminal for the voice-coil. The other end of the voice coil being attached to the speaker chassis (see the photograph on the left). The miniscule braided wires leading to the even thinner wires on the cone had completely corroded away!


The photograph on the right shows the hole into which the loudspeaker is soldered. Although not very clearly shown, I also discovered at least one lifted track (C1) and a couple of areas where component leads are being used make up for damaged tracks. The metal chassis is also connected to the 0V track and is even used to provide a 0V pathway, so it is important that it is soldered to the tracks in a couple of places. According to the service sheet, the PCB wasn't hand-soldered but 'dipped', by this I'm assuming an early form of wave-soldering.




In order to gain 'safe' access to some of the damaged tracks, to repair them and remove excess solder, it is necessary to partially lift the chassis away from the underside of the PCB. To do this, it is necessary to un-twist three chassis lugs as well as de-soldering them. As can be seen in the photograph on the left, this proved somewhat tricky since one of the lugs is situated right underneath the detector diode D4, which in this case is a Tungsol TS117. Alternatively, some TR1s used a Raytheon CK706A. Like the Texas transistors, this is a Germanium device.



I managed to repair the loudspeaker by replacing the missing wire with self-fluxing SWG-40 wire. It was now possible to listen to an injected IF-signal at 262KHz. However, the Local Oscillator transistor was not oscillating ... evident by the voltage on the Emitter being only about +1.5V. The transistor gave odd results on my transistor tester, and I thought this might be an actual transistor failure. I was unable to find an old but unused NPN Germanium transistor with a suitably low hFE, so I had a second look at the original transistor and decided to clean the legs. Bingo! It tested well and returned an hFE of 35. The voltage on the transistor Emitter was correct and the oscilloscope showed a healthy Sine wave.


Sadly, there is precious little of interest on Medium Wave these days, unless you like listening to guys with an unhealthy obsession in football who are able to talk incessantly on the subject for hours at a time ... or you can listen to Radio Scotland ... and that's more or less your lot! And as it turned out, the little TR-1 burst into life and I was able to perform a full realignment as detailed in the Service Sheet. Although it was a bit intermittent with Radio Scotland dropping out periodically. This issue appeared to be associated with the wiring between the PCB and the Ferrite Rod Aerial (or Loop-stick as they call it across the pond). Note the frighteningly thin cotton-covered wire in the photograph on the right.


I found another aerial-related intermittent contact. I thought there was a problem with the windings, but it turned out to be more subtle. The arrow in the photograph on the left points to a solder-point where the join between the two windings is connected to the chassis which turned out to be loose. The solution was made easy since the chassis is tin-plated. I simply soldered the wires to the edge of the chassis, see right. It was necessary to ensure that the joint is not proud of the adjacent screw. What looks like a loose wire in the photograph on the left turns out to be little more than excess wire ... which I trimmed off.


Above, left to right ... The TR-1 PCB showing the four replaced electrolytics and the new 22nF C1 next to transistor X1 (yellow can). The middle photograph shows the repair to the open-circuit loudspeaker and the final photograph shows the finished (repaired) Regency TR-1. Note that a previous owner has widened the slot for the volume-control/switch by literally gouging the plastic away. The volume control knob is a tad problematic. The knob itself has clearly split in two at some time and has been glued back together. Contact between the knob and the potentiometer shaft is by friction alone and thus it is prone to 'migrating towards the end of the shaft.

Conclusion:
Despite the slightly battered look of the 70 year-old plastic case. The Regency TR-1 actually performs rather well. The audio quality isn't too thin and sensitivity is surprisingly good. OK, I did replace five capacitors, but at least the semiconductors are all still original ... and that's what makes this such a remarkable little piece of history.

A final quaint quote from the Service Sheet ... Because of the extremely small size of the receiver and because of the use of transistors, the technician is in "strange territory" as soon as he opens the case.
Regency, TR-1