Surprised that people are forgetting how relatively easy it is to detect CW (and when it came along, SSB) with a regenerative detector. As others has mentioned, regenerative detectors were just about universally used with simple one ortwo stages of RF amplification ahead of the regenerative detector, and even with no amplification stage ahead of the detector. I was puzzled by the comment that one could not DF with a TRF receiver. Regardless of whether or not an IF stage was used in
a superheterodyne or whether the detection was performed at the signal frequency, DF, at least by use of a steerable loop antenna to determine the null in a signal level of a transmitter of interest can either be done on a CW signal with either a regenerative detector or BFO operating at either the signal frequency or an IF frequency. Low frequency radio beacons used to, and where still active, send their station ID and sometimes long dashes using MCW tone modulation of a steady carrier so they could be received by receivers that did not have a BFO or regenerative detector.
The 1930s vintage design RAK VLF/LF
receiver is a TRF receiver with two stages of TRF amplification and a regenerative detector. It has very high selectivity over its entire 15 kHz to 600 kHz, obviously hugest at the lowest frequencies, and comparablen contemporary superhets using the common nominal 455 kHz IF frequencies of the day.
It’s follow on RBA in WW2 was also a TRF was also a TRF design, although I believe it used a tracking RF oscillator at the signal frequency for a BFO. I believe the injection level of the RBA was adjustable to prevent overwhelming the incoming signal that was to be detected.
Chris AJ1G Stonington CT