How deaf is the FT817 at 136, 472 and 500kHz?

You may have read that the FT817 is deaf below 160m. This matters if you want to use the rig "as is" without preamps as a receiver for 136, 472 or 500kHz. So I decided to measure the performance using a signal generator. I measured the MDS on CW (the level at which I could just still hear a  CW tone without additional filtering), the level for S1, S5 and S9 on the FT817's meter. I also checked whether having the IPO switched in or out made a difference

These were my results:

136kHz (best results with IPO on)  DEAF!
MDS -103dBm
S1 -72dBm
S5 -68dBm
S9 -48dBm

472kHz (best results with IPO off)    SLIGHTLY DEAF
MDS  -120dBm
S1  -84dBm
S5 -81dBm
S9 -60dBm

500kHz (best results with IPO off)    A BIT DEAF

MDS  -124dBm
S1 -85dBm
S5 -82dBm
S9 -60dBm

What conclusions can be drawn? Well, the FT817 is definitely very deaf on 136kHz and needs a preamp to be usefully sensitive. On 472/500kHz, although the MDS is worse than at 1.8MHz, the sensitivity is arguably OK without a preamp. Yes, a small amount of high dynamic range gain with good filtering to stop overload from adjacent broadcast stations may give you a slightly better sensitivity, but the question is whether this matters when external noise may be the limitation. I have used to FT817 on 500kHz for several years and heard most of what was going, including transatlantic stations. At 472kHz it is slightly worse but still useable I think.

Crystal set sensitivity experiments

This afternoon I had a little play with a simple crystal set to see what sort of minimum discernible signal (MDS) was possible. My previous experiments with single germanium detectors had yielded an MDS of around -55 to -57dBm in the 5-30MHz range with a high impedance crystal earpiece.

Today I tried the same test with both germanium diodes and a hot carrier diode (HP2835) with and without some bias applied to the diode(s). Best results were with the HP2835 and with about 0.11V forward bias applied. Only a few uA are needed so battery drain is next to nothing. MDS was around -62dBm for a well modulated AM signal. For the tuned circuit I used a T50-6 toroid with around 20 turns with an antenna link winding adjusted for best sensitivity/selectivity and a 365pF variable capacitor .

Incidentally, forward bias for the diode could be obtained by rectifying other signals available on the antenna such as MW/LW broadcasters using a separate tuned circuit, rectifier and reservoir capacitor. I must give this a try sometime. Several hundreds of mV DC should be possible.

Then I tried switching the diode with a second signal generator at a power of between 1-10mW to see what sort of sensitivity could be obtained on a CW signal. MDS improved by around 25dB to around -86dBm i.e. I could just hear down to around 10uV. This would be a just usable level as a basic HF SSB/CW receiver.

I also tried the same tests with a pair of high sensitivity ST-3 headphones that Michael AA1TJ kindly managed to find for me. These are much lower impedance than the crystal earpiece so I experimented with antenna match and the match of the diode to the tuned circuit, expecting a better performance at the best settings. Disappointingly, the results were several dB worse than with the crystal earpiece.

Now, the results with the crystal earpiece are very close to those (without bias) that I obtained some years ago, so I don't think my hearing (at 63 years old) has changed much for the worse. So, Michael must have incredibly sensitive ears as there is no way I was able to hear down to the levels he could with his passive receivers.

At the moment I am playing with the AM crystal set with -62dBm sensitivity listening to shortwave broadcasters. Fun.

How sensitive can headphones be?

Having made several short-wave crystal sets in my time I have nearly always used the little "deaf aid" very high impedance crystal earpieces as the acoustic transducer. I thought these were very sensitive being able to copy HF AM signals down to around -55dBm to -60dBm. However I recently read somewhere that with the old WW2 DLR5 low impedance headsets one could hear an applied audio signal as low as 2uV (-101dBm), which is remarkably low.  

So, I began to wonder just what ARE the best headsets to use if one is after the maximum efficiency converting an audio frequency electrical signal into sound pressure in the ear?

This graph (from Wikipedia) shows the sensitivitity of typical humans at different ages. Someone of my age (63) is likely to be less able to hear as low as someone aged 30 years.

As some of these old military surplus headphones can be hard to find nowadays in good condition, I wonder what is the most sensitive headset/earpiece that can be bought new "off the shelf". Any ideas please?