Al Klase - N3FRQ - 23 OCT 2017
My test set is based on an Hewlette Packard 353A Patch Panel. These were originally used with an audio oscillator and AC voltmeter to mahe measurements on telephone circuits. There are a lot of these kicking around, hamfests etc.,and they are generally pretty inexpensive.
I stripped out all the original components except for the precision 600 ohm attenuator, added Ben's "FILVORA" and built the following circuit.
I also included an 8:600 ohm transformer, but the radio I'm using for a source delivers enough voltage not to need it.
Due to the 4 dB pad and the fixed 26 dB loss of the FILVORA, the attenuator dials, that have been offset 30 dB, indicating the power delivered to the load in dBm regardless of the load impedance.
- Set the input level for 0 VU on peaks.
- Set the attenuator for approximately -50 dBm
- Adjust the FIVLORA knob for maximum observed level from the head set.
- (Impedance matching, especially with actual audio,
is like horseshoes and hangrenades. You only need to come
close.)
- You'll find yourself settling on one of 2 or 3 possible settings, and will ikely observe frequency response changes from one setting to the next.
- Turn the attenuator down until you can just make out the speech. Perhaps missing a word here or there.
- Record this setting as the headset minimum discernable signal..
A lot depends on the sensitivity of your ears. I'm down at least5-10 dB from 20 years ago.
For younger ears:
- Traditional 2000 ohms headsets, crystal earplugs, etc.: Sensitivities in the -70 dBm neighborhood.
- Sound powered headsets: Below -80 dBm
- Some modern one-dollar earbuds are quite sensitive.
Check transformer losses.
One might actually measure the sound pressure level coming out of the phones. Add -3 dB if you only measure one phone at a time. To do this correctly you need artificial ear fixtures, one for "over the are" 'phones and another for "in the ear" phones. I couldn't figure out a practical way to do this.