So the real solution is indeed less power and less gain, but you might want to test with a cheap L pad (two resistors).
But have to worn you all that I have very sensitive hearing, as I can hear changings in my setup, that I made via PC/DSP software and without even DSP connected to PC !!!
So I'm real audiophile !!!
I wish I could get $1 for each audiophile that claims that but then fails in a proper, controlled double-blind test.
An lpad will burn power, so that is only an option if you have power to spare (including for other drivers in the case of a plate amp with a shared power supply).
Now regarding autoformer vs having a higher impedance driver to start with, the autoformer has a cost and will also bring some distortion of its own (albeit very low in this range).
With a 32 ohms compression driver (eg JBL D2430K or D2415K) you don't need any autoformer or lpad with a decent amp and gain matching.
Come one, that is in almost every hypex amp datasheet.
Eg for the amplifier of interest in this discussion the nc100hf, page 10 :
https://www.hypex.nl/img/upload/doc/ncore_mp/nc100hf/Documentation/NC100HF_03xx.pdf
6dB of distortion reduction when going from 4 to 8 ohms (and of course 3db of noise reduction)
If anyone in the UK wants to buy my Fusions, they have been used for testing and have pretty much remained in their boxes with only Wago connectors attached.
No problems, I just want to return to a simplier system that doesn’t require any knowledge to operate, and something with a volume knob and easy source switching.
I have 2 of the FA122’s with remote sensor. 20% off new prices?
Dm me if your interested to not fill the thread with irrelevant info.
No problems, I just want to return to a simplier system that doesn’t require any knowledge to operate, and something with a volume knob and easy source switching.
I have 2 of the FA122’s with remote sensor. 20% off new prices?
Dm me if your interested to not fill the thread with irrelevant info.
That's only the half of the story.....
Power levels in high frequencies are low and much lower with high efficiency drivers. The problem is that amps tend to have higher distortion at low levels. Hypex does provide the data:
In a loudspeaker with a passive crossover padding down a high efficiency driver keeps the amp in "normal" operating range. If you do this padding in the dsp tends to operating levels with higher distortion levels!
Power levels in high frequencies are low and much lower with high efficiency drivers. The problem is that amps tend to have higher distortion at low levels. Hypex does provide the data:
In a loudspeaker with a passive crossover padding down a high efficiency driver keeps the amp in "normal" operating range. If you do this padding in the dsp tends to operating levels with higher distortion levels!
Attachments
Max power goes down, and distortion climbs when it is reached (be it voltage or current limited), but within this limit a higher load always result in lower distortion from an amp.
Look at headphone amps for example, which are often tested with loads ranging from 8 or 16 ohms to 300 or even 600 ohms. Distortion always goes down with impedance.
And the myth that damping factor has any positive effect on sound quality should fall like old walls. Resistive padding should be done as soon as midrange and tweeter has substantially higher dB/V sensitivity than bass driver. Amplifier will swing happily with the voltage and digital output of DSP/DAC will provide as much bits of audio resolution and SNR as possible. Midranges with semi inductances in motors can benefit from series resistance in terms of linearity.
It is not totally a myth. Woofers (with high cone mass) benefit of reasonable amounts of damping factor, but beyond 20 or so it doesn't matter.
Amps are amps. What is interesting with headphone amplifier is the range of impedance they typically have to drive, and the measurements that follow.
But I might be biased as I do use a pair of THX 789 headphone amps to drive my compression drivers
Well of course, you loose 3dB each time you double the impedance once current is not the limiting factor.
That is all a matter of proper sensitivity matching between the drivers and amps at hand.
They are voltage limited once the impedance reach a certain value, like any amp.
Most desktop headphone amp cannot output the same power in a 600 ohm load than in a 100 ohm load for example.
And of course most headphone amps out there are mobile amps running from a phone battery, and are very much voltage limited
Yet, they still have less distortion with high loads.
If "gating" works correctly, would you need to worry about an anechoic environment?
In theory, all you need to do is measure the distance from speakers to microphone, calculate the time for sound to get from the speaker directly, then calculate the time for sound to go via the nearest reflective surface (the obvious ones being the floor below and the ceiling above) and then make the "gate" open just after one and close just before the other.
But I'm not sure how/whether you could do this with a swept signal, because it seems that you would need a period of silence after each sample (at a given frequency) to let echoes in the room die away before the next sample.
Is that how it's done? If not, how?