Single Ended, Single driver: In-room distortion measurements

[Gerrit Boers]

The past few weeks I measured the in-room IMD and THD for several amplifiers driving my Fostex208S in Replikon. I used my Single ended tube amp and a NAD C320 solid state amp for the tests. Later-on I also tested a Fisher KX 90 pp amp.
A quick comparison with two headphones was made and I started studying the effects of passive filters.

Summary of results:
-The distortion (IMD+THD) of a SET amp is significantly higher when measured with a resistive dummy load than when loaded with a loudspeaker.
- The SET amp has lower in-room distortion (IMD+THD).
- Headphones are not as good as their reputation, the AKG K 270 sucks.
- Passive filters seem to have a negative effect on the in-room distortion when a SET amp is driving the loudspeaker.

The details can be found on my website:

System distortion describes how I got started and shows how different loads affect the distortion of my SET amp.

System distortion part 2 contains more measurements showing a SET outperforming the NAD.

System distortion part 3 shows some preliminary results on the effects of filters.

If you can find any errors in my procedures or calculations, please let me know so that I may correct them.

I also would like to pose a question:
What are the best results for in-room THD and IMD that you ever came across? I'm looking for some benchmarks to strive for.

[planet10]

Very interesting work Gerrit, keep it coming

dave

[Gerrit Boers]

I did some additional work on full range vs. 2-way and the effects on in-room distortion. The results can be found in System distortion part 3.
It certainly looks like single ended tube amps and passive crossovers are not a good combination.

[planet10]

Originally posted by Gerrit Boers
t certainly looks like single ended tube amps and passive crossovers are not a good combination.

What happens if the XO is just a cap way up? (10k +)

dave

[Gerrit Boers]

What happens if the XO is just a cap way up? (10k +)

Good question, but this would be more relevant for a wide range driver like the Fostex208ES.
For my system there is not much to be gained form adding a supertweeter (without getting rid of the whizzer) therefor I did not investigate this.

If you look at the result for the Low+high two-tone you'll see that there is not much difference between the full range and the low pass filter, only when the system is turned into a 2-way the distortion (IMD and THD) goes up by almost 100%. Could this be because the amp now 'sees' the capacitor of the high pass filter? Maybe I'll do some more tests to figure this out.

I've been thinking about adding a Raal-ribbon as there are many good things said about this unit. The impulse response and CSD look very good (note that the scale for the CSD is just 25dB) but if you take a good look at the distortion plot you'll see that it's actually not much better (if at all) than the Fostex208S. The 2nd harmonic distortion for the Raal 140-15 is about 0.177% above 2.5KHz (85dBSPL@1m). I would really like to know how the Raal does on IMD.