Questioning Conventional Wisdom

[Bossobass Dave]

While riffing on one of my bass guitars, a thought popped into my head:

 

Conventional Wisdom: "Odd-order harmonics are more offensive to humans than even-order harmonics."

 

Why did I ever blindly accept that thought? Harmonics are, um... harmonic. Any combination of harmonics with the fundamental are consonant. Consonance is not 'offensive' to humans. Dissonance is 'offensive' and is any combination of tones that are NOT harmonic.

 

I also began to think about my in-room frequency response and how harmonics are related to room gain. Below the modal region, which is, of course, the frequency whose length is greater than 2 times the rooms longest dimension, the frequency response should have no ripples, whether it is flat or rolled off by filters in the signal chain.

 

In my room, the modal region ends somewhere between 20-21 Hz. So, why do I see a rippled response below that point?

 

It's simply because the resonant frequency of my floor system is approximately 5.8 Hz. Below that point, my signal chain causes a roll off and at that point the response rises as the resonance is excited by the subwoofer output at that frequency.

 

BUT, a resonance has harmonics of its fundamental. If that's true, then I should see a rise in response at 11.6 Hz and again at 17.4 Hz when a sine sweep, which contains those harmonic frequencies of the resonant frequency, is used as the input. Check the graph attached.

 

So, where I had been thinking that there is a universal dip in every room between 10-20 Hz, and I always had difficulty in determining where to spot the response curve to determine the +/- of the non linearity of the in-room response, I now believe that the curve should be spotted at the bottom of the resonance ripple effect (as the graph shows) and that the universal dip is more likely the response that lies between the resonance peaks.

 

This correct spotting of the basic response without room induced non linearity also serves to allow one to precisely calculate where room gain begins and at what rate it increases as frequency decreases. Check the 2nd graph attached.

 

I'll be looking at the dozens of saved in-room responses posted by others over the years to see if there emerges a similar pattern.

 

Just thought I'd pass these random thoughts on as I had a few minutes to do so.

 

 

[Ricci]

I have always heard that the odd harmonic thd is more offensive but I believe the correct teminology would be more easily heard.

 

Here is a bit of info from my old room and my new room. My exact gain in the new room is a bit of an estimate but this should be close. I have some data on it but it is on another hard drive. (I recently had my old measurements pc take a dump.Haven't swapped out the drives yet.) The old room has a spike at 4Hz while the new one has a large gain at 7Hz. The newer more solid brick and cement room seems to have more uniform response with less variation. You can see where my signal chain starts rolling off below 6Hz.The old room has something going on at 4Hz increments, 8,12,16,20,24,28,32Hz.

 

 

 

 

 

 

[MemX]

Interesting thread, thanks for posting, Bosso 

 

Ricci - could it be the case that your old room response (and, indeed, any wood-frame room response) was more varied because there are more components in a wood-frame room, meaning each will have different resonant frequencies, whereas a concrete and brick room will have a lot fewer components (in as much as each element is more solid and bricks are all one size, as opposed to a wood-frame room that might have batons of different lengths, plasterboard of different widths, support bars of different heights that alter the unbraced panel size of the plasterboard on top...) and therefore only have one or two resonance peaks?

 

(Longest and most confusing question in the world  lol)

[3ll3d00d]

It's simply because the resonant frequency of my floor system is approximately 5.8 Hz. 

How do you know your floor system has a resonant frequency of 5.8Hz?

 

If I look at mine then I can see peaks at 6.2Hz, 11Hz, 19.6Hz with 1st mode at ~41Hz

 

[Bossobass Dave]

How do you know your floor system has a resonant frequency of 5.8Hz?

I once measure FR at different positions in the room. The resonant peak was the same in all positions. Adjacent to the HT is a laundry room that shares the same floor system. I then took the mic into the laundry room and closed the door. Of course, the entire bandwidth of the FR was attenuated... except the resonant peak, which stayed virtually the same as all mic placements inside the HT.

 

I simply deduced that it was the floor resonant frequency that caused that result for lack of a better explanation.

[Ricci]

MemX I suspect that is not too far from the truth.

[LTD02]

higher order harmonics are more offensive.

3 is greater than 2.

3+5 is greater than 2+4.

so any series of n length will always be worse for the odds simply because it will always be higher in order.

my guess to where the 'rule of thumb' comes.

 

but be careful, harmonics alone are very different from harmonic distortions.  the former is critical to good sound, the latter are introduced when trying to recreate natural sounds in unnatural ways.