• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Close in jitter?

OP
Blumlein 88

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
Here are the jitter results for the Benchmark DAC2 done by Stereophile. That is about as good as it gets. 16 bit and 24 bit. The spikes in the 16 bit signal are supposed to be there. They should be at certain exact frequencies and levels with nothing else. The result below is essentially perfect. The 24 bit file has even lower spikes which are too low to pick up which is why you don't see them.

214BDAC2fig09.jpg


214BDAC2fig10.jpg
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
1.1kHz in-room

upload_2017-5-24_3-28-2.png


1.1kHz with ceiling fan on "2"

upload_2017-5-24_3-29-48.png



1.1kHz with ceiling fan on "3" (fast)

upload_2017-5-24_3-31-25.png


1.1kHz with ceiling fan on "1" (slow)

upload_2017-5-24_3-34-2.png


Control: Ceiling fan on "1" (slow) but no tone.

upload_2017-5-24_3-38-57.png
 
Last edited:
OP
Blumlein 88

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
Oh man brilliant. So sides bands of all the tones in music from rotating fan blades is like hundreds of times more than similar from jitter. I remember as a kid standing in front of the window fan and singing because of the effect.

So serious test listening must be done blind, low jitter and no ceiling fans.

I know I read a blog which examined how much moving air from HVAC would upset sound more than electronic jitter, but not ceiling fans.

So all we really need is a low frequency leslie to examine the effects of close in jitter. I do believe this dodge to close in hard to measure jitter is like the God of the Gaps argument.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
I figure it's a dopplerized reflection from the blade motion.

But I'm getting a little headache trying to figure how a receding vs approaching blade does what it is doing.

This fan has five blades, would a 4 (or other even number) look particularly different?
 
OP
Blumlein 88

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I figure it's a dopplerized reflection from the blade motion.

But I'm getting a little headache trying to figure how a receding vs approaching blade does what it is doing.

This fan has five blades, would a 4 (or other even number) look particularly different?

I think you basically have it. Doppler effect. The side approaching is reflecting sounds closer and closer as it moves. The receding one is doing the reverse. It is interesting the lower sideband signals, presumably the red shifted one are consistently a little higher in level than the other side of the main tone. That could have been a result of the angle of your mic, the source speaker and the fan however.

I wonder if you get the same effect with those rotary woofers that look like fans.

https://www.soundandvision.com/content/eminent-technology-trw-17-rotary-subwoofer

trw17-opener.jpg
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
It is interesting the lower sideband signals, presumably the red shifted one are consistently a little higher in level than the other side of the main tone.

The blade approaching the speaker angles d- the receding blade angles up (relative to the speaker), and would appear to have a better shot at reflecting from the speaker toward the mic.

Edit: Ok, got that backwards - the blade approaching the speaker angles up (blowing air down) and having a better speaker to mic reflection angle.

Tomorrow (it's not tomorrow until I wake up), I'll reverse the fan rotation for more exciting measurements.
 
Last edited:

Brad

Active Member
Joined
Nov 8, 2016
Messages
114
Likes
35
Or it could be phase modulation of the tone, due to the change is speed of the air.
I read an article once that intermodulation distortion between coax drivers is actually phase modulation, but the maths looks very similar to the Doppler effect
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Or it could be phase modulation of the tone, due to the change is speed of the air.

Watching the fan spin down (takes a minute or so even from the slow setting), and the side-spikes come closer and closer to the fundamental, I'd say it is reflections, since there is next to no air movement when it is creeping to a stop, but there are still sidebands, shrinking in .

Here is a fan spin-down movie from the slow setting - about 60rpm (wow, it takes 3 minutes):

https://www.screencast.com/t/doseZcPh

And another, from the fast setting spin-down to stop, with averaging and the Hann window to clean things up some:

https://www.screencast.com/t/1xD65tcayT

Walking across the room, R-L, then L-R, then blowing air around out in front of the mic with my mouth.

https://www.screencast.com/t/LMUiY9EwT

Sitting at my desk, 5 feet to the right and slightly behind the mic, moving my hand toward then away (arm's length) from the speakers, twice:

https://www.screencast.com/t/dD8eJfDGfLxy
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Ok, I'm done with that.

If I go to an audio show, and somebody moves, I'll demand the demonstrator restart the track, and glare menacingly at anyone else left in the room.

"Why?"

"I'm trying to listen to this equipment, dammit!"
 
OP
Blumlein 88

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I logged on here to corroborate Ray. I have two identical ceiling fans in my listening area. One 3 ft in front of the speakers and another over the listening position which is where I placed the Umik. Same results as you. In my case middle speed sidebands spaced every 9 hz to each side. Same with either fan on. Fan nearer the speakers results in slightly high levels in the sidebands.

Higher frequency tones result in higher level sidebands than lower tones though maybe not a direct relationship like with jitter. Turning the fan off and letting it coast to a stop (over 120 seconds with my fans) results in sidebands getting closer and closer to the tone.

Looks like a suitable way to simulate high levels of close in jitter. Also indicates such would show up amply on the FFT if it was there. With tones you can hear the effect in my case anyway. Sidebands up around a 2500 hz tone were only about 22 db lower in level. For experiments one could vary blade size and speed. Looking at the FFT I suppose it simulates having that $99 Schiit DAC.
 

Nightlord

Active Member
Joined
Mar 15, 2017
Messages
207
Likes
102
Location
southern Sweden
Now for the fun part... Figure out how to modulate the fan speed to produce pleasurable distortion levels based on the music played.... And what name to market the product under. :cool:
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Given the steady tone, and a steady rate of rotation of the fan...

Why are there multiple bands around the fundamental?

I suppose it has to do with the velocity of the fan blade toward the speaker, which would graph as a sine wave (of velocity) - greatest speed when the blade is at right angles to the speaker, minimal when pointing at (or away from) the speaker.

And variation of the relative velocity from the tip to the center of the fan...

So there is a continuum of fan blade velocity. And more than likely no steady "sync" of the fan speed to the tone... And the tone wave would interact with the fan at different angles (and different relative speed to/away from the speakers)...

So, why not a wide skirt, instead of bands? Why aren't the bands moving? Why whatever?

And why does The King of Audio seem to so seldom jump into these odd observations I come up with?
 
Last edited:
OP
Blumlein 88

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
I have had many of the same questions. Best I can tell taking video and counting spins with tape on a blade and a stopwatch my fan is turning at about 110 rpm. Or about 1.8 revs per second. This is a 5 bladed fan so 5 times 1.8 rps is about 9 blades per second. I saw the sidebands on my measurements at just about 9 hz spacing. The sidebands were not super sharp rather being a bit broad. So I will have to consider what is really going on here.

Looking at your medium speed (what I was also using) it appears your fan is also creating sidebands close to or just above 9 hz.

I cannot speak for the King of Audio's apparent disinterest in your observations. :cool:
 
Last edited:

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
This is a 5 bladed fan so 5 times 1.8 rps is about 9 blades per second. I saw the sidebands on my measurements at just about 9 hz spacing.

Hmmm... Reflections per second and not reflecting surface velocity? Why spaced bands? Hmmm...


I cannot speak for the King of Audio's apparent disinterest in your observations.

Hopefully he's diligently searching for a scholarly paper that discusses the scenario.


30 spins in 35 seconds = 30/35*60 = 51.43 rpm x 5 blades = 257.1 blades per minute = 4.3 blades per second

I see 4.3Hz spacing on the bands around 1000, 1500, and 2000hz tones.

Ok, so it looks like it's related to reflections per second and not blade velocity, so it's not primarily a Doppler effect (unless the numbers magically align for that too, somehow).

Increasing the volume makes more bands come out of the noise floor. I exposed 7 or 8 or 9 bands at 90dB and 1500hz.

upload_2017-5-26_22-28-56.png


I would have been happier (even if wrong) if it had stopped at 5 bands for 5 blades.

I can't explain "number of bands".

I don't see any bands around 100Hz. My guess is the wavelength is too big to interact with the blades.

Too many bands to count around 10kHz.

upload_2017-5-26_22-42-47.png


Hmm...
 
OP
Blumlein 88

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,521
Likes
37,050
You really need to be smart about this Ray. Never mind what is really going on. You have documented an actual measurable sound pollution of high end playback. One far worse than jitter. You need to slowly reveal this problem and get people all nervous about this plus the effects of air flow from HVAC systems (never mind if the latter is real or not). Then you need a researched solution, details trickled out over about a year. Then sell it to hundreds waiting with baited breath. I don't know the solution, but maybe closely spaced contra-rotating blades? Obviously this is the real reason fans in high fi gear are so bad. Not the actual noise level, just the fans spinning. Plus most computers still have fans. Clearly people have mistaken fan effects for noise on the USB connection. You can rectify that at great profit if you handle it right. You can buy some scholarly research later.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
I'd have to hire a salesman, because that is something I'm not.

I've got the hermit recluse part going for me, though.
 

Brad

Active Member
Joined
Nov 8, 2016
Messages
114
Likes
35
The bands are caused by harmonics and the bands are not continuous as there are 2 points where the reflection of the fan hit the mic (higher Frequency side band from blade moving towards mic and lower side and from blade receding from mic)

For 1500Hz, 51rpm, assuming 1m long blades, I calculate a Doppler shift would be 23Hz.

What does the ETC plot look like?

A thread on reddit suggests the effect is an intermodulation distortion - the 2 sources from the fan blade and the ceiling interfere.
 
Last edited:

DonH56

Master Contributor
Technical Expert
Forum Donor
Joined
Mar 15, 2016
Messages
7,834
Likes
16,496
Location
Monument, CO
I tend to agree with that. This is not really jitter, though I suppose could be treated as deterministic jitter. It is (Doppler/frequency) modulation of the signal. I have a mini-split so my fan is outside the room.
 

RayDunzl

Grand Contributor
Central Scrutinizer
Joined
Mar 9, 2016
Messages
13,198
Likes
16,981
Location
Riverview FL
Top Bottom