IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
Some very weird measurements H4n, Could someone explain this to me?
ktf
post Mar 20 2014, 17:47
Post #1





Group: Members
Posts: 360
Joined: 22-March 09
From: The Netherlands
Member No.: 68263



Hi all,

I was reading the SoX manual and found the spectrogram effect. I played with it, and by incident I found something which is think is really weird.

I thought, I can measure the audio output of my laptop. I hooked the line out of my laptop to the inputs of my Zoom H4n recorder, which I thought would have a better input than the combined mic/line input of my laptop. I made a linear 20Hz to 20kHz sweep in Audacity, played it through aplay, set levels, recorded it through arecord, and I got this...

H4n at lowest sensitivity, levels set to get -6dBFS


H4n at middle sensitivity, levels set to get -6dBFS


H4n at lowest sensitivity, levels set to get -12dBFS


H4n at lowest sensitivity, levels set to get -24dBFS


I was very surprised to see this, so I did the tests with my Sansa Fuse playing the files instead of my laptop, but it gave me the same results. Disconnecting the device from USB and running it standalone, recording to SD-card got me the same results as well. To me it looked like amplifier overload (instability) or something like that, so I tried a 440Hz sine combined with a sweep to check, but I got something even weirder!

Sine + sweep input


H4n at lowest sensitivity, levels set to get -6dBFS


H4n at lowest sensitivity, levels set to get -24dBFS


So my question to you is: what the hell is this? I know, this device is battery powered and can record for more than 10 hours on 2 AA batteries, but still, this looks like to much of a quality vs power consumption trade-off, right?


--------------------
Music: sounds arranged such that they construct feelings.
Go to the top of the page
+Quote Post
saratoga
post Mar 20 2014, 17:54
Post #2





Group: Members
Posts: 4925
Joined: 2-September 02
Member No.: 3264



FWIW I would use RMAA and not a spectrogram to do understand what is happening, as a sweep is not the best way to understand nonlinear effects.

But from those plots it looks like your recorder just has a lot of nonlinear distortion. Probably its a crappy ADC.
Go to the top of the page
+Quote Post
ktf
post Mar 20 2014, 18:31
Post #3





Group: Members
Posts: 360
Joined: 22-March 09
From: The Netherlands
Member No.: 68263



QUOTE (saratoga @ Mar 20 2014, 17:54) *
FWIW I would use RMAA

I've been looking into that, but AFAIK it's Windows-only. Is there anything comparable for Linux, to be specific: Kubuntu?


--------------------
Music: sounds arranged such that they construct feelings.
Go to the top of the page
+Quote Post
Arnold B. Kruege...
post Mar 20 2014, 20:10
Post #4





Group: Members
Posts: 3705
Joined: 29-October 08
From: USA, 48236
Member No.: 61311



QUOTE (ktf @ Mar 20 2014, 12:47) *
Hi all,

I was reading the SoX manual and found the spectrogram effect. I played with it, and by incident I found something which is think is really weird.

I thought, I can measure the audio output of my laptop. I hooked the line out of my laptop to the inputs of my Zoom H4n recorder, which I thought would have a better input than the combined mic/line input of my laptop. I made a linear 20Hz to 20kHz sweep in Audacity, played it through aplay, set levels, recorded it through arecord, and I got this...

H4n at lowest sensitivity, levels set to get -6dBFS


H4n at middle sensitivity, levels set to get -6dBFS


H4n at lowest sensitivity, levels set to get -12dBFS


H4n at lowest sensitivity, levels set to get -24dBFS


I was very surprised to see this, so I did the tests with my Sansa Fuse playing the files instead of my laptop, but it gave me the same results. Disconnecting the device from USB and running it standalone, recording to SD-card got me the same results as well. To me it looked like amplifier overload (instability) or something like that, so I tried a 440Hz sine combined with a sweep to check, but I got something even weirder!

Sine + sweep input


H4n at lowest sensitivity, levels set to get -6dBFS


H4n at lowest sensitivity, levels set to get -24dBFS


So my question to you is: what the hell is this? I know, this device is battery powered and can record for more than 10 hours on 2 AA batteries, but still, this looks like to much of a quality vs power consumption trade-off, right?


The H4N tests clearly show nonlinear distortion plus aliasing. That's the usual interpretation of a spurious response that goes down in frequency when the test signal is going up in frequency.

The 6 dBFS Fuze test shows groups of multiple diagonal lines that are monotonic, which appear to relate to several different orders of nonlinear distortion. The fact that the diagonal lines are in groups and are interspersed with vertical spikes is a new one on me. Seems like there is hum or noise in the test.

I agree with the comment about spectragrams not being the usual tool that is used for this sort of thing. Spectragrams are more familiar to most people as tools for analyzing real world signals like speech and music as opposed to technical problems with hardware.


Go to the top of the page
+Quote Post
pdq
post Mar 20 2014, 20:27
Post #5





Group: Members
Posts: 3394
Joined: 1-September 05
From: SE Pennsylvania
Member No.: 24233



Does the H4n record as PCM, or as lossy?
Go to the top of the page
+Quote Post
ktf
post Mar 20 2014, 20:30
Post #6





Group: Members
Posts: 360
Joined: 22-March 09
From: The Netherlands
Member No.: 68263



QUOTE (pdq @ Mar 20 2014, 20:27) *
Does the H4n record as PCM, or as lossy?

All measurements displayed here were done with the H4n as an USB interface, which I presume is lossless. As a standalone recorder recording to PCM, I got the same results.


--------------------
Music: sounds arranged such that they construct feelings.
Go to the top of the page
+Quote Post
Kees de Visser
post Mar 20 2014, 21:04
Post #7





Group: Members
Posts: 663
Joined: 22-May 05
From: France
Member No.: 22220



QUOTE (ktf @ Mar 20 2014, 17:47) *
I hooked the line out of my laptop to the inputs of my Zoom H4n recorder
Does the H4n have line inputs ?
Go to the top of the page
+Quote Post
ktf
post Mar 20 2014, 22:07
Post #8





Group: Members
Posts: 360
Joined: 22-March 09
From: The Netherlands
Member No.: 68263



QUOTE (Kees de Visser @ Mar 20 2014, 21:04) *
Does the H4n have line inputs ?

That depends on what you call line input. It has one 3.5mm jack input with plug-in power and two XLR/jack combo jacks that can both handle a line signal if it isn't too hot.


--------------------
Music: sounds arranged such that they construct feelings.
Go to the top of the page
+Quote Post
Alexey Lukin
post Mar 21 2014, 18:10
Post #9





Group: Members
Posts: 191
Joined: 31-July 08
Member No.: 56508



H4n has a built-in limiter and an auto-gain. Depending on your settings, they may be contributing to the nonlinear distortion.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 2nd September 2014 - 18:26