Topic: 42khz spike?

[page]

Today I went nature recording down by the river. It was mostly cause I could, so I set my (stock) FR2-LE to do 24bit/96khz. Using my BSCS-L (which is publicly stated as 30hz-20khz) I recorded 3 samples (XY, Omnis, blumlein). I get back and pull them up in audacity and for kicks look at the spectrum analysis. Here is what I see.

So my question is, what might be the cause of that spike? I havn't tried 24/96 recording in another environment, so I don't know if this is environment, mic, or recorder induced. Thoughts?

[rhinowing]

you didn't downsample or dither them, did you?

[F.O.Bean]

weird. I have no clue

[cleantone]

I can chime in to say that I recently used Izotope RX to downsample and dither someone's 24/96 recording to 16/44.1 for my own collection. I think it was made on a SD744 unit. There was a variety of constant frequency bumps all above 20khz. After the down conversion there was no longer any frequencies jumping out like that. I can't say I know the cause but my mind first jumped to an internal mechanical type of inherent noise. As with your pics these were completely constant volume and frequency spikes.

[rowjimmytour]

Today I went nature recording down by the river. It was mostly cause I could, so I set my (stock) FR2-LE to do 24bit/96khz. Using my BSCS-L (which is publicly stated as 30hz-20khz) I recorded 3 samples (XY, Omnis, blumlein). I get back and pull them up in audacity and for kicks look at the spectrum analysis. Here is what I see.

So my question is, what might be the cause of that spike? I havn't tried 24/96 recording in another environment, so I don't know if this is environment, mic, or recorder induced. Thoughts?

Maybe wind gust caused the spike?

[digifish_music]

Today I went nature recording down by the river. It was mostly cause I could, so I set my (stock) FR2-LE to do 24bit/96khz. Using my BSCS-L (which is publicly stated as 30hz-20khz) I recorded 3 samples (XY, Omnis, blumlein). I get back and pull them up in audacity and for kicks look at the spectrum analysis. Here is what I see.

So my question is, what might be the cause of that spike? I havn't tried 24/96 recording in another environment, so I don't know if this is environment, mic, or recorder induced. Thoughts?

diginoise?

[page]

As with your pics these were completely constant volume and frequency spikes.

These were done with the river going over the falls, no volume spikes or fluctuations were used during these samples. This exists regardless of the section that you select.

Full gain, full mic trim brings it up to about -15db. No EQ/downsample/dither/editing was done prior to screenshots. If it's digi noise, I don't hear it even on loud playback.

Here are two from a CD > RCA > 1/4" > FR2-LE. Taken from separate sections in the song. About 10 seconds each.

[F.O.Bean]

I cant offer and answer, but thats CRAZY how much info is past the 20k mark

[cleantone]

Maybe wind gust caused the spike?

Wind creates (in general) low frequency content. It pushes the diaphram in the mic which generates voltage that is equated to low frequency content. The microphone responds to the wind as if it were soundwaves. The force of the wind pushing your diaphram is as if you were recording some massive low frequency content.

I don't hear it even on loud playback.

You will be hard pressed to hear 42khz at all never mind finding a speaker that can reproduce it. Does your dog's head explode when you crank it? That is more than double the frequency that human hearing tops out at.

I cant offer and answer, but thats CRAZY how much info is past the 20k mark

But is there really information there? Probably not. Not that many microphones will be giving you any information up there. 

This exists regardless of the section that you select.

That is what I meant what I meant when I said it is constant. I am not sure where it is coming from but I suspect it is some sort of mechanial noise or false interpretation in the AD conversion. I have now seen these constant high frequency pitches from two difference 24/96 devices.

Can you unit record data with no source? Like a blank slate of line in material? Something that in theory would be recording only the noise floor of the device itself? I would be curious to see if the spikes are still there.

[page]

That is what I meant what I meant when I said it is constant. I am not sure where it is coming from but I suspect it is some sort of mechanial noise or false interpretation in the AD conversion. I have now seen these constant high frequency pitches from two difference 24/96 devices.

Can you unit record data with no source? Like a blank slate of line in material? Something that in theory would be recording only the noise floor of the device itself? I would be curious to see if the spikes are still there.

I'll try either tonight or tomorrow.

[F.O.Bean]

That is what I meant what I meant when I said it is constant. I am not sure where it is coming from but I suspect it is some sort of mechanial noise or false interpretation in the AD conversion. I have now seen these constant high frequency pitches from two difference 24/96 devices.

Can you unit record data with no source? Like a blank slate of line in material? Something that in theory would be recording only the noise floor of the device itself? I would be curious to see if the spikes are still there.

I'll try either tonight or tomorrow.

id love to know thata s well!

[page]

edit: these were all recorded with full gain on the FR2-LE.

3 pics: just so everyone is reminded, this is all after +50db amplification in audacity (for visual effect)

1) with cords, TRS, after +50db (cause it looks like nothing on the wav file without the amplification)
2) with cords, TRS, DVD player hooked up and powered on, after +50db amplification
3) no cords, just TRS, after +50db amp.

1 and 3 look similar, so I'm guessing thats the analog stage loaded and the A/D chip. I was surprised, it quieted down in sections when it was fed nothing by the DVD player.

[page]

Probably a timing signal from the unit.

If it was a timing signal that the unit produced, wouldn't it also produce it for the TRS input when it's receiving nothing? You see something nearby for the CD > TRS, but when you record at nothing, the spikes are elsewhere. I would have thought that the timing signal would have been something that had occured within a smaller range.

[page]

Hard to say, but it could vary depending on load.  Or it could actually be an intermodulation of a higher frequency signal.  At any rate, if it's -90dBFS and above 30kHz, I wouldn't worry too much.  It's going to be eliminated on downsampling anyway.

Fair play. I appreciate the responses.

[Digital Quality]

If it was analog tape, I'd say it's the bias frequency.

[page]

If it was analog tape, I'd say it's the bias frequency.

First set was a live nature recording, second came from a commercially pressed CD (DVD player > RCA > TRS > FR2LE)