Topic: cut waveform! Why?

[nicegrin]

I was just transferring some audiotapes the other day and noticed the following:

One of the channels seem to have it´s upper half of the wave form "cut off" (see attached picture!)
while other parts of the same channel and the other (right channel) don´t have this!
Very curious about what is causing this. Am I doing something wrong when transfering?

This is how I´m doing:

Cassette player -> CA-9100 pre amp -> Line in on Edirol R09HR with gain set to 40 of 80. The rest of the gain from the CA-9100.

Thanks!
N
 

[Gordon]

I'm no expert on cass transfers but I know you shouldn't run it through a pre amp.  I've done them using the V3 but only using the A/D of the V3 not the pre.

[nicegrin]

why not?

Got much better sound running through the CA-9100 then running directly into the R09-HR.

Sound is still good just wondering what is causing the waveform to look like that. Might be the tape copy aswell, but still curious what can make a waveform look that odd...

Thanks!

[it-goes-to-eleven]

I'm no expert on cass transfers but I know you shouldn't run it through a pre amp.  I've done them using the V3 but only using the A/D of the V3 not the pre.

Curious, since there is no "0 gain" setting for the v3 and I am aware of no way to bypass the pre-amp.

Are you sure it isn't an issue with the original tapes?  How do the channels compare on a deck with good meters?

[SmokinJoe]

I agree, take the preamp out of the equation.   For one thing you preamp is feeding 9V of plug in power upstream to a cassette deck output wasn't intended to take that, it might not burn it out today, but it might someday soon.  And secondly, because it appears that perhaps you may be brickwalling the preamp... feeding it a hotter signal than it can take, and that's the cause of your funny looking wave.

If you run Cassette > recorder and don't get the funny waveform then it's clearly the problem.  That flattopped waveform is "distorted" in the sense that it's a "distortion of the true original signal".  It may not sound horrible, but it will probably sound better when it's not modified.

As far as the preamp making it "sound better", that's a particular flavor the preamp adds, probably a slight high frequency boost.  You can add that with EQ when you get it to the computer.

Comparing a V3 and a ST9100 in this regard is kind of apples vs oranges.  Yes a V3 is a preamp, but in this role it's just an A/D, and it has plenty of headroom before you brickwall it.  To get zero gain on a V3, you set the jumpers to 20db attenuator, and set the gain to 20.

[it-goes-to-eleven]

To get zero gain on a V3, you set the jumpers to 20db attenuator, and set the gain to 20.

But it is still acting as a pre-amp, with no bypass.

[nicegrin]

Thanks! Sounds reasonable, but why is the waveform just cut in one end? The picture shows only one channel! I would have expect to see both ends cut if the signal is too hot for the preamp!

Thanks!

I agree, take the preamp out of the equation.   For one thing you preamp is feeding 9V of plug in power upstream to a cassette deck output wasn't intended to take that, it might not burn it out today, but it might someday soon.  And secondly, because it appears that perhaps you may be brickwalling the preamp... feeding it a hotter signal than it can take, and that's the cause of your funny looking wave.

If you run Cassette > recorder and don't get the funny waveform then it's clearly the problem.  That flattopped waveform is "distorted" in the sense that it's a "distortion of the true original signal".  It may not sound horrible, but it will probably sound better when it's not modified.

As far as the preamp making it "sound better", that's a particular flavor the preamp adds, probably a slight high frequency boost.  You can add that with EQ when you get it to the computer.

Comparing a V3 and a ST9100 in this regard is kind of apples vs oranges.  Yes a V3 is a preamp, but in this role it's just an A/D, and it has plenty of headroom before you brickwall it.  To get zero gain on a V3, you set the jumpers to 20db attenuator, and set the gain to 20.

[live2496]

You only need 60-74 db of dynamic range to capture the cassette recording.

There isn't any need to involve the preamp since levels are matched already anyway. (ie. line level output = line level input)

The level will show to be lower in the DAW software, but you can boost it. Really you only need to capture 12 significant bits to reproduce this recording with accuracy. Below that will be the noise floor of the cassette.

So I would leave some headroom. Capture in 24-bit format. And boost the level (normalize to some desired level) when processing the file.

[live2496]

Thanks! Sounds reasonable, but why is the waveform just cut in one end? The picture shows only one channel!

That's weird that the positive and negative values of the same file even look different. It might be that the software did not draw the waveform correctly. I have seen some other stuff like that here.

Gordon

[anr]

I get exactly the same on the audio from my relatively cheap camera.  It is essentially a poor design in the electronics, which is typical when the primary use or most expensive parts are elsewhere (e.g. the lens or in a tape deck the transport). 

To "restore", try one of two things.

1. Run Waves C4 in "Uncompressor" mode
2. Run a De-Hum / DC Offset programme at whatever your mains frequency is  (50/60). 

[Belexes]

I see this on my old analog masters when doing transfers. It's not the transfer itself, but the original recording in which one of my channels brickwalled while the other was perfect.

[live2496]

Yes, it's possible that you have some DC voltage entering the left channel. This can eat up headroom and therefore you see it brickwalling.
http://en.wikipedia.org/wiki/DC_offset

If you want to salvage what you recorded, remove the DC offset via software and re-open the file. Another trick is to filter the low frequencies at anything below 20 or 30 Hz.

[SmokinJoe]

Brickwalling frequently squashes on the top and not the bottom.  I can't say why, just that I've seen it plenty of times before.  Why on the Left, and not the right?  No idea.

[live2496]

It could mean that there is leakage of some DC into the left side but not the right.

As far as the top/bottom, perhaps it has to do with the charge of the DC voltage + or minus.