Need some help recording vinyl

[aireq]

I'm a DJ and I'm trying to record all my vinyl to my computer so I can use CDs instead. But I'm having a number of issues

Here's the equipment I'm using

Technics 1200-M3D Turntable,
Tascam X-9 DJ Mixer
M-Audio 2496 Sound Card

Now for the first issue. I have the turntable connected to PHONO 1 on my mixer. I start playing a record and adjust the trim for that channel so that the level is just barely below 0 db (not clipping). My mixer has a menu option to bypass the master fade so I set it up that way for recording. Then I hook up the mixer to my sound card through SPDIF (Coax connection). As the record is playing the meter on the mixer says the signal is just below 0 db, but the M-Audio Detal control panel meter on my computer says the signal is a lot lower at around -10 db. Wavelab's meter show the same thing. What's going on here? Should I just turn up the trim on the DJ mixer until the soundcard meter shows the right level and ignore that the mixer is showing +10 dB? I don't see any option anywhere to adjust the input level at my soundcard?


I should also mention that the meter for program 2 on my mixer is not functioning. I'm using program 1, and audibly everything seems ok with it, but the program 2 meter just doesn't show anything. This might be related, or it might not be. BUt I thought I mentioned it.


OK the next issue. I go ahead and set the trim on my mixer so it's showing +10 DB and my sound card and wavelab are showing about -3 db. On my sound card control panel I reroute the SPDIF in to my speakers so I can monitor the recording. Everything sounds ok, and I go ahead and record the record to a 44.1khz/16bit WAV file. As it's recording the song sounds fine.

After the recording finishes and I try to play it back the recording is full of static. There are regular loud pops all through out the recording that I did not hear while the recording was being made. Here's what part of the waveform looks like in Wavelab. The red arrows mark some of the pops. I've also tried another recording software (Audacity) and got the same static and pops in the recording. Does anyone have any idea what is going on?

Also, I get the same issue with static and inconsistent metering using RCA instead the SPDIIF connection. So it would seem it's some issue with my sound card?

 

[Treeline]

Have you tried to record it analog into the soundcard? See if you still have the instability going on. If not, then you will have begun to isolate the problem. Sounds like an A/D conversion outside the computer is misbehaving, or somehow there are two conversions in the signal path for some reason. The pic looks like real instability - bad or wrong resistance cable for the signal and so forth. Might sound better with a straight analog approach as the sound card A/D converter might be better than your mixer when dealing with such a low strength signal.

Another thing is to check your mixer manual to see if the phono input anticipates an actual RIAA signal, or a line level signal. Do you actually have an RIAA grade preamp in the signal path? Mixers are inconsistent with this, so don't assume anything about the phono preamp. It could be looking for a line level signal, which could explain your inconsistent level readings and at least some of the poor performance.

 

[aireq]

Have you tried to record it analog into the soundcard? See if you still have the instability going on. If not, then you will have begun to isolate the problem. Sounds like an A/D conversion outside the computer is misbehaving, or somehow there are two conversions in the signal path for some reason. The pic looks like real instability - bad or wrong resistance cable for the signal and so forth. Might sound better with a straight analog approach as the sound card A/D converter might be better than your mixer when dealing with such a low strength signal.

Another thing is to check your mixer manual to see if the phono input anticipates an actual RIAA signal, or a line level signal. Do you actually have an RIAA grade preamp in the signal path? Mixers are inconsistent with this, so don't assume anything about the phono preamp. It could be looking for a line level signal, which could explain your inconsistent level readings and at least some of the poor performance.

Yes it behaves exactly the same with the RCA outputs. Same static same low meter in the computer. I actually, just figured out the problem with the static. I had the recording program set to the WDM driver and not the ASIO driver. I switched that and now no static.

Though the level thing still seems a little weird.

 

[aireq]

OK I take that back I'm still having problems with the static as well.

I'm recording off of the SPDIF, but monitoring off of the RCA's out of the back of the sound card. It sounds find as I'm recording, but then the actual recording has a lot of pops and crackles

 

[Treeline]

Try analog to see what happens.

 

[SouthSIDE Glen]

Is that meter on your mixer a VU meter or a dBFS meter? If it's measuring on the analog side, and can go higher than 0, then it's a VU meter, and there's nothing wrong with your levels. Just keep recording around 0VU. Yes, that will come out well below 0dBFS on the digital side, it's supposed to.

Bring the levels down and who knows, that may help get rid of those pops, which could possibly be caused at least partially by your over-driving the preamps in the mixer by pushing everything up past +10dB VU.

If you don't like the resulting digital levels, you can just push the gain on the resulting file. As a DJ, you're going to want to do some volume balancing between your tracks anyway, because peaking everything at 0dBFS doesn't mean that everything is going to play back at the same volume. So the initial digital recorded volume is just not all that important.

EDIT: I'm not sure if this is significant at all, or if it is, what it may mean, I'm just pointing it out for maybe someone else to comment on: I notice that every one of those arrowed spikes, save one, occurs in the same general voltage region, they are all (save one) happening within just a few dB just on the negative side of zero crossing. And the only one that is not - the third arrow on the right channel - may just be something spurious as that is the only one of the group that occurs on ony one channel.

Possibly coincidence, I don't know, but any ideas from the EEs out there on what might cause such noise on both channels at the same time always around about the same negative voltage?

G.

 

[aireq]

Is that meter on your mixer a VU meter or a dBFS meter? If it's measuring on the analog side, and can go higher than 0, then it's a VU meter, and there's nothing wrong with your levels. Just keep recording around 0VU. Yes, that will come out well below 0dBFS on the digital side, it's supposed to.

The meter on the mixer list labeled dB and the scale goes from -24 to +14
The meter in WaveLab is also labeled dB but it's scale goes from -60 to 0

So take it my mixer is a VU meter and my software is a dBFS meter? I was always under the impression that I should record as loud as possible with out clipping to maximize the dynamic range of the recording? So when I see the meter in the software bouncing around 10 I always though I should make it louder.

Bring the levels down and who knows, that may help get rid of those pops, which could possibly be caused at least partially by your over-driving the preamps in the mixer by pushing everything up past +10dB VU.

Well again everything sounds ok as I am recording, as well as if I monitor off the Mixer itself. So the pop/crackles are happening in the computer somewhere not the mixer's preamp.

If you don't like the resulting digital levels, you can just push the gain on the resulting file. As a DJ, you're going to want to do some volume balancing between your tracks anyway, because peaking everything at 0dBFS doesn't mean that everything is going to play back at the same volume. So the initial digital recorded volume is just not all that important.

Again maybe this is a VU vs dBFS issue again, but I was under the impression that it's better to get as close as you can to 0dbFS to maximize the dynamic range.

EDIT: I'm not sure if this is significant at all, or if it is, what it may mean, I'm just pointing it out for maybe someone else to comment on: I notice that every one of those arrowed spikes, save one, occurs in the same general voltage region, they are all (save one) happening within just a few dB just on the negative side of zero crossing. And the only one that is not - the third arrow on the right channel - may just be something spurious as that is the only one of the group that occurs on ony one channel.

Possibly coincidence, I don't know, but any ideas from the EEs out there on what might cause such noise on both channels at the same time always around about the same negative voltage?

G.

 

[aireq]

I should also mention that a while back I recording at least 100 records that sound fine with the same equipment (and the same SPDIF cable. But this is a new install of windows because I got a virus a few months back. So considering that I'm thinking the issue is software.

What I don't get is I have the SPDIF digital line going into the sound card, and then the AD converter is converting that to analog which is going to my speakers. It sounds find while it's being recorded, but the actual recording is what has the pops/crackles. But at that point it all digital? It's basically writing the SPDIF stream to the hard disk?

Another thing I should mention. It's not consistent. The first half of the recording may be fine, but then the 2nd half might have pop/crackles. So I'm starting to wonder now if it's some other software on my computer that interfering?

Eric

 

[SouthSIDE Glen]

The meter on the mixer list labeled dB and the scale goes from -24 to +14. The meter in WaveLab is also labeled dB but it's scale goes from -60 to 0

So take it my mixer is a VU meter and my software is a dBFS meter?

That sounds like either what's called a peak program meter or a peak-reading VU meter, but is certainly not a dBFS meter. As such, it is not using the same scale as the meters in your software, and is measuring what is hapening on the analog side, not the digital side. Because of this, you want to set these meters for levels that the analog circuitry is designed to handle. On a peak program that would mean letting you meters hover around +4, on a VU meter that would mean averaging around 0 (occasional peaks may jump a little higher.)

From there, it's up to the A/D converters and how they are calibrated as to exactly what dBFS level those numbers convert to. This varies from converter to converter, but 0VU/+4PPM can equal something anywhere in the range of -20 to -14dBFS.

For the full dope on all this, check out the metering section of my "Metering and Gain Structure" applet located in the lower right-hand corner of www.independentrecording.net.

I was always under the impression that I should record as loud as possible with out clipping to maximize the dynamic range of the recording? So when I see the meter in the software bouncing around 10 I always though I should make it louder.

That's a common misconception (also discussed in the applet). When you boost the gain, you're boosting everything, including the volume of the analog noise floor. Considering that your typical album/turntable/cartridge/phono preamp signal chain isn't going to have a dynamic range any greater than about 65dB at the best (and often closer to 50-55dB), and that 16-bit digital recording has 90dB of room and 24-bit (preferable) has 140dB of room, you have PLENTY of room to get the full analog range on either one without having to press everything to the top.

But that is kind of a side issue to what I meant. What I meant was that recording everything to peak at 0dBFS does nothing to guarantee that the tracks will sound as though they are the same volume. Perceived volume is not determined by peak levels. You could have three songs all with peaks at 0dBFS and all of them sounding as though they had different overall volumes.

Letting the converters set the volume well below 0dBFS will preserve headroom for you if you want to (as is becoming more common with digital DJs) adjust relative levels of each of your tracks so that they play back at mopre or less the same perceived volume without having to pre-cue and preview playback levels, setting them on the faders on the fly. You can just cue up your playlist and either decide to "look busy", hit on the ladies or head out back to finish that spliff (or all three ) instead of standing there setting levels on the fly.

Well again everything sounds ok as I am recording, as well as if I monitor off the Mixer itself. So the pop/crackles are happening in the computer somewhere not the mixer's preamp.[/QUITE] Yeah, I didn't catch that the first time around. I think you're on the right track with that in looking at the interface drivers. If switching inputs does not help you, I'd look into the support services for that software and see what may be happening on the driver version front (newest is not always best).

G.