R/E/P Community

Please login or register.

Login with username, password and session length
Advanced search  

Pages: 1 2 [All]   Go Down

Author Topic: Recording Levels  (Read 6261 times)

channel29

  • Newbie
  • *
  • Offline Offline
  • Posts: 19
  • Real Full Name: Victor Ciampini
Recording Levels
« on: April 26, 2013, 05:14:20 PM »

Hi everyone, recently recorded a demo for a friends band, front end was M-Audio Profire 2626 (using only the 8 analogue ins) going into PT8 on a Macbook Pro.  My question is where are the possible danger spots in the chain for clipping.  I was very careful (or so I thought) to set the recording levels showing in the green on the PT meters and no clip lights on the interface itself and was able to capture good kick and snare tracks, at least free of digital distortion and the well well below 0dbvu, but the Overheads were another story. During level check I asked the drummer to play normal volume, then to hit as hard as he could and still, no clip lights, good levels in the PT meters.  Later though when I listened to the tracks, the overheads sounded great when just the bottom of the kit was being played and the ride cymbal, but either the crash or the openhat/or both caused the dreaded paper ripping distortion.  The recorded waveforms showed no spikes over 0 and as mentioned I don't believe I ever let a clip light show but who knows.  Is it possible the condenser mics themselves (AT2020) got crushed by too much SPL?  I had them over the drummers head in a sort of x/y pattern but I suspect I had them pointed a little too hi at the top of the kit instead of down into the bottom of the kit.  Anyway, any help or advice would be appreciated.

Cheers

Fletcher

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 590
Re: Recording Levels
« Reply #1 on: April 27, 2013, 10:35:52 PM »

It could have been almost anywhere in the chain - the AT's should take the level without an issue [if they're in good condition], so it could be the pre-amps or perhaps some of the processing if you used any.

If the levels you recorded look OK then you should be OK with that part - so the distortion might have been in front of the conversion / level to storage.
Logged
CN Fletcher

mwagener wrote on Sat, 11 September 2004 14:33
We are selling emotions, there are no emotions in a grid


"Recording engineers are an arrogant bunch
If you've spent most of your life with a few thousand dollars worth of musicians in the studio, making a decision every second and a half... and you and  they are going to have to live with it for the rest of your lives, you'll get pretty arrogant too.  It takes a certain amount of balls to do that... something around three"
Malcolm Chisholm

channel29

  • Newbie
  • *
  • Offline Offline
  • Posts: 19
  • Real Full Name: Victor Ciampini
Re: Recording Levels
« Reply #2 on: April 28, 2013, 10:01:02 AM »

Thank you Fletcher I appreciate the reply.  No processing was used on the way in so I think you are right about the pre-amps.  I will engage the pads next time, I think I made the mistake of pushing the levels too much where it's not needed with 24 bit.

Cheers

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 599
Re: Recording Levels
« Reply #3 on: April 28, 2013, 12:53:56 PM »

If it was only the high frequencies of cymbals that caused distortion, it may be caused by the analog signal conditioning of those converters. If everything else in the chain checks out and you have THD only on those sources, the circuits may be suffering from bandwidth related problems or slew limiting.

Here I use amplitude vs THD sweeps to find those problems. You feed it with a 10k sine wave and ramp up the levels until clipping. If 10k clips before 1k hz, you have some design problems to deal with.
Logged

channel29

  • Newbie
  • *
  • Offline Offline
  • Posts: 19
  • Real Full Name: Victor Ciampini
Re: Recording Levels
« Reply #4 on: April 28, 2013, 08:02:02 PM »

Thanks Jim, I never thought of that but I guess I will give a closer listen to the tracks again and see if the distortion seems to be in the upper frequencies or not and try the 10k/1k comparison. I HOPE it was user error (cheaper to correct!) and not related to the interface!

Cheers!

Patrick Tracy

  • Newbie
  • *
  • Offline Offline
  • Posts: 34
Re: Recording Levels
« Reply #5 on: April 29, 2013, 04:53:39 PM »

It's possible to clip a condenser's internal circuitry even if nothing else downstream is clipping. No downstream gain adjustments will correct this. A pad on the mic can help with this, but I don't think the 2020 has one.

Dan Lawrence

  • Newbie
  • *
  • Offline Offline
  • Posts: 46
Re: Recording Levels
« Reply #6 on: April 30, 2013, 11:25:08 AM »

I also have had similar issues, not just recording drums, but sometimes w/ (very) close mic'd acoustic guitar, as well as other instruments. I also seem to have it at times when I've done very strict monitoring of the meter levels. (LED based meters).

My question which may relate to the starters thread, is: Is it not possible that some high frequency transients are spiking levels at rates so quick that the meters don't register them, though they do push the circuits into distortion?

dan
Logged

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 599
Re: Recording Levels
« Reply #7 on: April 30, 2013, 12:06:56 PM »

Meters have all sorts of ballistic issues and are set to react to the speed of the human eye rather than the speed of the waveforms.

"Back in my day" (I hate that phrase!) analog meters were the guessing game with transients. Usually I found they went to -10 dbu to really hit 0.

If you have a pulse generator that will show you how slow they really are at set frequencies. Otherwise, record with conservative levels.
Logged

channel29

  • Newbie
  • *
  • Offline Offline
  • Posts: 19
  • Real Full Name: Victor Ciampini
Re: Recording Levels
« Reply #8 on: April 30, 2013, 01:33:50 PM »

Great stuff everyone and I think the one big change I will definitely make is as Jim said, simply recording at more conservative levels in future, which at 24 bit shouldn't be problem, and in fact will try and keep levels as close to the -18dbfs as possible which apparently equates to 0 db in the analogue world.  Patrick, you're correct, the 2020's don't have a pad, so I will use ones that do in future.  I think also as Dan mentioned, there could be fast peaks (intersample peaks??) that clip but don't register on the meters, further reason to keep the levels more conservative.

Dan Lawrence

  • Newbie
  • *
  • Offline Offline
  • Posts: 46
Re: Recording Levels
« Reply #9 on: May 02, 2013, 02:07:54 AM »

Thanks for that tip on the pulse generator Jim, it will help to do some real analysis on my system.

Cheers

d
Logged

Fletcher

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 590
Re: Recording Levels
« Reply #10 on: May 02, 2013, 09:59:13 AM »

keep levels as close to the -18dbfs as possible which apparently equates to 0 db in the analogue world

Ah... a topic for debate for a whole lot of years [like 30+].  There is no "standard" for headroom on the "dbfs" scale vs. the "VU" scale.  Personally, I like my machines set up so -20dbfs = 0VU [more headroom that way... which means a get a little more room for error... which in my world is a good thing as I tend to err early and often].

If I remember correctly -- AVID sets their machines for -18dbfs = 0VU, but I'm not 100% sure on that so if you're running Pro-sTools I'd suggest you consult the manual.  You can set the "headroom" anyway you like.  I know some guys that have been running -12dbfs = 0VU since the 80's when only 16 bit recording was available -- it doesn't seem to negatively effect their work... so seriously - "whatever works" is fine.

Peace
Logged
CN Fletcher

mwagener wrote on Sat, 11 September 2004 14:33
We are selling emotions, there are no emotions in a grid


"Recording engineers are an arrogant bunch
If you've spent most of your life with a few thousand dollars worth of musicians in the studio, making a decision every second and a half... and you and  they are going to have to live with it for the rest of your lives, you'll get pretty arrogant too.  It takes a certain amount of balls to do that... something around three"
Malcolm Chisholm

channel29

  • Newbie
  • *
  • Offline Offline
  • Posts: 19
  • Real Full Name: Victor Ciampini
Re: Recording Levels
« Reply #11 on: May 02, 2013, 10:51:29 AM »

Haha well I can tell you what DIDN'T work, equating 0 dbfs to 0 VU lol!!  The hilarious part of it also was that I ran the first take with the -20db pad engaged at the interface, but as it was recording I was observing the waveform and fell into the "wow that looks way too small" trap so I disengaged the pads, DOH!!!

Fletcher

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 590
Re: Recording Levels
« Reply #12 on: May 02, 2013, 04:06:39 PM »

LOL... from your description I think you'll need to add the note "no electrons were harmed during the making of this recording... though some might have been bruised a bit".

Peace
Logged
CN Fletcher

mwagener wrote on Sat, 11 September 2004 14:33
We are selling emotions, there are no emotions in a grid


"Recording engineers are an arrogant bunch
If you've spent most of your life with a few thousand dollars worth of musicians in the studio, making a decision every second and a half... and you and  they are going to have to live with it for the rest of your lives, you'll get pretty arrogant too.  It takes a certain amount of balls to do that... something around three"
Malcolm Chisholm

Dan Lawrence

  • Newbie
  • *
  • Offline Offline
  • Posts: 46
Re: Recording Levels
« Reply #13 on: May 04, 2013, 12:40:45 PM »

I wonder what the pro level take on this metering issue might be.

I often track a drum machine, and will use a really simple 4-8 beat loop to check levels. This is especially important regarding the different drums pads' levels in relation to each other, and how they impact finished levels (especially near transient levels)

I've noticed that I can play that loop continuously, and yet sometimes the exact same moment of the loop will reach into the yellow, and other times it won't. Do y'all think that's more likely an unsteady output in the drum machine, or a minor level of inaccuracy as can come w/ LED meters.

Watchyall think?

thanks

d
Logged

Fletcher

  • Moderator
  • Hero Member
  • *****
  • Offline Offline
  • Posts: 590
Re: Recording Levels
« Reply #14 on: May 05, 2013, 12:43:38 PM »

Hitting the "yellow" isn't necessarily a bad thing... I think it was Terry Manning who once said "yellow is the new red" -- and he's right!!

Peace
Logged
CN Fletcher

mwagener wrote on Sat, 11 September 2004 14:33
We are selling emotions, there are no emotions in a grid


"Recording engineers are an arrogant bunch
If you've spent most of your life with a few thousand dollars worth of musicians in the studio, making a decision every second and a half... and you and  they are going to have to live with it for the rest of your lives, you'll get pretty arrogant too.  It takes a certain amount of balls to do that... something around three"
Malcolm Chisholm

Dan Lawrence

  • Newbie
  • *
  • Offline Offline
  • Posts: 46
Re: Recording Levels
« Reply #15 on: May 08, 2013, 01:17:24 PM »

I do understand that about the "yellow zone." In the case I'm speaking of I use that marker to know when I am keeping a sub mix w/in a certain window prior to mixing. W/ dotted LED meters, it seems that is the easiest place to register visually and consistently. That way, when mixing, I don't have surprise spikes from cumulative issues as I find myself sneaking the drums up. Confession: I have sort of a crappy drum machine (compared to what's available through a DAW today.) I suspect the samples, though not totally offensive, contain an extraordinary degree of transient spiking. Case in point, most often I can barely get the crash cymbals to an audibly robust level before they are spiking too hard, though the rest of the "kit" seems in balance and w/in limits. I often still have issues w/this even after post EL-8 compression (twice, once at recording, again at mix)

The second place I am finding this important is that when I go to experiment w/ final sound, and give "mastering" my best go, especially for discs w/that ... um ... "radio ready" sound ... I find something often triggers the stereo bus compression too hard, and I think its most often the cymbals.

Hence my metering questions, and this puzzle about how the same loop (solo'd) can vary in levels at the exact same point in the loop. I wouldn't be surprised if the machine was inconsistent, but I wondered if folks had experiences w/ LED meters being prone to this, or perhaps a sound physics issue I can't forsee.

Anyways, thanks for the thoughts

d
Logged

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 599
Re: Recording Levels
« Reply #16 on: May 09, 2013, 11:16:59 AM »

If you use a pulse generator, you can widen the pulse until you see the meter move. That will tell you how it was set up. A 5k hz pulse will probably not show up unless the meter release time is slow enough to see it decay, IF it fires from it in the first place.

Don't have a pulse generator? Grab a set of drum sticks and a mic and start smacking them. Do you see the meters rise to the same level as a steady tone? Do they respond with a slow release to see that decay or does it follow the decay rate of the input waveforms? If you smack them and see a -3 db level on the meter, does it sound as clean as a tone set to -3 db?

That should give enough info to determine if you can trust them, or better yet, learn/understand their timing limits.

"A man's got to know his limitations" ~ Dirty Harry
Logged

Dan Lawrence

  • Newbie
  • *
  • Offline Offline
  • Posts: 46
Re: Recording Levels
« Reply #17 on: May 09, 2013, 06:32:55 PM »

Thanks for more details on using the pulse gen, Jim. I do seem to recall one available in some of the freebie ware I've downloaded, just haven't figure out how to use it all yet. I'll do some experimenting and see what turns up.

d
Logged

Jerome Malsack

  • Newbie
  • *
  • Offline Offline
  • Posts: 29
  • Real Full Name: Jerome Malsack
Re: Recording Levels
« Reply #18 on: May 10, 2013, 11:15:03 AM »

For a pulse, with RT60 testing you can work the test by popping a balloon.  Would the same popping a balloon provide the pulse test and recording test so you can find the best setting. 
Logged

Jim Williams

  • Hero Member
  • *****
  • Offline Offline
  • Posts: 599
Re: Recording Levels
« Reply #19 on: May 10, 2013, 12:25:08 PM »

That could take a lot of balloons. Better have a party first!
Logged
Pages: 1 2 [All]   Go Up
 



Site Hosted By Ashdown Technologies, Inc.

Page created in 0.051 seconds with 23 queries.