the Money, Part 2: Vocals, Vocals, Vocals
So Matt asked a question in one of the comments to a previous post that seemed like I could do an entire post on:
I’m definitely always looking for ways to improve the sound of my VOX, especially our lead VOX, any chance you can outline what kind of effects or plugins you tend to throw down on your VOX channels?
Vocals seem to be a big theme lately which I think is a good thing. I did a bit on sort of the philosophy of vocals in an earlier post so I’m basically going to answer Matt’s question with my general approach to handling vocals.
Mic choice is very important to me and always the first place I start. While 9 out of 10 times you can get away with a 58 on a vocal, in my opinion there are a lot of better options out there for vocals these days. The key for me is to find the right mic for each singer because not everything works the same on everybody. I’m a big fan right now of the Neumann KMS105. I also like the Audix stuff; they can give you great stage rejection, but don’t work so good on inexperienced vocalists that don’t get right up on a mic. I guess I tend to prefer tighter pickup patterns and condensers when possible, but I’ll still go for the tried and true 58’s and Beta58’s at times. In fact, in a one-off situation where I’m stuck with a box of mics I’ve never used or a bunch of 58’s, I’ll always grab the 58 because I can get it to work. But for my current gig where I get the same singers rotating through on a regular basis I like to try different stuff over time to see what works best.
Beyond mic choice, my biggest “processor” is just a simple onboard parametric EQ. In my opinion you can get a decent vocal on a good mic at least 90% of the way there with just a decent EQ on a good console. There have been many times over the years where an EQ was all I had. I ride my lead vocals pretty much all the way home, so this is really all I need. When I EQ a vocal I tend to stick pretty much to subtractive EQ. My goal is to get rid of any mud or garbage and get the vocal sounding as natural as I can; sometimes it helps to listen to singers in person and not through a PA to hear how they really sound. I’ll high pass vocals probably around 150–sometimes lower and sometimes higher depending on the vocalist, but I always try and go as high as I can. I never just immediately start EQ’ing; I always listen first and then go after any problem areas. Vocals are maybe one of the only instruments I’ll spend a lot of time working on solo’ed instead of in context. Putting them in context with the song is important, but the vocal always needs to sound good on its own so hello solo-in-place… Sometimes I even do the Scovill thing and start with the vocal and then push the band up around it…
EQ starting frequencies on a vocal for me are probably 200-250, 400-600ish, 800-1000ish. BUT BUT BUT BUT BUT these are just starting points for me in the relative ballpark; I always listen and often dig around. Moving on…2-4k for me is where the hurt can start so I’ll watch those frequencies, but the big stuff for me are the low mids where it gets muddy from proximity effect. I rarely find myself boosting EQs on vocals these days. Part of that might be my current PA’s, but I just like to cut first. If I am boosting it’s probably in that 2-4k range–I said I watch it, but I’ll still play there–and then sometimes up top, 12k and up to give it some air. One thing that’s nice with the Venue is you get to visually see what your EQ curve is. While some guys might look at it to see what the EQ is doing, I like to look at the darker area to see what’s left or what will actually be heard. Make sense? I actually have an EQ preset on the Venue where the EQ is flat, but my ballparked starting points are pre-dialed which is nice because in an emergency situation I can just grab the knobs and turn things up or down without having to look at specific frequencies. This works nice because then I find myself actually listening vs. watching myself mixing on the screen or making sure I have picked out some specific frequency I think is the right one in my head without listening. I just pull bands down and then slide them around ’til it’s right. Screens on a digital console can be a blessing and a curse…but I digress…
I think the biggest place a lot of engineers get into trouble on vocals is with EQ. I hear a lot of guys leaving a lot of the low/low-mid stuff in for some reason. For my current tastes, mic proximity effect isn’t so hot, and it’s especially bad on all those 58’s you find everywhere. If you look at the frequency plot of a Beta58, 2″ vs. 1/8″ distance is almost 10dB of gain at 200 Hz. Yuck yuck yuck. The result to me is most often unintelligible muddy gibberish vocals. I like my vocals to cut…you know, like on a CD. Of course don’t suck all junk out so it sounds like a telephone, but if you’re struggling to get your vocals to cut I would start around there. If things start to get harsh or hurt, try putting some back before you grab those hurt frequencies. While the rest of the stuff I use on vocals definitely helps take it to the next level, in a way it’s all sort of just icing on the cake after EQ. If I’ve got it I’ll use it, but I think it’s important to be able to get a vocal to work with just a mic and the console EQ.
My next preference after EQ would be for some dynamics processing. I’ll start on the analog side, and I’m not overly picky here. I’ve got some Drawmer stuff I use right now–DL241 I think. I’ve used DBX stuff before, too; I just prefer something relatively standard with attack and release controls. If there’s good stuff around, I’ll always use it a bit, though, but if all I have access to is lower end stuff, I might opt to not use any comps. A lot of times it just depends on what kind of mood I’m in. an>I tend to compress maybe 3-6 dB with a 2-4:1 ratio, but the comp doesn’t always need to be working. My experience has been that you can screw up good singers with compression, but with the not-so-good singers it can make a big difference. With vocal compression I’m really just trying to even out the level of the vocalist a bit and raise the average level, but not too much because dynamics are a good thing. I’m also careful to make sure that I’m not boosting too much of the stage bleed or sucking the life and tone out of the vocal.
Staying on the analog side, we have some BSS 901ii dynamic EQ’s in the rack that I like but try not to overuse, although when I have them available I go to them before I go to an overall comp like the Drawmer. My typical signal chain on a lead vocal has these at the top followed by my regular comp. I like the dynamic EQ for smoothing out inconsistencies that an EQ alone won’t get. For example, they’re nice for singers when they pull back from a mic to belt. If I compensate too much for proximity effect with the channel EQ, the vocal might thin out if the singer backs off, but with the 901 I can let the gear automatically dial back the EQ at that frequency when my singer pulls back to maintain a little more consistency. While the 901 is great to have, I still prefer to do my general EQ on a vocal and then use the 901 for a little extra tweaking. I installed a Rane C4 about a week ago to use the dynamic EQ mainly for de-essing. However, I might use it for some of the same things as the 901 because I think I like it a little better, but I haven’t played enough to know where I’ll end up on that. The dynamic EQ is just another nice toy to have, but not a necessity. Although, in my current gig that Rane C4 for de-essing is pretty much necessity.
Continuing in analog world, the next thing I’ll throw on a vocal is maybe some verb and delay, however, this is an area where I’m pretty conservative. Lots of verb on vocals is a huge pet peeve of mine because 1987 was 20 years ago. I haven’t been in a lot of live rooms that were so dead they needed a ton of verb so I lean towards the natural sound of the room and add verb when necessary to get a vocal to sit a little better. In the past I haven’t spent a lot of time with verb because I don’t like most of it, although I’m spending a little more time lately. Basically I like something fairly tight like a short plate. Lately I’ve also been experimenting with setting pre-delays to song-specific tempos. I always set the delay for a song’s tempo and ride it for effect, but I’m going to save that explanation for another day because it’s harder to explain than it is to just demonstrate. I think I can go a little overboard sometimes with delay when used in an effect capacity so I’m consciously trying to dial back my use a bit right now; I’m starting to feel like it’s just too easy and a cheap effect. On the other hand, I am starting to use delay on a vocal for an entire song at times, but it’s pretty rare–I did it last weekend, though… If I can get enough processing to add another delay on the Venue, I’ll probably add a couple more instances to have maybe one for “space” and another for “effect”.
Moving to the Venue, I get a little more specific and picky for vocals because I typically have more time to spend on a vocal along with more available options. I tend to do most of my EQ’ing and compressing with McDSP’s Channel G plugin. Like I said above, I have a preset I’ve been using that ballparks me so I’ll try and remember what I’m doing there. The EQ is the Neve emulation and I tend to narrow the Q’s as much as I can–they’re pretty wide to start. Sometimes I’ll try some of the other stuff like the API or one of the SSL emulations, but I tend to stick with the Neve most of the time. FYI, the Channel G EQ emulations are G–SSL G series, E–SSL E series, N–Neve, and A–API. I think the compressor is API with a soft knee and using the feedback compression setting. Over compressing here can kill the top end of the vocal, though, so I’m careful with that. Sometimes I’ll ditch the Channel G comp for maybe a Fairchild or an 1176; it really depends on the vocalist. I like the Fairchild on singers with a lot of dynamics; it’s pretty smooth and you can hit it hard. For a while I was using the McDSP MC2000 multiband comp on the vocals as well in somewhat the same capacity as the 901. Lately I’ve been leaning on the Serato Rane Dynamics plugin for more of this. I typically use one instance on every vocal with the Dynamic EQ freq centered somewhere around 7k, soft knee, really fast attack, and a ratio of 2:1-10:1 for de-essing. I might also add a 2nd instance to clean up around 200 Hz or so, but this is rare right now. Last thing on the Venue is typically a Cranesong Phoenix(probably Dark Essence). I put the Phoenix on pretty much everything. Basically I’m pretty much out of control when it comes to the Phoenix…seriously out of control…
For verb on the Venue I typically use ReVibe; I hope to pick up the TC stuff at some point, though. I have a factory preset I might have modified a bit, but I don’t remember what it was called—probably short vocal plate or something like that. For Delay I use the digital delay on Echo Farm. At least I think I’m pretty sure it’s the digital one. There is also an emulation of a TC2290 in there that I’ll use from time to time, but you’d have to look up what it’s called in the manual.
The last thing I’ll do with vocals is sometimes called parralell compression, upward compression, NY compression, or as labeled on my console “Vox Spank”. Basically I take the lead vocal and double-buss it to the master buss and a subgroup which also routes to the master. I insert a compressor on the subgroup, and I’ll typically compress it a bit harder than what I’m doing on the individual channel. Then I slide the group up under the vocal to taste. This can go a couple ways. It can really help me get a vocal to cut through or it can really hurt my gain before feedback; sometimes it does both. On the Venue side I typically use Smack for the compressor with maybe a 6:1 or higher ratio, a medium attack, and then probably a medium to medium-fast release. I vacillate on the amount of compression; it’s more than what I do on the channel, but I don’t kill it. On the Venue I also stick the Phoenix on here(more compression…), and sometimes I’ll insert an expander to turn the processing down a bit when nobody’s singing to limit stage bleed and feedback potential–I put the expander BEFORE the comps, though. On the analog side I might EQ this a bit, but it all depends. I generally only use this for singing and pull it back for talking singers. I sometimes also ride the fader here with every line of the song instead of riding the general channel fader. I’m not sure it’s the best way to do it, but it’s just what I’ve gotten to work.
I think that’s about it. Let me know if you have any questions and PLEASE share things that you’ve found to work in your world because this is also an area where I’m also always seeking new tricks.