Two-mic technique

[b1j]

[Note: I modified the title of this thread on June 27 because the conversation migrated over to mic setup alternatives.]

I've recently taken to recording my J-45 with two sE Electronics sE8 SDCs, at XY about 9 inches from the sound hole, so they're aimed at the 12th fret and the bridge. The combination of close mic and that boomer guitar is giving me a little more bass than one might want in a mix. So far it's only been solo guitar and I kind of like the full sound spectrum. There's plenty of headroom and no clipping. But now I'm thinking about how best to tame the low frequencies for larger arrangements.

The sE8 has an 80 Hz and a 160 Hz switchable low-cut (high-pass) filter. (Before these I only had the Rode NT-1A which does not have a filter.) Would I be better off recording with the 80 Hz filter on the mic, or just waiting to EQ the guitar track in the DAW (Studio One 5 has a fine array of EQ options)?

Of course the other world of options is to try mic placements farther from the guitar to lessen proximity effect (the sE8 is cardioid, but has only about 2 – 3 dB drop at 45º), and with different aiming patterns. But for now I want to try to fine-tune close micing to lessen artifacts from my less-than-perfect room, and XY pattern seems to me a good way to avoid phase issues.

Filter at the head of the chain or EQ in the mix?

[jim1960]

Originally Posted by b1j

The sE8 has an 80 Hz and a 160 Hz switchable low-pass filter.

It's not a low pass, it's a low cut. It prevents frequencies below a certain threshold (80hz and 160hz in this case) from being recorded. It's helpful, for example, if you're recording a boomy guitar and want to cut back that low end boom.

[sdelsolray]

A 80 Hz high pass filter may interfere with a guitar's frequency range, particularly if you use dropped-D (6th string) tunings. You can certainly use a plugin in Studio One to accomplish the same result, either when tracking or mixing (preferred).

[b1j]

Originally Posted by jim1960

It's not a low pass, it's a low cut. It prevents frequencies below a certain threshold (80hz and 160hz in this case) from being recorded. It's helpful, for example, if you're recording a boomy guitar and want to cut back that low end boom.

Yes, of course. Dumb mistake. I've corrected it in the original post. Thank you.

[b1j]

Originally Posted by sdelsolray

A 80 Hz high pass filter may interfere with a guitar's frequency range, particularly if you use dropped-D (6th string) tunings. You can certainly use a plugin in Studio One to accomplish the same result, either when tracking or mixing (preferred).

My track does have some important F# notes (92.5 Hz), and I wouldn't want those to be attenuated. I think the DAW's EQ will be more sensitive and adjustable, so I'll record everything onto the track instead of cutting at the mic, and then use the DAW's EQ when mixing.

It looks like the drop-D note is at 73.4 Hz. Interesting.

[KevWind]

I generally don't use mic filters and usually set the high pass filter in my DAW with a steep rolloff starting at 70 hz

[keith.rogers]

I'd expect some boom from that size guitar with that setup.

I'd say give it a try if you're setting XY that close to the soundhole.

I've always felt that experimenting is how to figure out how to use your equipment in your space on your guitar/whatever [playing]. Maybe you can get a better sound with a different mic placement, or maybe the HP switch will actually improve things. You could try it on one mic or the other or both. *Anything* that means I don't have to futz with EQ at mix time just to get the track sounding right (vs. mix decisions) is a better place to start, at least for me.

[TBman]

This was just being discussed in one of my threads recently and I was advised
to simply select 80 and below and cut it using the keyboard. I have Rx 7 by Izotope that allows sideways (lengthwise) selection of the track. I can listen to the selected section as well before I press the delete button to adjust if necessary.

[Doug Young]

I see little reason to EQ with the mic's high pass filter. It's nearly always best, when recording yourself, to defer decisions until you can listen to the track - hard to make decisions when you are engineer and performer.

I try to dial in the best sound by mic placement, which generally takes trial and error, again, since you're engineering while you play. If you want to add the high pass filter into the equation at that point, that's fine, of course, but it adds another variable to figure out. Of course the beauty of recording at home, is that the clock's not running, so you can take your time and try every combination, until you're happy.

At mix time, you do whatever you need to do - I nearly always apply a sharp high pass filter, just to remove frequencies below the guitars range (like say 50Hz and below). Then you treat the guitar however you need to - ideally needing very little EQ, but in the end, all that matters is how it sounds. For a boomy guitar, a Dynamic EQ can be helpful. It is pretty tough to EQ out a bad room tho.

XY is a fine option, but I think the idea of "phase issues" is somewhat exaggerated on the internet. Spaced pairs, which are often used for solo guitar, at least, has phase differences (not "issues") as a core part of how it works - how it creates the stereo sound. it all depends on how you plan to use the track. Guitar in a mix with other instruments? Solo? Plan to mix or play back in mono? My suggestion would be to find some reference tracks you would like to match the sound of, see if you can deduce how they were recorded, and/or try to match that sound. Could be mono, could be stereo XY, spaced pairs, etc. Each tends to have a specific sound.

[b1j]

Thanks for your thoughtful reply, Doug. It’s more encouragement to “get it right at the source,” as Joe Gilder preaches.

It may not be clear from my post, but I already have the tracks without the mics’ filters in use. To my ears I think the XY at that distance produced a nice, full sound, with any room effect a couple dozen dB below the guitar. I can now go to work gently on the EQ; it shouldn’t need much and the suggestions here seem right.

To your questions, this will be just “one” guitar behind female vocal trio, so I’m not troubled by a bold, very present guitar tone, as I might be if it had to play nicely with others in the mix.

I say “one” in quotes because I used a little feat I posted about a couple of weeks ago. This is a cover. The part is a fast, precise fingerpicking pattern that cannot actually be played by two hands and one guitar (maybe by Leo Kottke, but not by me). I noticed on the original they have the three low notes of the pattern on the left and the three high notes on the right, clearly played on two guitars! So that’s what I did.

Then, to make my life difficult I recorded each part into two mono tracks instead of a stereo track (for four tracks total), thinking that would give me more flexibility with panning into the stereo field and perhaps with selective processing: EQ, compression, reverb, who knows? (I do hope I’ll be happy enough with the result that I’m not tempted to go nuts with plug-ins. That’s why I took care with the setup and so far am pleased enough. I’ll see if I can avoid compression altogether.)

So now I have the low part panned hard L and hard R, and the same with the high part. The effect is nice and wide, and you don’t notice that it’s two guitars the way you do with the panning on the original. The L tracks both came from the bridge and the Rs came from the 12th fret. I suppose a more realistic placement might be 40% L and R, as it might sound from 5 or 6 feet from the guitarist, but I don’t think I mind the exaggeration. Wide is good. We’ll see after playing it in the car, etc.

If you can stand another side note, if the volume and tone balance is good I might use stereo tracks in the future with the XY setup. That would certainly speed up editing!

Yet another side note. I was idly playing around with phase on the two paired tracks. When I shifted the phase on one of the low tracks, the boom effect essentially went away. I can’t explain that. I’m tempted to leave it this way, but first I’ll double check that I haven’t made the track too thin with the phase shift. I suspect I might have recorded some of the takes with the guitar top not an even 45° from both mics. I might have introduced an extra couple of inches to one mic versus the other. Two inches might mean 12” and 14” from the two mics, which might be enough to create an out-of-phase result.

If you’re still awake, I’ll close here to say that I have settled on not relying on a mic high pass switch. Instead, I’ll adjust the mic placement, preamp gain, and playing intensity to give a rich, full tone with ample headroom. I’ll work on the high pass in the DAW.

It’s nice to have made some useful decisions. This is fun.

[Doug Young]

Sounds like you're on a good path. For that kind of recording, I'd probably do the guitar in stereo, the vocal in mono in the middle, which it sounds like you're doing.

The dual mono (using two tracks) vs 1 stereo track for a stereo guitar is just a workflow choice. Depends on how you like to work, your DAW, and what plugins you use. For me, I nearly always record to a stereo track (for solo or featured guitar). It lets me treat the guitar as a single entity. For example, putting a compressor on the track will equally compress each side, producing a more stable image, without having to side-chain them.

Editing is easier, as you mentioned, tho there are ways around that, depending on your DAW. Logic , which I use, lets me link tracks, so that editing one performs the same edits on any linked tracks. I use this when I use more than 2 mics, but for a simple stereo recording, it's just more steps to achieve the same goal.

If you want to pan somehow, there are plugins that can change the stereo width of a stereo track, and I find it easier to just think about the track width than to think in terms of panning two tracks for this purpose. Some people like to EQ each side differently, but again, there are plugins that make it easy todo that, and it sort of feels nicer to me to be able to see the EQ in a unified view.

But it all depends on personal choice and workflow. A lot of people who grew up with tape, where there wasn't such a thing as a stereo track, still feel better about using separate tracks, and that's fine. Whatever works.

[b1j]

Good points about workflow. I think the hard part, editing and comping two pairs of takes, is over. My plan now is to fine-tune a good volume balance among my four tracks, and then send them to one guitar bus and do all processing in the bus. Since the low and high portions of the guitar part are cleanly separated by frequency, it will be like, say, EQing a single guitar.

This is the first time my guitar recording hasn’t come out thin, or with some kind of comb filtering in the room, or too shrill. I think close mic’ing has a lot to do with it. Also, I wonder if the SDCs are helping with the guitar/room ratio, compared to the LDCs I was using. Whatever, I’m feeling good about being able to record my guitar so it sounds rich and full.

[rick-slo]

For what I record I do not switch on the high pass filter on my mikes. However in some recording situations using the mike high pass
filter may make the mike preamp happier. On solo acoustic guitar in the DAW I generally taper off the low frequencies starting around
50 to 60 hertz with a 12 Q.

[Sasquatchian]

I don't know, but with my Gibson Southern Jumbo, which is essentially a J-45, and boomy as well, the boom is really happening at about 200hz and the most effective method for me is to use the EQ built into any of the UAD Unison pre-amps that I use, my favorite being the Neve 88SR Unison which has a fairly robust EQ section where you can target both the frequency and the Q to quite effectively tame the boom. That pre-amp plugin also has a nice compression section and fully adjustable high cut and low cut filters as well and that beautiful tube emulation as well. That doesn't mean you need no EQ at all later but certainly a lot less and whatever EQ used later is applied given the context of how that guitar sounds with everything else being recorded. Even my 000-18, which is not supposed to boom that much, still does and needs a bit less of exactly the same treatment.

I use those same EQ controls when recording my gf's accordions, often boosting the bass or slightly cutting around 500hz on specific accordions or reed blocks that can get a bit chimey. I'll also use the Peterson tuning app on my phone in hz mode to pinpoint either guitar or accordion frequencies that need attention.

[b1j]

I was thinking about that, too: the boom is pretty audible, so I expect it might show up well above 80 Hz. My gear is considerably less sophisticated than some, but Studio One’s graphic EQ clearly pinpoints the trouble frequencies and gives fully discrete control over adjustments in central frequency, dB cut or gain, and Q — with several bands available across the spectrum.

In the past I have hastened to EQ every track before I have the right balance! I’m committed to stop doing that. For this song, I have been waiting to put the EQ on it until I have the balancing done. Nearly there.