Advice for recording a small folk style ensemble

Discussion in 'Location Recording' started by felixcat, Jan 9, 2005.

  1. felixcat

    felixcat Guest

    Hi All,

    Firstly, just want to say that there are some great posts going here. I've mainly been an engineer for rock/pop over the years, yet have always appreciated classical music, and I'm looking at using some classical recording techniques to record a few tracks of a group I'm currently working with.

    The situation is core group of two musicians (both play acoustic guitar and vocals) with other members added as the song requires. 10 tracks to the album, 5 of which are your standard rock band affair. The other five have quite an intimate feel, almost folk in style. The usual stick the mic 2 in from the source approach is not working, and I'm losing all the intensity of performance when recording/overdubbing them.
    Instrumentation: 1. Acoustic G, Flute, Congas. 2. Didgeridoo, tapping sticks, Acoustic G. 3. Acoustic G, Violin, Piano.

    Here's my thoughts on a new approach. I've got a lovely sounding church just up the street that will let us record there. I'm thinking about a stereo array from the "listening position" (probably front row of the pews) and spot micing over each instrument at a distance of about 1 metre (40 in?). In mixing, delay the spot mics so they arrive at the same time at the stereo array and use the spot mics for high lighting as required.

    Mics available:

    Avenson STO2 x 2, KM84 x 1, C1000 x 2, TLM 103 x 2

    Any advice to my options with mics (and whether this may well work) would be aprreciated.

    Thanks,
    John.

    PS. Just spent three great nights at the Philip Glass Qatsi Trilogy here in Sydney, with the premiere of Naqoyqatsi heading the concert series. Breathtaking!
     
  2. FifthCircle

    FifthCircle Well-Known Member

    Sorry for the delay in answering here- I've been extremely busy dealing with some other issues lately...

    Anyways, I wanted to make some suggestions...

    I see a couple ways that you can go about making these recordings. The first is the traditional multi-mic setup where you have a stereo pair out in front and mic each instrument individually. Similarly, you may find if the room is less than perfect that the stereo pair creates more problems than it solves or the room may be great and a stereo pair captures the goup perfectly and the spots just don't work... I don't know the Avenson microphones. I'd stay away from the C1000's as they are pretty bright mics.

    With one of my colleagues, I've been experimenting with an "older-school" way of micing small acoustic groups. One of the recordings he did of a folk ensemble has to be one of the best recording of any type I have ever heard. What it involves is using a single stereo ribbon mic (in this case a Royer SF-24) and placing all the musicians around it. When doing a recording this way, you have to move the musicians around until they sound right, but the natural sound is something you cannot achieve through multiple microphone recording.

    The blumlein pair is the secret to the setup. Each lobe of your figure-8 is rather well isolated from the other ones. What you are basically getting is 4 microphones with perfect phase compatability. One side is obviously out of polarity, but I have yet to find that that really matters in the end. Remember that the sides are reversed on the back so two players that are facing each other on the same "side" of the microphone will sound like that are coming from opposite sides of the stereo spectrum.

    I would set things up like this: 1- AG and Flute on the positive side of the mic standing next to each other for a nice L-R spread. Congas centered on the back side of the mic. Your result in the stereo image- guitar -> Congas -> Flute. 2- Dig and guitar set up like #1. Tapping sticks in the center of the rear lobe. Depending on the song, you may find that the Dig and tapping sticks may sound best reversed so the Dig is in the center of your image and the sticks are on the side. 3. Piano on the rear lobe, Guit and Violin on the front...

    The secret to making sure it works is that the players cannot see the opposite side of the microphone when they are playing. Otherwise you'll end up with some very strange phase and image problems.

    --Ben
     
  3. DavidSpearritt

    DavidSpearritt Well-Known Member

    I second Ben's excellent recommendation. I have done a Barbershop quartet with a Blumlein array, but they were all on one side and the result was sheer magic.

    Also we recorded the Ten Tenors, with five on one side of the Blumlein array (2 KM120's) and five on the other, facing each other. In each group of five, 2 were in the front row and 3 right behind them.

    This technique gives the best blend of vocals I have ever heard.

    Again Jan Erik Persson (Opus3) uses the same technique on small instrumental ensembles, as has Bob Katz, even utilising the out of phase quadrants for added reverb effects where coherance is not important.

    All demonstrates how wonderful Blumlein and two good 8's can be. Spine tingling results can be had.
     
  4. FifthCircle

    FifthCircle Well-Known Member

    Here is an example of what I was talking about- I mentioned it in the Blumlein/AKG thread but it should be mentioned here as well-

    http://www.royerlabs.com/wailin_jennys.html

    My colleague (and one of my mentors) made this recording recently in the style that is exactly as I described.

    --Ben
     
  5. Simmosonic

    Simmosonic Active Member

    I agree with everything that Ben and David have said here regarding Blumlein, especially if the room is good.

    I have made numerous direct-to-stereo recordings using both sides of a Blumlein pair, it's a good technique when you get everything right.

    I haven't heard the one on Royer's site that Ben suggested, but...

    The best I've heard so far was done by my engineering buddy Glenn Santry using *my* Royer SF24 (dammit!), and can be heard on the latest Karl Broadie album. Two acoustic guitars, vocals, and double bass. Straight to stereo, sounds beautiful. But Glenn's studio is set up in a church hall, so he has plenty of good sounding space to play with, and he has become very good at strategically placing absorption around the musicians to control their projection and the aspects of the room sound he wants.

    - Greg Simmons
     
  6. Simmosonic

    Simmosonic Active Member


    Hi John,

    Despite the glowing fondness for the Blumlein technique exhibited in this thread by myself and others, it occurred to me that it’s not possible with the microphones you’ve got, so probably not a lot of use to you…

    Due to your microphone options and the possibility that you will have limited time in the church and therefore want to get things as right as possible from the beginning, consider the following:

    1) Get the musicians to rehearse as an acoustic ensemble before going into the church. Experiment with their physical positions to get the best balance you can when standing in front of them, and then rehearse some more until they are totally comfortable playing in that configuration. Without headphones or PA or anything like that. It is important to make sure they are balanced, but also, they must be able to hear each other properly in order to play well.

    2) Set up the two TLM103s in ORTF (Mr Spearritt is going to hate me for suggesting that, but anyway). Set them up so there is 17cm distance between the capsule centres, and they are angled outwards at about 110 degrees (55 degrees from centre, each). ORTF is a very forgiving stereo technique. It’s not as amazing sounding as some of the other techniques, but a) you can do it with your existing mics, and b) it is not too hard to get a reasonable result. I would hate to hear that you hired some bi-directionals to do Blumlein and then found that the church was too live to make it work. ORTF does not provide the pin-point imaging of a coincident technique like Blumlein, but in exchange it provides a nice spatial sound (due to the arrival time differences between the microphones) that is unique to spaced microphones. The TLM103 wouldn't be my first choice for this job (probably a pair of DPA4011s or similar), but if it's what you've got...

    3) Project a line from the centre of the ensemble at about 45 degrees upwards, and place the ORTF pair on axis with this line, at the point where it is equidistant from all the instruments. This will probably be a good starting point, and you can then work on it from there – higher, lower, closer, further, to get the desired sound. (You ARE using a stereo bar, aren't you? It will be a nightmare of 'screwdriver work' without one.) Listen carefully to what it is giving you, and react in accordance. Closer will give you a wider stereo image and greater detail, but at the expense of less room sound and possibly losing the sense of ensemble that often makes a great recording. (If you are able, practice doing this in your studio or where the musicians are rehearsing in step 1, you will learn a lot after one or two experiments like this, and that experience will be invaluable when you get to the church. As a bonus, you might also capture some magic performances because they’re more relaxed in rehearsal…)

    4) Experiment with microphone positioning and instrument positioning until you get what you think is right. You will probably be monitoring on headphones while in the church, so make sure you are familiar with how those headphones present the direct/reverberant balance relative to speakers. Some headphones exaggerate the reverberation, others underplay it. The same applies to how they represent the stereo image…

    5) Always remember the words of the great Sherman Keene: “Air is the best mixer of multiple sounds”, and the words of the great Mike Stavrou: “Start rough, work smooth”. So, if it’s not sounding right, pull back a bit further to let the air do what it does best, and take another listen. If nothing makes sense, just get out there and move the microphone position dramatically somewhere else, anywhere else! Eventually you will find a position from which you can get your bearings, and then make fine adjustments. The more different positions you try, the more data points you have to work with, and the sooner you can triangulate (sorry, poor term but I use it all the time) the best position.

    You may also try blending in your Avenson omnis in a widely spaced pair, but I’m cautious of using omnis in any church in Sydney (you are in Sydney, yes?) because there are very few genuinely quiet churches in Sydney. What’s ‘quiet’ to our ears is rarely as ‘quiet’ to a microphone! Because of this, cardioids often provide a much more workable end result, and you can always add a bit of reverberation in mastering if it’s needed. Considering that these tracks will be sitting alongside rock songs, it hardly needs to be ultra-purist to the extent that there was no reverb added in mastering! (Another thing to consider is that cardioids have a natural LF roll-off, which often helps alleviate the LF traffic rumble that is inherent in most churches in Sydney and will be picked up beautifully by omnis.)

    Also, most importantly, check out the construction of the church itself. If it has a barrel-vaulted ceiling (a large curved or arched ceiling) I would say immediately to forget Blumlein. I know this sounds like a huge generalization, but I have recorded in numerous churches with such ceilings in Sydney, and Blumlein is rarely satisfying. Way too much reverberation, and if you move closer to the instruments, you are often too close. It is my favourite technique, and I have tried it (and other coincident techniques) over and over again in such churches, but I always end up getting a better result from a near-coincident pair.

    Which church is it? Perhaps I can give you some specific advice...

    Also, make sure there is some nice thick carpet (some churches have an abundance of this, in glorious red!) on the floor between the musicians and the microphones. This will stop first order reflections from the floor, and open up the sound considerably.

    If all this fails and the recording is a disaster, buy yourself The Cowboy Junkies 'Trinity Sessions' and eat your heart out.

    Okay, I hope this is helpful. Now I’m putting on my flak jacket and locking the door to the bunker…

    - Greg Simmons

    P.S. I’m currently traveling around Nepal and Tibet recording all sorts of things and having a great time, but the Philip Glass Qatsi Trilogy almost made me come home! I’m a huge fan of Philip Glass, I’m a huge fan of Philip Glass, I’m a huge fan of, I’m a huge fan of, I’m a huge fan of Philip Glass. I have also seen the Qatsi movies more times than I care to remember. And Baraka. Some years ago I saw the Philip Glass Ensemble performing Koyaanisqati live to the new print of the movie at the Sydney Opera House – awesome!
     
  7. DavidSpearritt

    DavidSpearritt Well-Known Member

    I will not, Greg. I am not a great fan of ORTF but in this case, with this venue and the mics available it is going to be a hell of a lot better than spaced omnis.

    I do not like ORTF, because cardioids are used, read ordinary off axis response (Jeremy I am going to clear up your confusion with this more, in the classical guitar thead), so ordinary sound.

    But its image placement is nonlinear and this upsets me. I recently recorded the Takacs quartet and the Macquarie Trio with ORTF and while the trio was OK because there was only L,C and R musicians, the Takacs musicians were displaced from their real positions. They were also a bit dry and scratchy, had to add SIR/Musikverein and that made it magnificent. I wish I had had my SF24 for that one though.

    No, its a cardioid and a large diaphram to boot, so the diffraction around the cage at HF will exacerbate the already ordinary off axis response. TLM103's are for close micing.

    Great advice to be sure, to be sure.

    Noise is an issue in churches but so is a woofly acoustic and this is what omnis will disturb even more. But I would try them first as a narrow spaced pair to augment the ORTF, not widely spaced.

    Good advice, although Blumlein is our main pr of choice in St John's Cathedral in Brisbane, we supplement it with a small amount of narrow spaced omni pair in front of them. We find blumlein drys up the acoustic in St John's nicely, especially with an audience behind soaking up the sound. We also use MS for the main pair for wide larger sources.

    Really Greg? Have you heard the minimilist piano stuff where the same stupid little ditty is repeated ad infinitum. Give me Ligeti any day.

    Cheers
     
  8. JoeH

    JoeH Well-Known Member

    I went to the Royer website for the "WailinJenni's" MP3 link, and was quite impressed.

    Even moreso to read how they did it:

    <<The recording chain was:
    One SF-24 and one R-122 fed into three channels of VacRack tube mic preamplifiers. The SF-24 was fed into a set of VacRack electro-optical limiters linked for stereo and set for 3 dB of limiting "to be safe." Signals were then fed into line level inputs of a Mackie 1604 VLV Pro mixer, where 1.5 dB of top end on the Mackie's EQ's (somewhere between 10 kHz & 12 kHz) was added and the Mackies 75 Hz high pass filter was also kicked in. A touch of Lexicon reverb was added, then the Mackies analog outputs were fed into the analog inputs (balanced) of an HHB CD recorder.>>

    I'll have to pick up that CD from them, and hear it for real as 16/44 recordings.

    Great stuff, indeed!
     
  9. FifthCircle

    FifthCircle Well-Known Member

    I don't know if that recording is commercially available, actually...

    It was done for a radio broadcast of a show called "Folk Scenes" that is produced here in LA. Broadcast on KPFK (Pacifica Radio)...

    --Ben
     
  10. Simmosonic

    Simmosonic Active Member

    Agreed with all of the above, for all the same reasons. And TLM103s would not be my first choice, or my second, or probably not even my third for this. Maybe not even my 10th, or 20th, or continuing in orders of magnitude. They wouldn't be my LAST choice, however!

    It's the great problem of trying to give good advice that is also achievable with the available equipment. So far, John's got a lot of glowing discussion about the joys of Blumlein, but no solutions to his posed problem of trying to make a recording of an acoustic ensemble in a church with a specific set of microphones. Perhaps there is no budget for hiring more appropriate microphones?

    ORTF is a very forgiving technique, it's easy to get an 'acceptable' result on the day without a lot of previous experience or familiarity, it's achievable with John's microphone selection, and it's a better option then two omnis (IMHO). Although he probably ought to try the omnis as well, because you never know your luck...

    If he took the ORTF with TLM103s approach, I suspect he'll get something acceptable, but it almost certainly won't be a 'finished' result without some help in mastering. I expect it will need some judicious EQ (encode to M and S, EQ the M separately to the S so that the centre sound sources - which are off-axis to both microphones - have a similar perceived spectrum as the side sound sources, tweak the M and S balances appropriately and decode to LR stereo). Then, because these tracks are going to have to stand alongside rock or pop songs, a bit of compression will probably be in order (Compression?!?! Wash my mouse out with soap!) and will help to smooth out any lumps in the balance that could not be fixed by physical musician/instrument placement at the session. And finally, if required, a touch of impulse reverb added after the compression to maintain a more natural sound, i.e. so the reverb itself isn't compressed. It could sound quite good - not necessarily from a purist audiophile point of view, but it will certainly maintain a nice acoustic contrast against the other songs, which is the aim of taking such an approach in the first place.

    I also dislike ORTF's non-linear imaging (angular distortion?), but John is in a situation where he can move the musicians around a bit to put them where he wants them in the stereo image, so it becomes less of an issue.

    I've just had another thought. Perhaps he could try his two TLM103s in XY, at 90 degrees. That won't be going so far off-axis as ORTF, and it requires a 180 degree soundstage to reproduce a hard left to hard right image. He could set the musicians up in a small arc/semi-circle, with the mic in the centre of the radius as a starting point, and work from there. (But if it's a church with a barrel-vaulted ceiling, perhaps avoid having the musicians and mic aligned with the centre of the room, which is where I find the reverberation builds up considerably.)

    I cannot think of anything else with the microphone selection he currently has available. We can discuss the wonders of Blumlein or of DPA 4011s and similar until the 'roos come home, but if he ain't got the microphones then we're not offering any help at all!

    Personally, I find it very frustrating when I start a thread with the aim of getting a useful answer, only to find it gets hijacked in a different direction that doesn't help me at all. I am worried that's where John's thread is going :roll:

    - Greg Simmons
     
  11. Simmosonic

    Simmosonic Active Member

    Sorry, it's bigmouse Simmosonic here again. Just after posting the above message, I had an idea which is probably more of theoretical interest, but anyway.

    Continuing with the theme of trying to find a solution to John's recording using only the microphones he's got available, this came to mind:

    1) Position his TLM103 cardioids in a coincident manner, one above the other, both facing hard left.

    2) Assign them both to the same bus. Reverse the polarity of the lower cardioid and match the gains until the cancellation null is as deep as possible.

    3) Rotate the lower cardioid so it is now facing hard right, remaining coincident with the upper cardioid. Summed together, the two cardioids are now acting as a single bi-directional.

    2) Position his KM84 directly between them, coincident, facing the centre.

    Voila! MS!!!!

    DON'T TRY THIS AT HOME, KIDS... It is bound to end in tears.

    - Greg Simmons
     
  12. FifthCircle

    FifthCircle Well-Known Member

    My, my Greg... You certainly are off your rocker, aren't you...

    Personally, I don't mind ORTF. I much prefer other pickups, but I've gotten plenty of good recordings with an ORTF pickup. I've even used TLM103's with results that my clients loved (I would have preferred other mics, but hey, if the client is happy....)

    One of my favorite mics is a large diaphragm mic- which by its very nature isn't going to be as flat off axis. It is all in what you want...

    I've done some recordings where I've used a spaced omni directional pair for ambience and basic room sound.... Then each instrument is mic'd individually. On my website- Example 6 is exactly that approach. I had my B&K omnis in the room (with the diffuse field grids) as a spaced pair. Each instrument was mic'd individually- the violins and cello with Beyer 160 ribbon mics, the lute and voice with TLM103's, and the harpsichord with a pair of MKH40's.

    In the end, as long as you have a concept of what you want it to sound like, the gear can almost become secondary to the final product.

    --Ben
     
  13. DavidSpearritt

    DavidSpearritt Well-Known Member

    Greg and John,

    Please accept my humble apologies for hijacking this thread and turning it into another soapbox to say how good blumlein is. I was originally confirming the advice Ben gave and for some reason got completely carried away. :oops:

    From what John has in mics and the church he has in mind and the small group, there is absolutely no question that an ORTF pr of the TLM103's with some small amount of omni outriggers thrown in will give the best result.

    I would experiment with the omnis fairly close together, ~ 500-600 mm, in front and higher than the ORTF pair. Listen to both pairs separately first to make sure they have the same distance perspective, then favour the ORTF in the mix, so that the omnis are just detectable.

    Cheers
     
  14. Simmosonic

    Simmosonic Active Member

    John? What say you?

    Maybe he finished the recording yesterday...
     
  15. Simmosonic

    Simmosonic Active Member

    Sorry. I've been taking antibiotics for an ear infection, and I had red meat last night. It always makes me aggressive.

    - Grrrrrrrrreg Simmons
     
  16. Simmosonic

    Simmosonic Active Member

    Okay...


    He's all yours.

    - Greg Simmons
     
  17. DavidSpearritt

    DavidSpearritt Well-Known Member

    So was I, very much so. Downloaded all the samples and relished them. Those girls are great singers, fantastic intonation, ensemble and blend, but it is a great recording alright.

    Speaking of great singers, its sad news that Victoria de los Angeles died on last Friday.
    http://msnbc.msn.com/id/6831172/
    One of the best.
     
  18. DavidSpearritt

    DavidSpearritt Well-Known Member

    Regarding Ligeti, a quote on the Sony website about him ...

    "Ligeti doesn't ever try your patience. Few composers are so attuned to their audience's concentration threshold." The Independent (London), December 1996

    A true quote IMHO and the very antithesis of Phillip Glass.
     
  19. Simmosonic

    Simmosonic Active Member

    Oh my gosh! I hope I catch this before someone else does!!!

    I am absolutely 100% perfectly wrong in the above statement. You can't make a bidirectional from two omnis back to back. You need to do it with two cardioids. To show how it works, here's the maths:

    Cardioid = omni + bidirectional

    If we have two cardioids, coincident but with one facing the other direction, we have:

    Cardioid #1 = omni + bidirectional
    Cardioid #2 = omni + -bidirectional

    If we reverse the polarity of Cardioid #2 and add it to Cardioid #1, we get:

    Result = (omni + bidirectional) + (-omni + bidirectional)

    The omni components cancel out, and the result is a bidirectional.

    Sorry for being misleading!

    John can still make a bidirectional by using his two TLM103s back to back, with the same process I described earlier. Then he could stick his KM84 on top, or between, them, and have MS. PLUS, he'll still have his pair of omnis to use as outriggers.

    I have corrected my previous post for posterities sake.

    Despite all of this, I wouldn't go there in practice. Too messy.

    - Greg Simmons
     
  20. felixcat

    felixcat Guest

    Not quite... Have been in heavy duty sessions over the last week on another project. Thanks to all for the *treatise*. Agreed on the bi-directional from 2 cardiods, v messy. Using the 103's in ORTF with the omni as 'outrigger' seems to be the agreed choice for the mikes available. We've got a bit of time with this one, so we can experiment. Some mike purchasing isn't out of the question, so any suggestions there (aside from the Blumlein, as there is quite a deal of reverb) would be welcome.

    We did get into the church on Thursday daytime and Friday evening for general rehearsal, demo recordings, to listen to how much noise the mikes heard etc. The church is St James Anglican in Turramurra. Located at the end of a small cul-de-sac, so not alot of traffic noise. Being a leafy kind of area, Mr Tree Lopper was in full swing when we arrived Thursday am.

    Greg, yes high vaulted ceilings, marble floor in the choir and altar areas, carpet over floorboards in the pews. The church also has a chapel attached on the left, wooden floor with carpet runs down the aisle. Ceilings alot lower, solid wood paneling scooped upwards into a hexagon shape. There are several of these running the length of the chapel's ceiling. Some interesting reflection from the ceiling in the chapel, but not suitable for our purposes.

    Evening is the best to record, although a storm at about 9.30pm finished our session for us.

    We tracked with congas, flute & acoustic. Congas centre, flute to the left, ag to the right at 45° off the congas, forward about 1 metre. Omni's on a Jecklin disc centre approx 2.5 metres from the congas, height approx 1.5 metres above the carperted area. Spot mikes on each (3 x 103's - 2 of mine, 1 borrowed for the day). Result was ok, not flash. We're their again this week, with the didge player this time, so will try thr ORTF and omni combo then.

    We got some good stuff out of this though. Firstly, finding the spot where the groups sounds the best in the church and allowing the group to rehearse as an acoustic ensemble.

    No offence taken on thread hijack. I'm deeply interested in trying the Blumlein technique. 8)

    Cheers,

    John
     

Share This Page