Acoustic Celtic harp... I've been trying to record the same song for two months!

K

Kcater

New member
Thank you for adding me to this forum!
I'm hoping someone might pinpoint something or things I've got wrong.
The biggest issue is I'm capturing so much clicky clacky finger noise, and the actual harp sounds dull, fuzzy -- just not clear and resonant. It's hard to listen to.
The harp is strung with fluorocarbon, which is bright and resonant and highly sensitive.
Mics are matched Rode NT5. I've tried AB, XY and ORTF in every conceivable position: high, low, side, front, back, up, down, near, far....
Scarlett 2i2
Reaper
With the headphones I have been using, I did produce one track I thought sounded pretty good, but when I played it on other speakers, the sound was hollowed out and tinny. Played on my cell phone, which I know is going to be bad, it was unlistenable. People thought it was a harpsichord. Whatever I make, it's got to be somewhat OK on a cell phone, since that's what most people are using now!
I've also tried several rooms in my house.
In Reaper I apply very mild compression, gently EQ the bass for mud, add a reverb file that's pretty nice, and I've tried to EQ the clicking, but it's all over the spectrum. When I tried to use Reafir to cut out the clicking, it made other problems. I wasn't convinced that a noise reduction tool was good for the issue.
Is there anything that's obvious to you? I welcome your thoughts, and thank you!
 
I suppose the first question has to be what the harp sounds like sitting there, in the room? A sample of the problem would help no end, but all those mic techniques capture space - usually filled with multiple distinct sound sources. AB, close in is simply two mics - like people use on guitars. NOT remotely stereo, but usually two distinct tones you blend together. Over such short distances, the time delays are tiny, so they are not really any of the techniques you mention. They're always just two mics with different sounds, rarely stereo in any conventional sense. Until you get to the size of vibes do they sort of morph into stereo with real locational info. The Rode mics are a nice stereo pair, but they are a bit tizzy in close, so it may just bethey don't flatter the harp, revealing all those annoyances in considerable detail. EQ rarely removes this kind of thing without spoiling the sound.

Is there somewhere in the room that the harp sounds nice to your ear? If so, this would become my starting point for ONE mic and then use your nice reverb plug in. This sounds weird, but have you tried any different mics to rule out a mutual mismatch. I once recorded on in Ireland using an SM57 - the only spare mic I had and it worked really nice, that warm old fashioned sound, and I don't remember much percussive noise from the fingers? Might be worth trying?
 
I always found Harps to be simple to record - following basic tenets -

The most suitable place for capturing the sound of a harp is an area with minimal but natural reverberation - this often isn’t a house -
if in a House you are going to need Gobos or the equivalent to block reverberation.

Curtail as much of the undesired reverberations or echoes that might be captured on the recording as possible.

The mic choice is critical - I use two Ribbon Mics (MXL R40s Believe it or not) in a very loose X/Y configuration - they are Figure 8 patterns and work well - and then a Gauge ECM-87 (Cardioid Setting) around the top of the harp's pillar - with the microphone pointing down towards the sound board - they are sent to individual inputs - and I choose the blend that sounds best to me.

Before recording, take the time to tune your harp.

Think of the Harp as a vertical piano sound board.
 
rob aylestone said:
I suppose the first question has to be what the harp sounds like sitting there, in the room? A sample of the problem would help no end, but all those mic techniques capture space - usually filled with multiple distinct sound sources. AB, close in is simply two mics - like people use on guitars. NOT remotely stereo, but usually two distinct tones you blend together. Over such short distances, the time delays are tiny, so they are not really any of the techniques you mention. They're always just two mics with different sounds, rarely stereo in any conventional sense. Until you get to the size of vibes do they sort of morph into stereo with real locational info. The Rode mics are a nice stereo pair, but they are a bit tizzy in close, so it may just bethey don't flatter the harp, revealing all those annoyances in considerable detail. EQ rarely removes this kind of thing without spoiling the sound.

Is there somewhere in the room that the harp sounds nice to your ear? If so, this would become my starting point for ONE mic and then use your nice reverb plug in. This sounds weird, but have you tried any different mics to rule out a mutual mismatch. I once recorded on in Ireland using an SM57 - the only spare mic I had and it worked really nice, that warm old fashioned sound, and I don't remember much percussive noise from the fingers? Might be worth trying?
Click to expand...
Thank you for all your thoughts! Here's the first edit I got all the way through and was ready to share on Youtube, and then discovered how bad it was on anything but my headphones! On anything else, it's tinny.
I've been trying to find a subsequent test to share -- I've put so many hours into trying to find a good sound. They all have issues. This one is mediocre, kind of dull, kind of clicky, volume seems to wobble. (Tried to attach, must be too large. I'll go save it in another format and try again.)
 
Kcater said:
Thank you for adding me to this forum!
I'm hoping someone might pinpoint something or things I've got wrong.
The biggest issue is I'm capturing so much clicky clacky finger noise, and the actual harp sounds dull, fuzzy -- just not clear and resonant. It's hard to listen to.
The harp is strung with fluorocarbon, which is bright and resonant and highly sensitive.
Mics are matched Rode NT5. I've tried AB, XY and ORTF in every conceivable position: high, low, side, front, back, up, down, near, far....
Scarlett 2i2
Reaper
With the headphones I have been using, I did produce one track I thought sounded pretty good, but when I played it on other speakers, the sound was hollowed out and tinny. Played on my cell phone, which I know is going to be bad, it was unlistenable. People thought it was a harpsichord. Whatever I make, it's got to be somewhat OK on a cell phone, since that's what most people are using now!
I've also tried several rooms in my house.
In Reaper I apply very mild compression, gently EQ the bass for mud, add a reverb file that's pretty nice, and I've tried to EQ the clicking, but it's all over the spectrum. When I tried to use Reafir to cut out the clicking, it made other problems. I wasn't convinced that a noise reduction tool was good for the issue.
Is there anything that's obvious to you? I welcome your thoughts, and thank you!
Click to expand...
OK, here's the raw mp3 test. I think the mics were about 42" away in AB for this one. In the edited recording on Youtube, they were ORTF in front of the harp, about 2 feet from the column.
 

Attachments

I think we are hearing the hole in the middle. Both mics missing the nicest area of the harp. four feet away probably means the inverse square law reduces the furthest strings too much, too quickly. Assuming they’re symmetrical, you create a blending f two mics that ‘blur’ the crossover point, which robs the detail a bit and gives a typical mic at a distance sound. If you had use a stereo coincident technique, any of them, ai think, that would have been an issue too. In the coincident modes neither mic would actually be hearing the harp properly either! If you imagine the cardioid pattern polar diagram, the 90-120 degree angles in all the usual techniques mean at best 50% of each mic’s capture area is empty space.

Too much room, too little instrument. In the piece played, it is very centre heavy in terms of where its being played, i.e. the finger contact is rarely at the top or bottom strings, but in the middle area where the mics are the weakest with any of these stereo configurations. My way of working with any new instrument i have not recorded before is to put out one mic and record a little bit and see what it sounds like. I have a personal rule too. Any instrument less wide than my monitor speakers will jot response well to any notion of ‘stereo’ because in a room my ears detect hardly any width at all. I always look at the whole area used by voices or instruments. If from the centre, they are within 70 degrees or so, side to side from the best mic distance, then I will use a stereo technique. If they are spread out less than that, my experience of stereo techniques is that they don't create realism to any degree. Pianos are a great example. Fair enough, they’re wide, but if you stand six feet from a six feet wide grand and close your eyes, the width is minimal. The bottom and top notes coalesce in the middle, not left and right. Orchestral harps have virtually no width, as I said, vibes and xyzlophones, marimbas and other percussive instruments that have multiple resonators do, so I consider them as a wide instrument. Others are point sources. Small harps have virtually no width, because the sound comes from the whole thing. I would record a harp in a church using a stereo technique. Any of the ones you mentioned, or maybe even something exotic would be great to capture the instrument in a wonderful space. In the studio, I would get closer with multiple mics, and blend them afterwards.

The video with the reverb added is quite usable I think but the raw file just says nice mics in the wrong place. Sorry this went on quite a bit but I have been getting live recordings in nice or nasty places wrong for a very long time, and getting it wrong wasted so much time. I now have a ‘thing’ about when stereo techniques are appropriate, and when they are not. Sorry.
 
Hi Kathryn, I am not anywhere near Rob's league but I have to say the clips sound pretty much like every harp I have ever heard reproduced in the last 70 years! But then I am clinically deaf! I am glad you used an MP3 for a clip, much easier for me at least. If you start with a 24 bit/44.1kHz .wav and code it to 320k MP3 you don't lose much quality.

I did not detect any "clicking" but if I can be a bit cheeky, even slightly rude? Clicks and squeaks in the classical guitar world are often down to poor technique (son plays classical G) I have NO idea about harps though. Another crumb of possible help from the guitar world? If, as was suggested you remove as much of the (bad) room reverberation you can end up with a dull sound. A trick the gitists use is to place a large sheet of reflective material around the player. Ply or hardboard would work.

Dave.
 
ecc83 said:
Hi Kathryn, I am not anywhere near Rob's league but I have to say the clips sound pretty much like every harp I have ever heard reproduced in the last 70 years! But then I am clinically deaf! I am glad you used an MP3 for a clip, much easier for me at least. If you start with a 24 bit/44.1kHz .wav and code it to 320k MP3 you don't lose much quality.

I did not detect any "clicking" but if I can be a bit cheeky, even slightly rude? Clicks and squeaks in the classical guitar world are often down to poor technique (son plays classical G) I have NO idea about harps though. Another crumb of possible help from the guitar world? If, as was suggested you remove as much of the (bad) room reverberation you can end up with a dull sound. A trick the gitists use is to place a large sheet of reflective material around the player. Ply or hardboard would work.

Dave.
Click to expand...
Hi Dave, this particular composition uses a technique harpists call bisbigliando, (like a tremelo), but causes my fingers to constantly be returning to reverberating strings, and with the incredibly sensitive fluorocarbon strings I'm going to get finger noise, which my recordings seems to amplify. An all out buzz is a mistake and I can re-record. I did try recording in a small reflective space (the nook for my glass french doors) and it was interesting, but I gave up on it. The harp sounded very loud, but was it better? My ears are exhausted :)
 
rob aylestone said:
I think we are hearing the hole in the middle. Both mics missing the nicest area of the harp. four feet away probably means the inverse square law reduces the furthest strings too much, too quickly. Assuming they’re symmetrical, you create a blending f two mics that ‘blur’ the crossover point, which robs the detail a bit and gives a typical mic at a distance sound. If you had use a stereo coincident technique, any of them, ai think, that would have been an issue too. In the coincident modes neither mic would actually be hearing the harp properly either! If you imagine the cardioid pattern polar diagram, the 90-120 degree angles in all the usual techniques mean at best 50% of each mic’s capture area is empty space.

Too much room, too little instrument. In the piece played, it is very centre heavy in terms of where its being played, i.e. the finger contact is rarely at the top or bottom strings, but in the middle area where the mics are the weakest with any of these stereo configurations. My way of working with any new instrument i have not recorded before is to put out one mic and record a little bit and see what it sounds like. I have a personal rule too. Any instrument less wide than my monitor speakers will jot response well to any notion of ‘stereo’ because in a room my ears detect hardly any width at all. I always look at the whole area used by voices or instruments. If from the centre, they are within 70 degrees or so, side to side from the best mic distance, then I will use a stereo technique. If they are spread out less than that, my experience of stereo techniques is that they don't create realism to any degree. Pianos are a great example. Fair enough, they’re wide, but if you stand six feet from a six feet wide grand and close your eyes, the width is minimal. The bottom and top notes coalesce in the middle, not left and right. Orchestral harps have virtually no width, as I said, vibes and xyzlophones, marimbas and other percussive instruments that have multiple resonators do, so I consider them as a wide instrument. Others are point sources. Small harps have virtually no width, because the sound comes from the whole thing. I would record a harp in a church using a stereo technique. Any of the ones you mentioned, or maybe even something exotic would be great to capture the instrument in a wonderful space. In the studio, I would get closer with multiple mics, and blend them afterwards.

The video with the reverb added is quite usable I think but the raw file just says nice mics in the wrong place. Sorry this went on quite a bit but I have been getting live recordings in nice or nasty places wrong for a very long time, and getting it wrong wasted so much time. I now have a ‘thing’ about when stereo techniques are appropriate, and when they are not. Sorry.
Click to expand...
Wow, thank you, this will take me some time to work through, as my comprehension of most terms is fuzzy. What I'm sensing you are telling me, is putting these two mics on a bar won't work in any configuration, but I'm not sure where that leaves me. The Scarlet 2i2 can only handle two mics, so do I put one close and one far, and pan both to the middle? This is basically my level of understanding.
The closer I get the mics, the louder the finger noise is. The further back I go, the duller and less present. I'll find a clip I did with the mics basically next to my head, hoping to capture what I hear when I play. It is interesting, but it picks up my breathing.
I have recorded many times in my flutist's studio, so I'm trying to duplicate what the engineer (her husband, who has crazy skills) did in their studio music room -- a room roughly 20x20 carpeted. He had professional equipment of course. He liked the ORTF, so I gave it a try. and xy, and ab.
 
Forget calling the pairing of 2 mics on a bar X/Y or ORTF, or DIN or the others - stereo, because close in, they're being used totally differently that the usual stereo application. The way these techniques work is to do with a couple of two main features. Differences in volume from one mic to the other, and differences in time between the two. At 300Ms - close miking with a modest distance between produces small delays that don't let the stereo system work.

You've discovered the finger noises - if you think in reverse, if you are not recording these, then the mics are not capturing sound from the finger area, which is where the most realistic tone is. What you could try is to focus one mic fairly close in (with the noises) and another perhaps on the other side a little further away. You can then blend the two together. You might need to swap the polarity of one as they will probably fight with each other and sound a bit hollow or phasey. You then get a realistic tone with finger noises and a thinner one with less noise - you can do mild EQ on both, and perhaps just a tiny bit of panning - not much, just a tad one way from centre on one and a similar amount on the other. Unless the room has an acoustic you like, then some judicious reverb to set the 'space' - and this will sound more stereo than the use of a coincident or spaced pair. Look at your ORTF pair - 110 degrees means you have the hole in the middle - so it's probably the worst to use on a smallish instrument.

As the player - remember your impression of a good recording is always modified by your perception of what your instruments sounds like. The trouble is NOBODY apart from you hears what you hear. Trumpets (and other brass) are often recorded really badly by trumpeters, because they never hear what an audience hear. Classical guitarists are often the same. They hear the perspective from their playing position, never from the audiences.

I guess what might be happening is you hear the finger noise, but is that because it's intrusive and wrong, or maybe it's really what your instrument and playing sounds like to others? Do you have a trusted musician friend? Maybe the flautist? What do they think? They will have been in the space with you and can comment critically on how close the recorded sound is to what they heard?

The strongest sound I can hear that you might really hate seems to be the thumb noise - that sort of percussive low frequency tone/thump noise that is tonally different to the fingers. The noise you hate - is that the thumb ones. NT5's can be a bit prone to sounding a bit thin, especially off-axis. If you can arrange it - try using one and getting somebody to hand hold it while you play, and they move the mic around - focussing on different areas of the harp from different distances and aimed at the same spot from different angles. Then listen carefully with the other person and see if you get concensus on positions that work better than others.

Does this stuff help? I don't want to just bombard you with stuff that is missing the point or confusing, so I'm happy for you to say whoa - stop.
 
Kcater said:
Hi Dave, this particular composition uses a technique harpists call bisbigliando, (like a tremelo), but causes my fingers to constantly be returning to reverberating strings, and with the incredibly sensitive fluorocarbon strings I'm going to get finger noise, which my recordings seems to amplify. An all out buzz is a mistake and I can re-record. I did try recording in a small reflective space (the nook for my glass french doors) and it was interesting, but I gave up on it. The harp sounded very loud, but was it better? My ears are exhausted :)
Click to expand...
Ah, I wasn't clear. When you have a room well 'lagged' so as to kill virtually all reverberation the resultant dull sound can often be improved by sitting on a few sq mtrs of hardboard.

But I can see you are in good hands.

Dave.
 
rob aylestone said:
Forget calling the pairing of 2 mics on a bar X/Y or ORTF, or DIN or the others - stereo, because close in, they're being used totally differently that the usual stereo application. The way these techniques work is to do with a couple of two main features. Differences in volume from one mic to the other, and differences in time between the two. At 300Ms - close miking with a modest distance between produces small delays that don't let the stereo system work.

You've discovered the finger noises - if you think in reverse, if you are not recording these, then the mics are not capturing sound from the finger area, which is where the most realistic tone is. What you could try is to focus one mic fairly close in (with the noises) and another perhaps on the other side a little further away. You can then blend the two together. You might need to swap the polarity of one as they will probably fight with each other and sound a bit hollow or phasey. You then get a realistic tone with finger noises and a thinner one with less noise - you can do mild EQ on both, and perhaps just a tiny bit of panning - not much, just a tad one way from centre on one and a similar amount on the other. Unless the room has an acoustic you like, then some judicious reverb to set the 'space' - and this will sound more stereo than the use of a coincident or spaced pair. Look at your ORTF pair - 110 degrees means you have the hole in the middle - so it's probably the worst to use on a smallish instrument.

As the player - remember your impression of a good recording is always modified by your perception of what your instruments sounds like. The trouble is NOBODY apart from you hears what you hear. Trumpets (and other brass) are often recorded really badly by trumpeters, because they never hear what an audience hear. Classical guitarists are often the same. They hear the perspective from their playing position, never from the audiences.

I guess what might be happening is you hear the finger noise, but is that because it's intrusive and wrong, or maybe it's really what your instrument and playing sounds like to others? Do you have a trusted musician friend? Maybe the flautist? What do they think? They will have been in the space with you and can comment critically on how close the recorded sound is to what they heard?

The strongest sound I can hear that you might really hate seems to be the thumb noise - that sort of percussive low frequency tone/thump noise that is tonally different to the fingers. The noise you hate - is that the thumb ones. NT5's can be a bit prone to sounding a bit thin, especially off-axis. If you can arrange it - try using one and getting somebody to hand hold it while you play, and they move the mic around - focussing on different areas of the harp from different distances and aimed at the same spot from different angles. Then listen carefully with the other person and see if you get concensus on positions that work better than others.

Does this stuff help? I don't want to just bombard you with stuff that is missing the point or confusing, so I'm happy for you to say whoa - stop.
Click to expand...
Yes, actually that is great information, and I appreciate your generosity so much! Thumbs can be an issue for sure. You are absolutely right that I'm only trying to record what I hear when I play. I think the harp sounds amazing in every room of this house! So much natural reverb, warmth and clarity. I'm going to try to work with your suggestions this week. I wish I still had another musician in my life to help me hear, but this board is giving me some much needed points to work on.
Panning is such a subject. I'm grateful you gave me permission to stray from 100% R/L.
 
Ha you don’t need permission. Recording has rules that everyone learns are really guidelines, that breaking once they fail is perfectly. Now i am ancient, my personal warning bells ring whenever you get asked to record solo violins and violas, any brass, and to a lesser degree, small woodwind. None of these players know what their own playing sounds like to others, and often hate it! In my career, i have discovered many people do not come back for a second recording, until they had had another by somebody else they hate more! My trick now is to play them a little of a past recording of somebody else, and ask what they like about it as a comparison. This seems to explain away what they are hearing on the recording, vs what they normally hear. “Ah, my instrument has much greater clarity, presence, detail” etc etc. you can drop in early why they think this from the playing position perspective. It sort of manages expectations.
EDIT
The harp video above shows very clearly the concert harp, and apart from sound, we have always provided lighting. Years back i lit an orchestra in a very large church, and one piece I lit in gorgeous red light. In the video, can you see the red strings? They use these to locate their hands on the correct strings. Not in red light they don’t. Every string was red. They could not play. Since then, no saturated reds for harpists!
 
Last edited:
You must log in or register to reply here.
Back
Top