Nature recording Q&A

Build a home-made parabolic dish

Sun, 25 Oct 2009 16:56:35 -0700

The barrier is made from flexible plastic cut to about 5.5" diameter. The windscreens are standard screens made by Shure Brothers for their SM81 cardoid microphone. I sliced away a little over 1/3 of each windscreen and then glued them to each side of the barrier. Two MKH 20s are placed side by side with a rubber spacer between them at the rear; they are then taped together at their rear ends using electrician tape. They are then inserted into the two windscreens, with each mike facing forward and fairly tight against the barrier, with their elements at the center of the barrier.

As the photographs show, the mike holder is fashioned using standard ABS plastic pipe fittings purchased from a hardware store. I had to enlarge the hole in the Telinga to accept the fittings. The tube that receives the barrier setup is slit at the top and bottom so that the barrier can be slid into the slits, thereby holding the barrier setup firmly in place.
A rubber gasket was made from a rubber pipe cap also secured from the hardware store. The gasket is placed at the rear of the parabola, and helps secure the fittings.

I made a wooden handle and secured it to the plastic pipe using a pipe clamp. The handle was coverd with gray foam, allowing for quiet handholding.

In Photo 12 you can see the two mikes as they appear from the rear of the unit. After attaching my cables, I stuff a piece of a plastic trash bag around the rear of the mikes to make it airtight and prevent the sound of me breathing from being transmitted through the tube to the mike elements.

Playback question

Wed, 25 Feb 2009 08:11:20 -0800

> Leo Boon wrote:
>
>>> ..I am looking for player and speakers
>>> which will [attract] wanted species...
>
>
> Dear Leo,
>
> The subject of field playback is very controversial. At a minimum you need to be aware of any local regulations regarding playback.
> Playback is prohibited in more places than you may realize. Most
> national and state parks in the USA regulate sound playback. These regulations protect species that might otherwise be subject to harrassment. Regulation of playback is especially important
> in "hotspots" (inside or outside parks) where an unusual species can be found, or where birders congregate (such as near a wilderness lodge). Many sound recordists "just say no" to field playback at any time. That said, I understand the value of playback when doing census work, or when trying to show an unusual species to a group. It's far less damaging to the environment to attract a bird over to your location than it is to make a new trail into the forest.

Walter Knapp makes this point as follows..................

A lot of the playback controversy is based on the assumption we know what the emotions of some animal are. That is a major error well known to those that study animal behavior (that study has a name, Animal Ethology). It was about the first thing brought up when I took courses in this subject. Playback is a area where otherwise fairly logical folks yield readily to the Bambi syndrome.

It is just as easy to make the interpretation of a excited male
responding to playback as building up his ability to ward off other males, building up his hormones as it is to think that he's being discouraged or confused. Used in moderation we are probably doing no damage. It is clear if you keep at it in the same area you can probably do harm to the dynamic of the established territories. That is the reason for many of the specific off limits areas. Other regulations are directly from someone under the influence of the Bambi syndrome. And it's easier to just blanket outlaw something than to look closer and limit it just where necessary.

It is also fairly well known that you cannot trust any vocalizations you get to be natural. So doing it so you can record is a bad idea. At best you should use it only to find locations and come back in a day or two without playback to try and record. For the most part, it will only attract males actively defending a territory. Or in the case of frogs, ones in breeding mode and inclined to sing.

The least expensive tool of this type is the Audobon Society "birdcall". This consists of a piece of wood with a metal rod loosely through it. You twist the rod to make various noises by it's action on the wood. You are attracting mostly by using a animal's curiosity, it's a strange sound. In the hands of experts I've seen very impressive use of this tool with many animals coming into view, not just birds.

Playback is a standard tool used in science. I use playback both playing recordings and my own poor vocal imitations of the call of the Pine Barrens Treefrog in my continued search for it in Georgia. In fact I just recently bought a amplified speaker for this. Not much help in what to buy, as it's way out of production, it's a Toshiba "Miniamp" speaker system I got for next to nothing off Ebay. About the same size as my Portadisc, it's from the days when circuits were made from individual transistors. Very convenient, runs off 8 D cells, has a mic with it as well as inputs for other sound sources. The 8 NIMH batteries I got cost
far more than the Miniamp.

I have no illusions about the coverage of such a speaker system, though it's rated at 100 yards. The way to evaluate that is start it playing and walk out and see how loud it really is. The falloff is dramatic. On anything you can run off batteries don't expect a lot of distance. In my case that's not a big problem as I only test specific locations and those can only be identified by being right at them.

In the Pine Barrens area in Florida they were known from three sites and were considered very endangered after several years of regular looking had turned up no more. By the use of vocal imitations and playback it was found in several hundred sites across several counties in one season in a successful effort to get it off the endangered list. Pine Barrens Treefrogs have a calling pattern that makes it very hard to survey them
just by listening. They form into "calling groups" of just a few
individuals, often several calling groups in a area. Typically only one calling group is active at a time. They call for a few minutes, and then are silent for 20-30 minutes before repeating. For survey that means you could, at best, check out just a few sites in a night, and their entire calling season is only a few months. They will, however, readily respond
to even crude imitations of their call.

Walter Knapp

how do i record rain sounds?

Wed, 4 Feb 2009 08:17:19 -0800

Is there a technique to recording Rain sounds?

Like ocean waves, this one requires some thought. Again, we hear what we see. If we're looking at a puddle in the rain, we tend to hear the raindrops hitting the water in the puddle. If we happen to be looking at a large leaf in our garden or in the forest, we'll tend to hear the effect of raindrops on the leaf.  There are two problems: One pertains to the type of mic you're using.  Condenser mics tend to be particularly sensitive to humidity and have a fairly high attrition rate over time under those conditions. Two types we've found help mitigate that problem - the Sennheiser MKH series, and a couple of less expensive Sony products like the ECM 55B lavs.  Best hope is to find a shelter in a biome in which you want to record, with a relatively high (8 - 12 ft - w/in the 3 - 4 meter range) shingled or thatched wood-construction roof. Record the rain at different periods, lite, moderate and heavy, being mindful to stay away from locations where water drips off of the roof edge (unless you want that effect). Because the heavy rain will sound like incessant white noise (the mic does not discriminate detail), you'll need to vary, in a mix, the different period samples in the timeframe you hope to create the illusion.  Good luck! Bernie Krause

what makes a good nature recording microphone

Wed, 4 Feb 2009 08:14:22 -0800

What makes a Good Microphone and how should I choose one?
To me a good mic is one that's still impressing me after I've used it for a season or more of nature recording. No amount of studying specs can by itself pick them out. I study specs, I read what others use them for and listen to the results, if possible I try and get one in my hands. I download the spec sheets and put them in my collection and compare them to those I already have downloaded, or even download more for comparison. I hunt up reviews and opinions. I note where someone has used them for a recording.  As far as the spec sheet the first thing I'll read is the polar pattern.  If it's not the right pattern for what I'm after the finest numbers in the world won't help.  Next is how it's powered, I prefer phantom power mics.  Usually the first number I'll read is the self noise. I'm looking for as low as I can get, but more I'm comparing that number to mics of the same type I already know. Where does it rank? I have a general cutoff at 15dBA, but that's not hard and fast.  Somewhere about here I'll look at the photos and dimensions and think about how it would be suspended and wind protected. How well it could be used hand held. Check the weight, try to guess the balance.  Next I'll look at the polar diagrams and sensitivity graphs. I'm looking primarily for a fairly flat sensitivity graph and smooth polar diagrams.  I make a guess as to how authentic and representative they are.  After that I'll go through the other numbers. Sensitivity is important, but in the class of mics I might look at not a big problem.  At that point, if not before I'm into thinking about using the mic in practical situations. For instance if the mic were one to be mounted in a SASS, having a even diameter along it's length and a well exposed diaphragm is important.  But building things around it is only one thing. I visualize sites I might record with it. By looking at the polar graph and spectral chart I try and sort out how it would handle picking up the site I'm visualizing.  I also pay attention to diaphragm size, generally nature recording mics have diaphragms on the smaller size, but I'd like to experiment more with large diaphragm mics.  I also evaluate the build quality and get some idea about repairs.  Nature recording exposes mics to quite a few hazards and you want a mic that can take that. If you are considering a mic that's not been used for nature recording but just studio there's a significant chance it will fail quickly outdoors, particularly in humid conditions. You won't find specs for this. The experiences of other nature recordists is about all you have to go on. And you have to know what sort of places they record in. The popularity of Sennheiser's ME and MKH mics, particularly the latter, for nature recording has a lot to do with their toughness out in the environment. Couple that with excellent specs and sound and it's easy to see why they are the safe choice.  > I have a mic that is used on a video camera. Its Specs are  > Unidirectional, Frequency response 70- 16 000 but you for get anything  > below 700Hz. Output Impedency 600ohms, Sensitivity -68dB and its noisy. > To assist in the poor sensitivity I use a umbrella for a reflector  > which many strange looks have come my way. I record on a minidisk  > mostly birds which I have to get close as possible to. Then I didn't  > know what I was getting but I had to start some where. I'm a little  > wiser now but there is a lot of things that still hazy.  That's really the trick, get out and record. Nothing helps with all this like hands on experience. At first the information you read will be a jumble, but it's organized primarily by using it.  Walt wwknapp@mindspring.com

microphone self noise

Wed, 4 Feb 2009 08:12:22 -0800

“I’m looking for a microphone to record sounds in nature. Since many of these sounds aren’t very loud I’ve determined that I want to get a mic with very low self-noise. Most good mics have a rating for this, and I assume lower numbers are better, but I don’t really understand the specification so I thought I should check.”  
Unfortunately it's not straight simple answers.  For instance, on "how much extra hiss I'm going to hear in the background with an extra 10dB". This depends, primarily on the gain setting you use on your mic input, it's variable. As you increase the gain, the mic's self noise is amplified right along with the sound you want. It's generally better to think in terms of the site you might can record. With 10dB lower self noise you can record a site that's 10dB quieter and still end up with the same background hiss level. Since in nature recording we are often dealing with recording quiet sites, self noise is a limiter on how many sites we can record well. If we try to record too quiet a site the sounds from the site will sink into the noise floor of the mic.  The self noise number is determined in testing by comparing it to a standard 94dB sound level. It's only really valid at that gain setting.  For nature recording we are usually recording quieter sounds, so more gain.  You may note that signal to noise specifications are generally just another way of saying the same thing. The self noise figure is subtracted from the 94dB test level.  Incidentally, when we say dBA we mean the test sounds had a frequency/intensity distribution that paralleled human hearing. One can also test using a even frequency/intensity distribution, which is usually what it is when the figure is given in dB. The two numbers may be quite different.  There's another aspect to self noise. It does not all sound the same.  With some mics it's a nice smooth sound that you hardly notice, with others it might be a sputtering or crackling sort of sound. Obviously with the smooth sound you can have it somewhat audible and not bother the listener, but with some of the others any audibility is undesirable.  At the beginning of this group one of the first discussions was just how good a self noise figure was needed. While there was no absolute agreement a figure of 16dBA or less was more or less the conclusion.  Above that figure it was going to often be limiting, below that figure it would be less of a problem.  Self noise is only one thing of importance. Others are sensitivity, handling noise, wind sensitivity, humidity and temperature sensitivity.  It's rare that either dynamic range or max sound level is a issue for mics in nature recording.  > I have looked at a number of sites that has a general advise on tools  > and methods. Also there have been a number of references on what  > makes a good mic in this group but there is hell a lot of messages to  > read to get a grasp of it all. I'm asking the impossible, can there be  > a set of guidelines amongst the group that would make an ideal mic,  > one that would do the job and those that will disappoint you for  > individual calls vs ambiance recordings. There is no need to mention  > specific makes of mics or other additions such as barriers as one  > needs to get a grasp of the basics before being side tracked down  > another path. Some one could explain in practical terms what the  > varies specifications are. This maybe a useful reference for the home page.  The problem with the concept of a ideal mic is there is no such animal.  A mic that shines for one use may be a dog for another. I tend to think in a purpose orientation. For instance if I want to record a entire marsh full of many species of frogs I would use a SASS mic for a overall recording of the marsh, the ambiance and all the calls together in a wide view. That's a good purpose for that mic. But, if I wanted to pick out a individual frog, or small group in the same marsh, I'd use the Telinga parabolic. Because it's stereo the Telinga will still contain some ambiance, but will bring out the wanted individual or group. By aiming at each group I could record the entire marsh piece by piece, something I do if recording for scientific survey. My different mic setups are chosen to cover a range of field widths and distances. For any particular use one or another is going to be ideal.  This is why our response to someone asking for the ideal mic tends to be to ask what they intend to record.  For picking out individual calls the focus is usually on directionality, we want to exclude sounds other than the calls we want. Two types of mics are common here. The parabolic mic has the sharpest directionality, it will reject the most of the surrounding sounds. And, because of the reflector it's got extra gain that occurs before the mic. Gain that does not amplify the mic self noise. A noisier mic can be used in a parabolic as a result. For those with a limited mic budget who are doing individual calls the parabolic is the most bang for the buck.  The alternative for individual call directionality is the shotgun mic.  It rejects sound from the side, though not as sharply as a parabolic.  It, however, provides no extra gain before the mic for the caller you are pointing at. You have to provide that extra gain amplifying the signal and thus the self noise. For this reason a shotgun mic has to be very low self noise, and ideally high sensitivity. That combination makes good shotgun mics for nature recording expensive.  Of course there is a third alternative for individual calls. That's getting close. That opens up the mic choice. And you can get into things other than mics. Like Ghillie suits and other stealth methods.  Ambiance adds a different problem. You want to record a sound field, all the sounds. You may or may not be doing this in conjunction with recording individual calls within the sound field. What this introduces is that all the quieter parts become important and need to be accurately recorded. Thus the need to have the lowest self noise in the mics. Which tends to mean expensive. It also introduces the concept of field size, you could record the ambiance of a small area all the way up to everything all around you to all distances (the last for which there is no ideal mic)  Ambiance also brings in space or physical volume. Which is best done in stereo. This not only introduces the added cost of a 2nd mic, but the problems of relative positioning of the mics to accurately portray the sound field. Here I can give you a readable reference, "The New Stereo Soundbook" a good coverage of the techniques that does not get over technical: http://www.stereosoundbook.com/  I should note that, although it may seem so, nature recording does not use the very highest priced mics much. The Sennheiser MKH mics, as high priced as they are do not compare to others at wallet deflating. Price is certainly a consideration, but it's easy to end up with a system where the recorder is far more capable than the mic attached to it. The mic is the most critical component. There is a lot of focus right now on high sampling rate uncompressed sound recording with very little notice that the mics used won't come close to providing the quality of signal needed. Just recently I saw one person discussing this and he figured the mics he could afford did not even challenge 16bit 44khz sampling.  For the most part we work with mic designs that were perfected for analog tape. Putting the word digital on them does not change that.  Walt wwknapp@mindspring.com 
A general statement which is almost true:
Considering "sound quality", all microphones today are as good as the best ones 20 years ago.So don't worry about sound quality, frequency response etc. It's good enough.  Then the noise problem:  Listen to "My Garden" on telinga.com, click sound gallery.The microphones in question had an inherent noise level of some 8-9 dbA. That's low.  In this category of mic's you will find the Sennheiser MKH series, the telinga MPS1 series and all those recently marketed low-noise mic's using the same China-made 1 inch capsule, like the CAD 179 for example. (Those China capsules are very different from one another! You need to match them to get a stereo pair!)  If you replay such a recording at a reasonably low level, like in reality or a bit louder, 10 dbA will not be audible at a recording like "my garden".16 dbA will be audible, but not disturbing. 23 dbA will be disturbing and something which you will try to filter away. Roughly.  Considering an owl, 50 meters away, deadly silent around - 16 dbA will not only be audible, but also disturbing.  Noise is related to size. The bigger the membrane is, the easier it is to make it low noise. All 5 mm electrets have a noise level about 30 dbA or worse. All 10 mm capsules have a noise level at about 26 dbA or worse. (All except one) All 20 mm capsules, like Sennheiser ME-series and Telinga EM23, make about 16 dbA noise, or worse. The Sennheiser MKH and Telinga MPS1 are here exceptional. They have 20 mm membranes, but a noise which is only 10 dbA or better.  Hope it helps a bit.  Klas. 
Another key factor that might also be added to the discussion about self noise is "sensitivity." The MKH's have both low noise and high sensitivity. The KM-23, though it has a very respectable rating of 16dB(A) self noise, also has high sensitivity. The Sony MS-957 has a considerably higher value of noise @ 25 dB(A) but the presence of this noise in the signal is more apparent because its sensitivity is only ~6 mV/Pa and the record gain in quiet locations must be increased for sufficient record volume. The Shure WL-183 has 22.5dB(A) self noise but its high sensitivity of 40 mV/Pa allows it to take greater advantage of its moderate self noise. The ideal combination of low self noise and high sensitivity contribute to a mic's ability to perform well under "very quiet" conditions. So, high "mV/Pa" numbers are good and low "dB(A)" numbers are good. Any number above 20mVPa is very good; any number below 10dB(A) is very good. Mic manufacturers' testing methods vary considerably, so the published numbers should only be regarded as ballpark "estimates"  until proven in the field. Many terrific recordings have been made with modest gear that is perfectly suited for the situation-- including sound levels. The placement of the mics in the sound field can make the recording sing or seem dull.
Rob D.

Mic noise is one thing - mic preamp noise is something else.
If the output / Pa of a microphone is low, - then you have to use a lot of preamp gain to get the signal up, but then you will hear the preamp noise!!  - not the mic noise!  If a mic has a 23 db self noise, its still has a 23 db self noise, even if the output is high.  In practice you could be correct: A low output microphone with a good self noise might give a poorer result than a high output mic with a poorer self noise. But then it is because of preamp noise, and could be improved by getting a better preamp.  The word "sensitivity" is also used by the Japs to describe self noise, instead of dbA. It's very confusing.  Klas