• Wondering which camera, gear, computer, or software to buy? Ask in our Gear Guide.

Recording Sound in Service Tunnel

Hello all, I'm the soundwoman in a classmate's second year film. It's a location film, and all of it takes place in a service tunnel. We're going to see the location tomorrow, but I'm (obviously) pretty worried about echoes.
Any suggestions? Anything I can cover the walls with in close-ups, anything I can do with the neckmikes?
Would greatly appreciate tips and advice!
 
Recording production sound in highly reflective locations is a real bitch.

Obviously a great boom-op will be a great help, but sometimes that is not enough. Lavaliers, properly placed on the talent, can definitely have a much less ambient sound. Using as much sound absorption as you can will help as well.

Something else that helps is the positioning of the actors. If you can keep them further away from the walls and try to keep them at angles from the walls the sound will have (marginally) less recorded bounce.

This is really a tough gig. I would suggest that, immediately after the shooting of the scene is complete, you record wild lines very close with the boomed mic in the location. When I put on my dialog editors hat I really like wilds as the actors as still in character and the "pacing" of the lines is similar.

You may want to post your question here:

http://jwsoundgroup.net

This is a group for production sound professionals hosted by production sound guru Jeff Wexler.

GOOD LUCK!!!
 
Thanks! So, did some tests at the location today. Boom is meh, definitely echo there, though the closer I can get the less echo I get. I have an option of taking a higher-quality Rhode gun mike or a lower-quality gun mike that's longer. Does that matter at all?

I'm recording with a Zoom Hn8. Are there any settings I should know of that can help cut off the echo? Someone told me to try to cut off the frequencies...

I will also be using lavaliers / neck mikes. I know from the tests that the dialogue of the person with the mike comes out well, but any other sound I get on it is echo-y. Actors will be wearing military uniforms + undershirts, the lavaliers will be wrapped in (cigarrete) filters. Any other tips?
Pretty worried about the gig, especially since I know the director probably won't invest in a professional sound mix.
 
Thanks! So, did some tests at the location today. Boom is meh, definitely echo there, though the closer I can get the less echo I get. I have an option of taking a higher-quality Rhode gun mike or a lower-quality gun mike that's longer. Does that matter at all?

You may want to think about a hypercardioid mic, as shotguns mics tend to exaggerate sound bounce in "hard" locations. I won't get into the physics, but a hypercardioid will have (again, marginally) less echo/ambience. Even if you use lavs record with a boomed mic. As an audio post editor and rerecording mixer I like having the room sound to mix into the lavs if I want too.

I'm recording with a Zoom Hn8. Are there any settings I should know of that can help cut off the echo? Someone told me to try to cut off the frequencies...

With the exception of excessive low end you SHOULD NOT try to do any type of EQing on location; this is a job for the audio post folks.

I will also be using lavaliers / neck mikes. I know from the tests that the dialogue of the person with the mike comes out well, but any other sound I get on it is echo-y. Actors will be wearing military uniforms + undershirts, the lavaliers will be wrapped in (cigarrete) filters.

I don't know what type of soldiers your actors will be, but one "trick" when shooting "modern" special forces types is to remember that they have sophisticated communications systems which includes headsets. You can also be creative with lav placement - next to a button, in a pen, under the brim of a baseball cap, etc. Just something to discuss with the director and wardrobe.

SOF%20headset%204%20March%202014.jpg


Any other tips?

Getting in close is always the best option with boomed mics; the closer the better.

Use as many sound blankets and other sound absorption as you possibly can; every little bit helps.

Don't forget the wild lines. If the director says "no" on location or does not use them during the DX edit at least you have done your job; it's not your responsibility if the director does not do his/hers.

Pretty worried about the gig, especially since I know the director probably won't invest in a professional sound mix.

Audio post is your directors problem; your job is to get the best production sound possible in the given circumstances. The director chose the location and should be aware of the limitations forced upon the production sound team. If s/he doesn't want to invest in a proper audio post that's not your problem.
 
Can you explain a bit more about how / when to do this?

Just record the actors saying their lines a few times with a mic in close - about 6" to 12" - while in the service tunnel after the shooting of the scene is completed. You'll have to use your judgment as to the distance re: bounce vs. clarity. As I stated in my first post the actors are still in character and the "pacing" of the lines is similar.

If you haven't purchased "The Location Sound Bible: How to Record Professional Dialog for Film and TV" by Ric Viers you should; a solid overview of doing production sound.

13592487._UY630_SR1200,630_.jpg
 
Thanks! So, did some tests at the location today. Boom is meh, definitely echo there, though the closer I can get the less echo I get. I have an option of taking a higher-quality Rhode gun mike or a lower-quality gun mike that's longer. Does that matter at all?

I'm recording with a Zoom Hn8. Are there any settings I should know of that can help cut off the echo? Someone told me to try to cut off the frequencies...

It is very important to know your equipment and know the right names for them.

"Rhode" doesn't exist. Rode though is an audio company that does exist, but they product many different mics, so you can't just say "a Rode mic".

Likewise when you said "Zoom Hn8", that doesn't exist at all!

Neither does an "Zoom H8n", and neither does a "Zoom H8" exist. Which really does make it tricky to figure out what on earth you might the hell be talking about it???

My next guesses are you are mean the Zoom F8n (but I very highly doubt that, as it is a brand new product which only just came out. I only today got the first F8n in my country!), or the older Zoom F8 or the Zoom H6 (or maaaaybe even the H5 or H4n).

But these guesses are little help here, seeing as how the F8 is such a radically different machine to the H6!

So please please, *learn* the names of what gear you are using, and refer to them correctly.
 
A hi-pass will not have any noticeable affect on ambient reflections. All it will do - in your example - is remove all frequencies below 80Hz.

EQ3Settings.png

I thought (and please do correct me if I'm wrong) that it depended if the reflective surface was completely flat because non linear (i.e. not flat) surfaces change the frequency of reflections. So, according to the laws of physics:

- Reflection:

Reflection occurs when a wave bounces from the surface of an obstacle. None of the properties of a wave are changed by reflection. The wavelength, frequency, period and speed are same before and after reflection. The only change is the direction in which the wave is travelling.

However, there are exceptions. One of the exceptions is if the denser medium is nonlinear. e.g. a tunnel.

The effect of a non-linear denser medium (i.e. the tunnel) will lower the frequency of the wave so the returning echo will be lower than the initial sound of the voice and can be partially removed by removing frequencies lower than the initial emitter. In this case, the voice. So I would pop a high pass for a woman at 160 and a man at 80 (depending on their voice) as if the tunnel is not a simple, flat wall and this would help.

I thought the Kirchhoff Integral theorem is a place to start as although it was focused on optical waves, it is also applied to sound. And research seems to indicate there is a frequency change such as this paper on the effects of concave surfaces on sound waves: https://www.peutz.nl/sites/peutz.nl/files/publicaties/reflection_of_sound_by_concave_surfaces.pdf

Have you been to Lascaux in France to hear this effect in action? If you go to the prehistoric caves, the sound of clapping reverberates and genuinely sounds like a stampeding herd. It is audibly different and you can hear the frequency change yourself.

Again, please feel free to correct me if I'm wrong!
 
I am going to disagree on the basis of practical experience. I've never had any success whatsoever removing ambience/echoes with a hi-pass filter.

As to the scienceā€¦ Yes, sound does bounce around in "hard" spaces. Yes, high end sounds diminish more rapidly as they lose energy more quickly. Yes, frequencies can change as they interact as they bounce around. But frequencies in the kHz range dropping to Hz range with volume levels significant enough to require a hi-pass filter? Nope, not in my experience.


My (very limited) understanding of the Kirchhoff Integral theorem is that the effect is more significant the higher the frequency. Light frequencies are far more energetic by several degrees of magnitude than sound waves, so the affect is close to unnoticeable at a sonic level in practical usage.

I've never heard of anyone solving ambience/echo issues with a hi-pass filter, but I could be wrong. You may want to bring this up on the JWSoundGroup production sound forum.

BTW, a man made tunnel IS linear when compared to a cave.
 
Last edited:
I am going to disagree on the basis of practical experience. I've never had any success whatsoever removing ambience/echoes with a hi-pass filter.

As to the scienceā€¦ Yes, sound does bounce around in "hard" spaces. Yes, high end sounds diminish more rapidly as they lose energy more quickly. Yes, frequencies can change as they interact as they bounce around. But frequencies in the kHz range dropping to Hz range with volume levels significant enough to require a hi-pass filter? Nope, not in my experience.


My (very limited) understanding of the Kirchhoff Integral theorem is that the effect is more significant the higher the frequency. Light frequencies are far more energetic by several degrees of magnitude than sound waves, so the affect is close to unnoticeable at a sonic level in practical usage.

I've never heard of anyone solving ambience/echo issues with a hi-pass filter, but I could be wrong. You may want to bring this up on the JWSoundGroup production sound forum.

BTW, a man made tunnel IS linear when compared to a cave.

I didn't state the problem was solvable with a high pass or low cut filter, I said (and I quote myself) "it will help."

And just to re-affirm (I think you're saying the same thing as me), frequencies cannot change through reflection per se as everything we know about physics would simple be wrong. Rather they change when reflected by a non-linear, denser medium and / or a moving, denser medium (the latter you would know simply as the Doppler effect). And service tunnels would fit the first, part of this description unless the service tunnel is simply a perfectly, flat wall without ceiling or protrusions.

So, a high pass or low cut filter will remove a little bit of the echo and help. The amount of help cannot be easily calculated (if you read the research paper I provided, the physics / maths is in there) as it depends on the how non-linear the environment is.

My principle is that if I do ten things, each of which removes a tenth of the problem, at the end I have no more problem. This is simply one of them.

But if you love sound, try that cave. It's an amazing hypothesis that early man was storytelling through pictures and sound and if correct, would make that cave the first, ever Foley studio. I wonder if there was a producer saying, don't worry if you mess up, we'll be able to fix it in post...
 
Back
Top