One of the things I’ve had almost no experience in until this year is onboard vehicle recording. To be honest, I probably subconsciously avoided it for so long because of how difficult/time consuming it can be to get right. There’s a lot of things to factor in: Number of mics, Mic placement, Wind protection, Gear protection, Performances to capture, etc.
For a recent job, however, I was tasked with creating a library of sounds for a Harley motorcycle. Luckily, before I even learned of this job, I worked up the nerve to attempt recording something a bit less manly: my moped… a 1987 Honda Spree. Apparently my subconscious can work for good too. The Harley record went great! More on that in a bit. First let me tell you about all of the things that went wrong with my moped record. Mistakes are more fun to talk about, right?
Forgive me in advance, I have no pictures of the moped record. Part of that had to do with it not being my prettiest or most successful recording session. I’ll do my best to describe the problems.
Moped Record – Problem 1: Mounting the Mics
Believe it or not, I got mic placement/choice mostly right on my first attempt. One near the engine and one by the muffler. I also had an MS rig mounted on the rear of the moped pointing toward the ground (this turned out to be mostly useless, however). The problem I ran into was mounting the mics properly. I’ll start with the engine and muffler mics…
I used an SM57 for the engine, and a KM184 for the exhaust. I ended up using duct tape to hold everything in place, which worked out great at first! I taped the SM57 directly to the inside body. I taped a small boom pole to the frame just above the exhaust and stuck a KM184 on the end of it. I got on the scooter to make sure everything was still in the clear when I sat on it. It was. I gave the scooter some big bounces and shakes to make sure nothing shifted. Nothing moved. I was pretty proud of myself at this point. This first attempt was obviously going to be a piece of cake. I started up the moped and took a short ride to the recording location, a large church parking lot.
I got there and everything still seemed alright. Mics were still there… they hadn’t shifted… no dings or scrapes. I started going through the list of performances I wanted to capture. I had recorded my drive to the location as a test, so I was pretty confident in my levels. For that reason I didn’t feel the need to playback the recordings as often as I normally would. I got about 1/4th of the way through my list when I smelled something burning/melting. I looked back at the exhaust mic and saw the foam was resting against the tire. “Ok, mics are probably just settling into place”, I thought to myself. I adjusted the mic and went on recording.
Maybe 15 minutes later I felt a clunk as I went over a bump. I stopped to see the engine mic had fallen and was resting on the gearbox. I was using some more duct tape to make things more secure when I saw the boom arm holding the exhaust mic was drooping. Taking a closer look at the attachment points for that I realized the REAL problem: Duct Tape and heat don’t get along well. The mic positions we’re fine to start. But as the duct tape began to warm up from the engine/exhaust heat, it lost it’s rigidity and actually started to melt. This allowed the mics to slowly droop throughout the record.
When all else fails, use more duct tape, right? Well in this case, that probably wouldn’t have helped much. I decided to call off the record. I still captured around half of what I had hoped to get. Also I had been recording by priority, so the sounds I did get were the most important to me.
Moped Record – Problem 2: Selecting a Location
When I was planning this record, I was thinking of places where there wouldn’t be a lot of people. And I succeeded at that! The church parking lot I recorded at was empty for the whole (half?) record. What I should have been thinking of, however was very long stretches of road. In listening back to some of the recordings I had realized that because of the proximity of the mics and the engine/exhaust, almost NO exterior sounds were noticeable. This was good, because although there were no people around, there were plenty of birds.
No, people and birds weren’t the problem. The issue was the amount of space I had to move around. The parking lot was very large, but didn’t give me enough room to keep my speed up for very long. The slow speed recordings weren’t much of an issue, but I really started to realize the conundrum as I picked up speed. The faster I went, the less time I had before I needed to stop and turn around. As bad as it was… I think the fact that it was a 49cc moped that could barely make 35mph helped me there. Had I been on anything faster, the location probably wouldn’t have worked at all.
Moped Record – Problem 3: Deciding on Performances
One area I spent a little too much time planning for(or not enough, depending how you look at it), was deciding what performances to capture. I thought about all of the possible types of performances I would need to capture to make sure I’d be covered for anything I needed when using it in post. I fell down the rabbit hole a bit. I’ll admit it.
“What if the moped is going at a medium speed and the speeds up to fast… slowly?”
“What if the moped is going at a fast speed and decelerates to a slow speed… quickly?”
“What if the moped pulls off slowly up to a medium speed?”
“What if the moped pulls off quickly to a slow speed?”
I ended up with quite a list of things to capture, and that was just the onboard variables. To be honest, had I planned what I REALLY needed to capture a little bit better, I probably could have gotten through the record before I noticed the problems with my mounting methods.
You see, somehow I have been able to cut every single car/motorcycle scene that’s been presented to me with a fraction of what I was attempting to record that day. I should have been thinking about that. Sure… there are ways to improve on most onboard vehicle libraries, but what I was trying to do was overkill.
Ok, enough talk about failure… let’s talk about success! For each of my shortcomings on the moped record, I was able find a solution and improve the Harley record. I also had some help from my great fellow sound guy buddy AJ Olstad, who helped with the set up and got to do the fun part: riding!
Harley Record – Success 1: Mounting the Mics:
Since I knew to stay away from duct tape, we decided on using some very tough zip-ties. I still kept mic position pretty much the same, but this time chose the points of contact very carefully. The goal was to keep them away from any significant source of heat. The exhaust mic was on a boom arm mounted to some frame work behind the seat. We found a nice little cavity for the engine mic that was far enough from the engine to still be safe. Aside from zip-ties we also used foam and moleskin to reduce vibration and protect the bike. We ran xlr cable (being mindful of moving parts!) to the saddle bag where we had a 702t to track everything.
I couldn’t have been happier with the result. We did a test record on the way to the location to get rough levels and make sure nothing was going to fall off. When we got there, everything was still rock solid! No mic drooping or melting foam.
Harley Record – Success 2: Selecting a Location
Selecting a location wasn’t too difficult. We were in the distant suburbs of Chicago during work hours, so that alone reduced the number of people out. I know I said earlier I found out that didn’t matter so much, but this time I was also recording from a stationary perspective (more on that later). I used google maps to scout out potential locations ahead of time. It’s such a great resource when setting up for a session like this. You can use satellite view to get an idea of potential problems like nearby highways/traffic. Furthermore, if available, street-view gives you a great idea of what things will look like when you get there. This was a HUGE help for my Diffuse City library 😉
We settled on a long stretch of road that almost no one drove on. It was far enough away from the city and any highways that the only troublemakers I ran into were a few birds. They were close enough I could usually clap loudly to scare them off. I’m starting to think I should carry around a starter pistol or something for these kind of sessions!
Harley Record – Success 3: Deciding on Performances
When I recorded my moped, I was on my own time. That was a good and bad thing. Good, obviously because I could take my time. Bad because, well… I could take my time.
This time, while we weren’t on a set schedule, we wanted to treat it like a work day and be done around 5. It took ~1.5 hours to drive each way, so instead of eight hours, we actually had around five. This meant planning what performances to get was very important. I wanted to get everything we needed and nothing we didn’t. I threw out my rabbit hole list and started anew. I referenced a few motorcycle SFX libraries to see what others had done. I came up with a general list and made choices on what could be added as well as what could be taken away.
I ended up with a list of just 14 essential performances that would cover almost everything you would want in a motorcycle library. Now, that’s not counting things like horn honks/foley movements/etc… but for this project, none of that was needed. 14 performances may not seem like a lot, but with mic setup we had, we ended up with a huge amount of versatile material.
Like I mentioned, there was a KM184 and SM57 on the bike. For the stationary recordings (pass bys/pull aways/etc) I had a 744t with two M/S mic setups: My trusty Oktava MK-012’s on a stand for left to right movement, and a handheld 418. This was very helpful because each performance accomplished multiple things. For instance, I knew we wanted steady onboard fast speed recordings. We also wanted fast pass bys. With just a fast pass by we were capturing both of those. Not only that, we were capturing two important stationary perspectives: left/right movement with the MK-012’s and a mic-follows-motorcycle perspective with the handheld stereo shotgun.
With each performance, we captured a huge amount of versatile content. And after 14 performance (with multiple takes, of course), we still had time to kill! There were a few other things specific to this project we wanted to get. We put some contact mics on various points of the frame, engine, and gear box to capture some “internal” type sounds. We captured the classic sounding tings that happen after a Harley motor is turned off. We even went on to record a few neat pass-bys and pull aways from a different bike. AJ had a friend nearby that was more than happy to help us out with that.
After bringing the recordings into a session and checking out what we had, I can’t tell you how happy I was with the results… not only in the sound quality, but also by the versatility of each performance. While I wouldn’t call it a comprehensive library, I’m comfortable enough to say I could cut just about any motorcycle driving scene with what we did capture, and I’m very proud of that.
It’s amazing to me how much you can learn from just one attempt at something; even if you fail miserably. You can research and plan all you want, but the only way you’re going to truly learn your own way is by doing it yourself. I’m just happy I failed on my own project, and not a client job! 😀
OK! I’ll shut up already. Here’s a little edit of some of the stuff we captured that day. Crank it up!
You can hear raw audio from the three perspectives we captured from each performance. The onboard mics turned out great (aside from a little wind noise at very high speeds). The handheld 418 gives a nice non directional stereo image. The stationary Oktavas capture the movement of the bike as well as the natural reverb of the space, which can be very useful in post work.
All in all, I’m very happy with what we got that day. Excited to try out some more onboard recording in the near future!