ART USB Dual Pre digital artifact

OK. Here it is. Set the microphone for Mid/Side and turn the Side volume all the way down. That’s a mono voice track. Announce into the rear of the H2n, the side away from the display. They don’t tell you how you’re supposed to see the volume meters when you do that, but do it in front of a mirror until you get the hang of it.

The H2n instruction book is too large to post.


The h2n set to mono yielded a nice result. The interface is tricky, I hadn’t figured out how to set the side mics to OFF until now. They were at 0, which turns out to mean mid way.

The problem, of course, is that the h2n still picks up a lot of background noise compared to the Coles, which is pretty much unbeatable in this respect. With the ART, I had a lot less hiss. I will have to try shielding it according to Cyrano’s instructions.

Time for the goal. Nowhere in this posting is the eventual goal mentioned.

Buried in the hiss is the other shortcoming of lip microphones. P-popping and “wet mouth noises.” In the wild, nobody cares if Piper 202 has mouth distortion when calling the tower for landing as long as most of the intelligence is maintained. To this end most communication systems filter out low frequencies and exhibit many of the characteristics of a telephone filter.

The techniques for signaling to an audience you’re in an airplane during a reading are: close-talking, P-popping, telephone filter and saying things like “Piper 202 request landing.”

So goals such as reading for audiobooks or other entertainment reading are unlikely to be met even if the setup didn’t receive electrical interference and radio shows.


The goal: Recording a language instruction course, specifically laboratory “tapes” for student practice. Basically the same goal as recording audiobooks.

I personally think the shortcomings of not recording in a studio but in a regular room, with bare walls, pages rustling, birds chirping outside, an occasional tractor or airplane, doors closing, people talking in the next room, etc., make the Coles a better choice than the h2n, if I can get rid of some of that hiss. Of course it’s tempting to do this the right way and invest in soundproofing.

I hooked up the Coles directly to the h2n, and I got a signal of about -30db. When I boost that in Audacity, the hiss is worse than with a preamp (no surprise). I’m wondering what the result might be if I used a recorder with higher gain, like the DR-100mkII. I read that compared to the h4n set on maximum (assuming the h4n and the h2n have similar sensitivity), the DR-100mkii was set at 4/10 to achieve similar levels. Eliminating an intermediary pre and an extra cable, and the XLR instead of mini jack input, would surely improve things a bit.

EDIT: Google tells me that the DR-100mkii has a nominal input gain of -70db, while the h4n is -47. I don’t know if that would make my mic 23db louder.

I have an earlier H4, not H4n. Nobody is writing the papers about the good quality of the H series external preamps. I didn’t think the H2n had a microphone connection. Or if it does, it’s an afterthought.

I do agree completely about using a non-computer recorder. People get into enormous trouble with USB connections and hidden computer processing. The forum is full of postings like: “Why does my AudioBook sound like a cellphone?”

Don’t let a “studio” scare you. There’s no shortage of winter coat closets posing as studios. One poster did it with hardware store plastic pipes smashed together (without glue) and moving blankets hung over them. I believe we helped her get published.

There are techniques where you make a little tunnel on the table for the microphone and hang a blanket behind you. You don’t have to soundproof the whole world. More later.


We did this soundproofing thing to death a while back and I saved it.

You don’t need a studio to make a recording. People make good recordings in difficult circumstances all the time, but not on a budget.

If that’s a Sennheiser shotgun, it goes for about $1000 usd, not counting low noise field mixer and recorder. That works out to about $2000 (not counting the camera). Sign here.

Most people find PVC pipes, moving quilts and clothespins a lot more economical. I can’t stress enough the value of a quiet room. It’s the difference between simple volume corrections to your chapter and you’re done, and painful filters, corrections, effects and adjustments. ACX has a quality failure they call “Overprocessing.” You can’t beat up your faulty recording in post production and have it sound good. Or worse, you may be able to do it three out of five times. The other two, you read again. Wouldn’t that be fun?

If no other good reason, if you do this enough, you will spend a significant amount of time in post production instead of planning your next show.


Thanks for the tips. I’ll be finishing a part of the attic when it gets warm enough to work without gloves. Aside from lots of fluffy material on the walls to reduce echo, how about soundproofing to reduce unwanted external noise (i.e. 6 boys wrestling and throwing objects and one girl screaming bloody murder)? For example, would it be useful to build a box (a room within a room)? I’m sure Google will give me some ideas.

screaming bloody murder

That explains the communications type microphone.

You’re talking yourself into recording somewhere else entirely. There are professional audio people who will refuse to record in an environment like that.

There is no up side. If you say yes and it works, you will be the Go-To Person for difficult/impossible recordings. If it fails, your name is mud.

Or you could just say, “Oh, I’m sorry. I hear me mum calling.”


Use mineral wool boards. Thickness: 10 cm or more. 55 kg per cubic meter density is ideal. More won’t hurt, but will probably be a lot more expensive and the gain is minimal. Your average heat isolation material is only 35 kg per cubic meter. Much less efficient.

A box-in-a-box is the best solution, but fairly tricky to make. Ideally, it should be completely floating, so get out the antigravity generator :smiley:

Doors, windows and floor are the hard stuff when you’re building a box-in-a-box. They make bridges and that undoes all your other efforts. And ventilation is a pita too, as the goal is the opposite of isolation.

Read up about the “triple leaf effect”. Most people would expect 3 layers with air in between to perform better than 2 layers. In practice, the 2 layer effort usually works better.

About 1/3 of the info you’ll find on the net, about acoustics, is only meant to sell overpriced stuff that doesn’t really work, like the foam you see everywhere (pyramid foam).

Another third is plain wrong.

It’s the last third part you want :smiley:


Do you mean boards like this? But 10cm thick? Do you mean wool like this?

Something like ROXUL SAFE’n’SOUND, but 4" in stead of 3":’n’sound

It’s a thermal isolation board, common in building materials shops. Just not the most common 35 kg/m3, but a bit denser, at 40 kg/m3. 50 or 55 kg/m3 would be even better.

It’s the same stuff you can buy in blankets, but boards, 10 cm thick. Cheaper than most other materials, usually in stock and easy to work with. Cut it with a large (electric) bread knife, put in place and glue or use long nails. For the ceiling, you can suspend it on nylon wire, more or less “floating”. You can’t easily use it for floors.

The stuff in the first link works even better, but is probably also more expensive and a lot more work. And the failure comes with the construction: the beams used to keep the two board constructions apart, are bridges again. Their isolation claims may be true, but they’re probably measured without these “bridges”. I couldn’t immediately find complete tech specs, which usually isn’t a good sign. And those they give, are mostly imperial, which is alien to my ISO mind :mrgreen:

The stuff in the second link probably works too. Only, I fear it’s even more expensive. But as I can’t figure out the exact specs, I can’t tell.

My method works well for the first part of the traject: putting something against existing walls and ceiling.

If you need to build new walls, the product from the first link is easier.

The product from the second link will work for stuffing the areas where a board is less practical, like corners or other small spaces.

I wouldn’t use glassfiber (which is even cheaper) as it is very nasty. You can’t stop scratching for days.

When the first part of the traject is finished (all existing walls, ceiling and floor covered), you can start building your box, if that is the way you decode to go. But it is very well possible that it is all you need, even if screaming girls are hard to dampen. :smiley:

While you’re beating up soundproofing, it’s good, since you’re starting from scratch, to not make the walls parallel. You can get away with far less cotton wool on the walls if they’re not flat to each other.

We had a sound room several millennia ago which had minimal actual soundproofing and it looked for all the world like a normal office with a little carpeting … until you measured it. None of the walls were flat to each other and the ceiling was tipped. No it didn’t have that pointy foam everywhere, but I had no trouble pushing multiple sound shows through there.

Flynwill on the forum was there for the design and he may have notes.

Time Marches On and I had a later office with surgically flat and aligned walls. My joke was I could clap and go to lunch and the clap would still be bouncing back and forth when I got back. That room would have been a nightmare to soundproof.

As we go.


I managed to try the Coles with a DR-100mk2. Here is the unmodified test recording. Any impressions?

Any impressions?

It’s noisy. I isolated the dialog portion of the test and applied volume and peak processing I’ve been using for AudioBooks. RMS and Peak work very well, but the noise level (fffff) is much too high.

That’s not the only problem. Why are you recording at 96000, 32-Floating instead of 44100, 16-bit? Several of the forum tools and MP3 exporting options stop working when you do that.

There’s something else wrong, too. The volume never goes taller than 0.5 (picture).
Screen Shot 2017-02-01 at 18.58.51.png
That’s wrong. Something is causing volume distortion. Any microphone system should be able to go all the way up and down to 1.0 on the blue waves when you yell. You don’t want to record a show that loud, but that’s where your vocal expression and theater emphasis go when you’re recording at normal volume. Yours get cut off and distorted.


Thank you very much for taking the time to check it out. To answer your question, I only had about 5 minutes with the device, which belongs to a cinema cooperative, and the sound guy is the one who set it. I assume he chose 96000 and 32 bit floating because that’s the best the device will do. About the loud sounds being clipped, I wonder if it was a problem with the settings (is that what a limiter does? maybe it was turned on).

(is that what a limiter does? maybe it was turned on)

One version of limiter can do that, but Audacity doesn’t apply effects, filters or corrections during recording. Stopping the blue waves at half-way is an alarm signal. Proper recording systems are never supposed to do that.

Yelling is a valid test. It doesn’t permanently break anything and it can tell you how your equipment works. When people pay you, it’s good to know not only how your gear works, but how it doesn’t and why.

I can’t do it. You have too many odd settings, symptoms and distortions. Sorry. Maybe one of the other elves will drop in.


It’s clipping at +/- 0.5
You can see it clearly if you zoom in close:

I’m not familiar with the DR-100mk2, but I’m guessing that it’s a similar issue that I’ve seen on the Zoom H2.

On the H2, there is a three position “gain” switch. This switch applies high, medium or low gain to the microphone signal. For recording loud sounds you need low gain, and for recording quiet sounds you need high gain. The more “gain”, the more the signal is amplified before being converted to digital.

The H2 also has another recording “level” control. This one is a continuous scale from (if I recall correctly) 0 to 100. This control adjusts the level after it has been converted to digital.

The potential problem:
If the three position “gain” switch is set too high, then the signal from the mic can overpower the A/D (analogue to digital) converter, and the resulting digital signal will be clipped. You can then turn down the (second) “level” control, and the recording level will look fine on the meter. The already clipped signal is reduced in amplitude, but it is still clipped.

One way or another, your recording shows too much gain before the A/D converter.

Yes, the DR-100mkII also has a three position gain switch (on high in this case), so that would make perfect sense.

You had to put it into a Faraday cage.

e.g wrapping Aluminium around it (caution with power plugs, better to first put isolating material around it or to use a shoe carton and the aluminium around that).
Or perhaps, you have a metallic biscuit box.
I sincerely hope that the cable company will do something.


I’ve had a chance to try some new gear. Here’s a recording made with the lip mic on a Zoom h5 through a Triton Fethead preamp. Any comments?