Need ideas for low-noise portable preamp

[dasbin]

Hi guys,

I'm looking for suggestions. I'd like to build a mic preamp for nature recording; in particular I'm thinking of things that are not often recorded due to the technical challenge - little bugs and such. In other words, I need massive gain, extraordinarily low noise, low power consumption (need to run off batteries), phantom power output, and small size.
There are some very low-noise opamps out there that I was thinking of building around. But I've never designed anything before and I'd like some pointers before I start, if you guys have any.
Thanks a million!

 

[PRR]

This problem probably has to start at the microphone. Do you have one picked out?

 

[NewYorkDave]

There's a worked design for a low-noise, very high-gain, battery-powered AF amplifier in the " low power design" chapter of The Art of Electronics. If you don't already own a copy, look for it at your local library or bookstore.

 

[dasbin]

[quote author="PRR"]This problem probably has to start at the microphone. Do you have one picked out?[/quote]

Until I can find anything with a better noise-to-price ratio, I was thinking the Rode NT1A.

 

[PRR]

> I was thinking the Rode NT1A.

An amazing spec for any price: 5dB SPL???

Sensitivity is 25mV for 94dB SPL. So the mike self-noise voltage is 0.88 microVolts. That's not an amazingly tough spec to meet.

Mike power, being Phantom, is probably up in the 200mW range. And I assume the preamp load is a line-level input, 10K or more. It sure should be possible to make a preamp that eats less power than this mike.

How loud is a bug? I could imagine working down to 15dB SPL peaks. Mike output is about 3 microvolts. If we wanted to boost this to "+4VU, we need a gain of 300,000 or 109dB. That's insane.

In fact I think you will have to use the deep dynamic range of modern digital recorders, let the bug-peaks sit far below "0VU", and boost it in the editing. But we do not want to get down in the digital grit. I think if the analog noise (air and electronic) is 10 or 20 dB above the digital noise, you get nearly full resolution.

If a recorder has 16-bit 96dB digital dynamic range, and 200mV input sensitivity, the digital noise level is equivalant to 3 microVolts at the input. If we want analog noise of about 1 uV to be 20dB higher than digital noise of 3uV, we need a minimum gain of 30, or 32dB, which is easy.

I have also heard swarms of insects exceed 70dB SPL. To put this on tape/disk at -20VU you only need gain of 20dB. In fact you could just use a 1:10 transformer, zero power.

With an amp, a gain of 1,000 or 60dB is quite practical, and gives 0VU for sound of 52dB SPL, -20VU for sound of 32dB SPL, -40VU for sound of 12dB SPL, and mike self-noise reads about -33VU, far above digital noise.

A 1:10 input transformer gives gain without power, but has weight and cost. And you still need some electronics.

The NT1A has direct-coupled (transformerless) output, you probably use short cables and typically FAR from all electrical noise. I would try a transformerless unbalanced preamp. A single NE5534 can meet the noise spec, and can give a gain of 1,000 with good bandwidth and distortion, with power consumption no greater than the mike.

 

[Gus]

Does anyone have a picture of the inside of a NT1a? I own a nt1000 that I have some of the circuit traced. If the circuits are the same in both microphones you might be able to get more gain from the NT1a. The case bottoms looks close in both microphones


My nt1000 looks to be a charge amp design about 100pf of capsule cap and what looks like to be a 100pf feedback cap, for an overall gain of 1.

If you look in the Neumann PDF microphones you will see what look toi be the charge amp circuits used in the KM84 and Fet47. The nt1000 has a zener regulated 15V for the gain circuit. You might be able to up the microphone gain to x2 or x3 with reducing the feedback cap value to 47pf or 33pf. This would help with the gain you might need from the external microphone preamp.

 

[CJ]

Next question now that we know the mic, what are you recording with?
Sony mini-disk, etc.

 

[dasbin]

What am I recording with? I have yet to answer this question myself definitively, but, for the moment I am *planning* on using using a PDA and the Core Sound SPDIF system along with their portable 24-bit ADC (Mic2496). This ADC also has a built-in preamp that is extremely low noise, but it simply doesn't provide the gain I need (only 54 db, I am thinking I will need at least in the 80's). If I can rip out their preamp from this box, and replace it using my own design running off the 9 volt battery, all the better. I could most likely keep their phantom section then too.
This is of course after I get a whole plan figured out here and have the other gear ready, not to mention the money to dive into this venture.

The other option, which I am seriously considering, is to build my own portable ADC and bundle it with this preamp idea. And of course use it with the Core Sound SPDIF PDA interface. This would probably be the most economical. The only problem is I don't trust my own skills enough to build an ADC board from scratch. I could use an eval board, but I doubt any of them run on the power levels I'm thinking of here.

 

[CJ]

If you do a Yahoo search on Field Recording, you will get a ton of hits.
I have a couple of links at home I will grab tomorrow.

 

[Mbira]

When I did field recordings in Zimbabwe a few years ago, I used a Shure FP-33. I really loved that thing. I sold it when I got back, but it was built LIKE A TANK. It lasted 2 years in the bush and never had a problem. It runs on 2 9 volts.Who knows, maybe someone here will tell me that they suck, but it felt very transparent to me.

Joel

 

[NewYorkDave]

I've used those Shure boxes in TV remote trucks. I can't vouch for the fine points of their sound--we weren't overly concerned with "silky top end" in the news business--but I can confirm that they are indeed rugged and reliable. Shure also makes a headphone amp in the same format that's proved to be quite handy at my present job.

[EDIT]: Whoops, I just checked the Shure site, and I was thinking of the FP22 and another box that was probably the predecessor to the FP23. The preamp I remember looked more like the FP22 headphone box and used 9V batteries. From looking at Shure's discontinued products page, I think it may have been the FP11.

 

[dasbin]

Ruggedness and the things that usually make a good field device are not at the top of my list. I'm looking to branch out very specifically into recording quiet nature sounds, many of which are at the threshold of hearing or possibly lower. This is a field that few people enter into, so to get the best gear for a reasonable price, I kind of have to hack my way into it rather than taking the usual field-recording suggestions.
Thanks for the advice on the Shure anyways, though; I appreciate it.

 

[Kev]

A look at the way some classical Mic-pres are made might help.
The Mic itself might be the most important. Perhaps something very directional and designed for getting in close is needed.

A small pre-amp mounted in the Mic right at the capsule to give say 20dB fixed gain and then the 70+ at a special battery Mic-pre. You can develop your own battery pack and use whatever voltage suits.

PZM Mic and get the bug to walk over the plate ??? :green:

One problem could be that the environment may not be quiet enough to run the very high gains. A rain forest is a very LOUD place at times and I fear you will never be able to record a BUG stamping his feet even if he does have SIX.

 

[Rob Flinn]

When I did field recordings in Zimbabwe a few years ago

Now I know why you call yourself Mbira (or thumb piano to the uneducated). Do you play an Mbira ?