Continue to Site

Welcome to our site!

Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

  • Welcome to our site! Electro Tech is an online community (with over 170,000 members) who enjoy talking about and building electronic circuits, projects and gadgets. To participate you need to register. Registration is free. Click here to register now.

Way too much bass in my mic pre-amp

Status
Not open for further replies.

TheGuy

New Member
Heyup,

I know comparatively little about electronics, so please bear with me!

I've been working on a little project to record sound with my Zen MP3 player, which has the ability to record at 160kbs. However, it only has a line in, so to get sound from a mic, it needs to be amplified.

Now, I've got it pretty much sorted by using the pre-amp and amp circuit from one of those REALLY old Sony Walkmans, and a little circuit to power the electret capsules. Although this works very well, and the resulting sound is more than I could have hoped for, there is WAY (!) too much bass. Although it can be EQ'd in editing software, it makes the signal overload very easily.

Is it possible to make a simple high-pass filter for the mics before they enter the amp, and would it solve the problem?

Also, why exactly is this happening? Could the amp be boosting the low frequencies for some reason?

Thanks!
 
If you were using a self-built preamp for which you have a schematic diagram, tailoring the frequency response would be simple. Since you likely do not have a schematic of what you are using, it is very difficult to suggest a fix.
 
I've just done some tests and it's definitely the amp that's boosting the bass (they must have designed it for a cassette head that lacked a lot of bass or something). As you say, without a circuit diagram, I'm stuck. :( I'm guessing that it'll be extremely naive of me to wonder about using a photo... lol


So, I guess the only way around it is to make an extra strong high-pass filter for the mics. Are they reasonably simple to make effectively with some capacitors?
 
You're using a tape preamp, NOT a mike preamp, so it's completely wrong - it would be easier to build a proper mike preamp than to try and build a filter to correct the other, which would reduce quality as well. AudioGuru has posted a suitable mike preamp circuit here MANY times - and it's available as a 'sticky'

https://www.electro-tech-online.com/threads/basic-opamp-circuits.35258/
 
IIRC, magnetic heads (tape heads) have a voltage output that rises at 6dB/octave (with increasing frequency), which has to be compensated in the preamp by a 6dB/octave fall in response. If you drive the amp with a mic that has flat response, you will wind up with massive bass boost.
 
Last edited:
Creative Zen has voice recorder I thought :p, I use Transcend MP650 that has an excellent voice recorder with options like group/hall environments. Even there's a lots of voice recorder players are available at damn cheap rates.

You could try these kind of little crackers. Make it for battery operation as it won't consume much current. Assemble inside a tiny enclosure with battery. I'd say you can re-design it for even tiny and work on 3v button cell and it'll just look like a keychain connected to line-in sock.

These are very effective and extremely compatible and portable!
**broken link removed**
 
IIRC, magnetic heads (tape heads) have a voltage output that rises at 6dB/octave (with increasing frequency), which has to be compensated in the preamp by a 6dB/octave fall in response. If you drive the amp with a mic that has flat response, you will wind up with massive bass boost.

It's not as simple as that, it applies different levels of boost and cut at different frequencies, similar to the RIAA curve.

Like I said, dump it and build a proper preamp.
 
Microphone preamplifier

Hi TheGuy,

I made a nice PCB layout of audiguru's preamp. The board is purely single sided with a small cheat (a zero Ohm resistor to avoid a trace between between connector pins)

Board size is 43.5X40mm (1.7125X1.5625").

The electret microphone is located on board, but there are two large pads to connect to cable if necessary.

Here are some screenshots: schematic, PCB-layout, layout with ground fill and silk screen.

If you're interested in building the board please PM me for the Eagle files.

You can download a free Eagle version from CadSoft Online: EAGLE Layout Editor

Regards

Boncuk
 

Attachments

  • ELECTRET-MIC-AMP-SCH.gif
    ELECTRET-MIC-AMP-SCH.gif
    11.1 KB · Views: 879
  • ELECTRET-MIC-AMP-BRD.gif
    ELECTRET-MIC-AMP-BRD.gif
    22.3 KB · Views: 656
  • ELECTRET-MIC-AMP-FILL.gif
    ELECTRET-MIC-AMP-FILL.gif
    24.2 KB · Views: 472
  • ELECTRET-MIC-AMP-SILK.gif
    ELECTRET-MIC-AMP-SILK.gif
    15.1 KB · Views: 542
My, my, so many helpful replies! :D

You're using a tape preamp, NOT a mike preamp, so it's completely wrong - it would be easier to build a proper mike preamp than to try and build a filter to correct the other, which would reduce quality as well.


Ah. So much for a cheap short-cut then, haha! I guess I'll have to build one.

AudioGuru has posted a suitable mike preamp circuit here MANY times - and it's available as a 'sticky'

One of the reasons I chose to try a tape amp is that it amplifies the signal to a level that can take full advantage of the bandwidth that LINE-in offers (bandwidth might not be the correct term to use, hmm). Would this preamp circuit amplify the signal enough, or will I have to also build another amp to go after it? Also, is the signal to noise ratio... good?

IIRC, magnetic heads (tape heads) have a voltage output that rises at 6dB/octave (with increasing frequency), which has to be compensated in the preamp by a 6dB/octave fall in response. If you drive the amp with a mic that has flat response, you will wind up with massive bass boost.

It's not as simple as that, it applies different levels of boost and cut at different frequencies, similar to the RIAA curve.

Ahh, I see. Makes sense now!

Creative Zen has voice recorder I thought :p

Indeed it does. :p However, the quality in that mode SUCKS! 8Khz or something like that. Fine for voices, but I'm aiming for high-quality recordings. ;) The MP3 player you mentioned costs quite a bit too... £45 at least. I bought my Zen for £8 purely for this project, as it's the cheapest thing I can find with a decent recording bitrate.

@ Boncuk, that circuit certainly looks interesting. What's the output quality like, may I ask?
 
hey theguy, are you trying to done something on a budget? You might want to look into what kind of parts and techniques actually make things sound better. Are you hotwiring the mic input of the zen player? (I'm not familiar with the product) You may also be limited to a lower resolution because of the zen's recording processing, as I'm guessing it converts analog to digital at ten bits, and anti-aliases it to not sound as jagged. 10 bit audio doesnt sound too good, even if you used thousands of dollars worth of preamps. You might want to try hotwiring a preamp you know works for you so you can see what the player is actually capable of before getting too deep into it.
 
Oh, no, no, it's got line in, which records at CD quality (sorta... 128kbs is apparently "CD quality" to some people. This is 160kbs WMA, which is about twice as good as MP3 @ 128kbs). I've tested it, and damn, comparing it to my Zoom H2's 24bit 94khz wavs... to the casual listener there's no difference.

It's the only product in the price range with such a good quality input, which is why I chose it.
 
@ Boncuk, that circuit certainly looks interesting. What's the output quality like, may I ask?

Hi TheGuy,

I don't either know about output levels or about quality. Since the circuit was designed by audioguru I guess it has maximum performance.

audioguru stands for high quality AF-design. :)

An electret microphone has a wide frequency range and I guess building a board with those few parts won't be wasted money.

When building the circuit I suggest to use metal film resistors at 1% tolerance and metalized polyester capacitors (WIMA MKS-types)

Turn VR1 until you get ~750mVpp at normal speech. That level is normally required by power amplifiers.

Boncuk
 
Last edited:
My mic preamp uses a fairly old and inexpensive audio opamp. It has fairly low noise (hiss), low distortion of 0.003% and a bandwidth to 100kHz.

An OPA134 opamp is newer and costs a little more. Its noise is 3.5 times less, its distortion is 38 times less and its bandwidth is 2.5 times higher.
You might not notice any difference.
 
I'm pretty sure no human being can hear 100KHz. :)
 
The bandwidth must be high so that the intermodulation distortion at high audio frequencies is not too high. The intermod distortion causes high audio frequencies to beat together and produce lower frequency audible distortion.
 
Ive always had issues with consumer equipment compressing everything I send it to all hell. i.e. my camera. So this thing doesn't compress then? Hummm...
What kind of stuff are you recording into it?
 
as for the Recording quality, the device may be doing something to the audio you like. I would guess the zoom is designed to record raw uncompressed files so you can have that type of control later.

Even if the zen doesn't compress, it is still enduring a pretty significant line of change to the audio signal.

First it captures it into digital
then it anti aliases it
then it converts it to a .kb/s format which is a compressed (not for audio, for file size)

The last step of compression to wma will have a significant effect on the sound. You will notice the highs are a little warmer but not as detailed. If you are working with acoustic guitar this may have a "Tube" sounding effect. It would also tone down the high digital edge of the recording device. This in combination with a bass-biased input signal would equate to a LOT of bass.

You may also want to investigate how the recording device clips a hotter input signal. What you do is take a signal generator, (you can get a program for your laptop) and run it in to the device and crank up the gain through clipping with a preamp. (if you have nothing, use a stereo) Get access to the recorded waveform and evaluate where it clips and whether it clips the peak signal by sheer limiting or top end compression. You'll know what you have if you jam it and you either see what looks like a square wave, or a scaled version of your signal. If you see the latter (everything is rounded off), making a preamp for the device may effect the sound but in reality you might just be forcing the device to compress the audio with it's own algorithm by overdriving it.
 
Last edited:
if you knew what was what with the tape preamp, you could find the frequency equalization components in the feedback loop (assuming it's an op amp in there somewhere) and replace them with a resistor. tape preamps and RIAA preamps have one thing in common, the use of two sets of parallel RC networks to tailor the frequency response to closely match a standard curve. removing the capacitors from the feedback loop will flatten the frequency response without altering the overall gain (but it will really boost the gain above 1khz compared to what it was).
 
Last edited:
Ive always had issues with consumer equipment compressing everything I send it to all hell. i.e. my camera. So this thing doesn't compress then? Hummm...
What kind of stuff are you recording into it?

Depends what you mean by "compression". If you mean levels compression, then no, the Zen doesn't compress. If you mean file compression using a codec, then yeah, it compresses, but not too much so that it becomes obvious. It'll be used for recording dialogue, so 160Kbs totally fine.

This isn't for professional use anyway, so I'm alright with a bit of audio degradation on the recording side of things. I'd have to spend much more than £8 to get something better. I doubt anyone would notice anyway, because, as I said before, even comparing an uncompressed wav from a professional recorder, the Zen sounds almost the same to the ear.

then it anti aliases it

Doesn't anti-aliasing take place only when re-sampling to another sample rate? I doubt the recorder oversamples, and then downsamples. I dunno though, I'm no pro :)

How did I start talking about the recorder anyway? :D This is meant to be about the amp...

So, audioguru, I'm going to take a crack at your electret mic preamp. I do have two questions before I start my electronic adventure, though.

Does your circuit provide power for the electrets? If I was any good at electronics I'd be able to see from the diagram, but I'm not, so... :D

Also, because the Zen has NO amplifying at all, it requires quite a loud input to be just under clipping. If you'll excuse the non-precise comparison, it requires something similar to an MP3 player on 2/3 volume. So, because your circuit is a pre amp, it'll also need another amplifier afterwards, correct? Or will your circuit amplify the mic quite a bit more than I'm thinking?

Thanks! :)

@ unclejed613: MikeMl said something similar. It'll be easier to build a new one, it seems, as I don't know what is what with the tape amp. :)


I've uploaded a sample .wma file from the recorder, just in case anyone is interested in a budget recorder but isn't sure on the quality. LINE-2009-04-09_06h55m11s.wma - File Shared from Box.net - Free Online File Storage
 
Status
Not open for further replies.

New Articles From Microcontroller Tips

Back
Top