Building an ULTRA sensitive PC MIC

[Peter_wadley]

By the way,

The manager thought the circuit was great

it was also my lucky day...

It turns out the store was getting rid of a large portion of their electronic components..

Long story short: I struck an under the table deal with the manager and paid $20 for over $200 worth of parts..

-100's of capacitors
- chokes, coils, and inductors
- 20 buzzers and chimes
- LEDs
- switches
- lamps
- bayonet holders
- fans
- stepper motors

It was comparable to christmas.. sorting through a huge box of brand new parts... picking whatever I felt like...

I have a question about 2 lights I grabbed..

They are XEXON STROBE bulbs and have 3 leads.. I have no clue how to hook them up!

Later
Peter Wadley

 

[Peter_wadley]

Oh yeah forgot to mention...

He gave me a 12v 200ma wall wart and said he didnt want to be switching 9v batteries...

So I knew right away I would need to deal with the buzzing that comes with these pieces of S***

When I first plugged it in, the circuit was unusable because of the noise.. so I ran it through a 7809ct 9v regulator.. this cleaned it up quite a bit but it still was humming..

I then added a .15uf bypass capacitor and it now sounds completely clean.

Is this the proper method of dealing with the hum?

Did I choose the proper size capacitor.. is there a formula (seems like there is formula for everthing!)..

Thanks again

 

[audioguru]

Look at Xenon Flash Circuit in Google. There are many links to circuits and articles.

 

[audioguru]

The hum from the cheap power supply was connected directly to the input of the opamp then it was amplified 101 times.
Reduce the hum with a resistor-capacitor filter before it reaches the opamp's input like this:

 

[Peter_wadley]

Is it the standard practice to have a small cap (.1uf) and a large cap (100uf) for bypass filtering loud supplies?

I seem to always see .1uf and 100uf..

Does the .1uf rectify small changes in voltage whereas the 100uf rectify the larger change/waves of voltage change?

hope my that makes sense :|

Edit: Does the .1uf cap that is on the opamp go to ground on the other side, thanks!

 

[audioguru]

100uF is a good filter for low frequencies. It is electrolytic and is made by winding an insulated foil around and around that adds inductance which makes it a poor filter of high frequencies.
A 0.1uF ceramic disc capacitor has low inductance so it is a good filter for high frequencies. Many voltage regulator ICs have a 0.22uF ceramic disc capacitor in parallel with a big electrolytic capacitor at their input.
A 1000pF ceramic disc capacitor is a good filter for very high radio frequencies. Many radio circuits have three capacitors in parallel across their suppliy with the three different values.

 

[Peter_wadley]

Thank you for clarifying audio!

 

[TheVictim]

One of the smaller modules I have for my prototyping system consists of a 100uF, .1uF, and .001uF cap in parallel mounted in a paint pot. I use it whenever I'm testing any kind of audio amplifying circuit.

 

[Peter_wadley]

Thanks for sharing!

 

[ozarkshermit]

Hi all:

I'm responding to this rather old thread, since I need a good electret AGC preamp. After much checking, I think I'll go with the circuit developed by Audioguru in post # 15.

I have developed a small circuit board (with the modifications shown), using the TL071. I'm ready to etch the board, but don't want to make something no better than what I presently have. I am interfacing the mic to a speech recognition chip, (HM2007), and have been using a small pre amp I made using a LM358, which works OK, but does not have AGC. I'm not looking for "super sensitivity", just something that will allow me to speak with slightly different voice levels. As I said, the circuit I am using works OK, but it's hard to speak in a consistent volume, which can sometimes result in recognition errors. I plan on using a POT for R3 (100K fixed resistor in the circuit as it is shown), to adjust and "tune" the output to the HM2007. The Mic was also picking up a lot of electrical interference before I added a shielded cable, and that seemed to cure most of those problems. I have to add a 9 volt regulator to the circuit, since I do not have that voltage available. I will heavily filter and de-couple the regulator part, so hopefully that will not present problems.

Just looking for any other advice from the experts. Digital I can handle, but I'm not the best at analog.

Ken

 

[audioguru]

I selected a TL071 opamp because it is low noise, has very low distortion and has a bandwidth that is high enough for hi-fi audio.
An LM358 is designed for low power so it has a lot of noise (hisssss), a lot of crossover distortion and a bandwidth narrow like a telephone.
An opamp does not need a regulated power supply. Filter its reference voltage instead with a resistor and capacitor.

 

[Torben]

Hi all:

I'm responding to this rather old thread, since I need a good electret AGC preamp. After much checking, I think I'll go with the circuit developed by Audioguru in post # 15.

I have developed a small circuit board (with the modifications shown), using the TL071. I'm ready to etch the board, but don't want to make something no better than what I presently have. I am interfacing the mic to a speech recognition chip, (HM2007), and have been using a small pre amp I made using a LM358, which works OK, but does not have AGC. I'm not looking for "super sensitivity", just something that will allow me to speak with slightly different voice levels. As I said, the circuit I am using works OK, but it's hard to speak in a consistent volume, which can sometimes result in recognition errors. I plan on using a POT for R3 (100K fixed resistor in the circuit as it is shown), to adjust and "tune" the output to the HM2007. The Mic was also picking up a lot of electrical interference before I added a shielded cable, and that seemed to cure most of those problems. I have to add a 9 volt regulator to the circuit, since I do not have that voltage available. I will heavily filter and de-couple the regulator part, so hopefully that will not present problems.

Just looking for any other advice from the experts. Digital I can handle, but I'm not the best at analog.

Ken

Hi Ken,

I thought you needed a preamp with AGC? This circuit has no AGC--it's just an electret preamp. It is a nice sounding and very sensitive little unit but yeah--no AGC at all. The circuit you're talking about was my first try at an electret preamp which I posted a few years back and Audioguru corrected (see Peter's link to the original thread). I just never got around to posting the corrected circuit after Audioguru told me which bits I'd stuffed up.


Regards,

Torben

 

[ozarkshermit]

Torben:

Thanks - - - after I looked more closely at the circuit I realized it was alost identical to the one I used with the LM358 and yes, I definitely want something with AGC.

I've been surfing till my eyes age glazing, and have found a few circuits with AGC, (not sure how good they are) but now am considering a MAXIM 9814 - it looks like it would work. Only problem is the surface mount package, but then it's about time I started trying my hand on making PC boards for them.
After I digest some of this for a while I'll decide what to do hopefully.

Ken

 

[ozarkshermit]

On second thought, after looking more closely at the dimensions of the MAX chip, there is no way I could use it - I don't have a microscope !

Ken

 

[audioguru]

Look in Google for Audio Compressor Circuit. There used to be NE570, NE571 and NE572 ICs that worked well but they are obsolete. Most discrete circuits use a Jfet to attenuate signals that are too loud.

 

[ozarkshermit]

Audioguru:

I managed to locate some NE571's on eBay, might order some - but am leaning more to using the circuit using a TL072 with the FET and associated components. The only problem (for me) would be the split 15 V supply. Since this is being added to an existing circuit, I do not have the liberty of split supplies. Is it possible to use that design with a single supply?

Ken

 

[Torben]

Audioguru:

I managed to locate some NE571's on eBay, might order some - but am leaning more to using the circuit using a TL072 with the FET and associated components. The only problem (for me) would be the split 15 V supply. Since this is being added to an existing circuit, I do not have the liberty of split supplies. Is it possible to use that design with a single supply?

Ken

Hi Ken,

When you say ". . .the circuit using a TL072 with the FET and associated components", what circuit do you mean?

You can indeed use a TL072 on a single supply--that's what the circuit from post #15 does, for instance (well, with a TL071 which I had in there but but that's just half a TL072*). In fact there's a sticky on this very topic which includes Audioguru's famous diagram explaining how it's done: https://www.electro-tech-online.com/threads/basic-opamp-circuits.35258/

Do you already have a schematic for the AGC portion?

* -- Truth be told I wound up using a 5532 which is why there was a compensation cap on the op-amp--Audioguru crossed that cap off the diagram though since a TL071/81 doesn't need it.


Cheers,

Torben

 

[audioguru]

Here is a hard limiter circuit that uses a single supply voltage. It might not work if the Jfet is at its spec limits. I would use a Jfet with tighter limits like a 2N5485. This circuit is not optimised for low distortion.

 

[ozarkshermit]

Torben:

Here is the circuit I was referring to

**broken link removed**

It was highly recommended by Audioguru

Ken

 

[anjana.larka]

My circuit is not working : (

dear audio guru,

i have made this circuit as per schematics with TL071 and ur modifications, completed last night. Circuit seems working ok ass pre amp but it is not sensitive as i have to eat mic to get sound from my amplifier.

Circuit images are attached here with if you plz shed some light on it...

Basically i want to use it with my computer so that it can pic my voice commands from some distance but it is not working in that way : (

can i use simple a cell phone mic or a hand free mic in palce of this whole circuit.. i am thinking in this way ... but this circuit sould work what you say

thanks in advnace...