building a very sensitive listening device

[speeedy6]

Hi group,
My name's Mark and I just found this forum and it sounds exactly like what I need.
I'm looking to build a portable ,frequency specific listening device like the Gen-ear made by General Wire Spring Co. for finding water leaks underground.These things seem incredibly expensive for what they are.
I originally began looking at portable headphone amps typically used to plug into an electric guitar and go directly to headphones for private playing and, also a guitar equalizer pedal for effects, to filter the input.But,the pedal would only enhance the frequency I wanted while deminishing the frequency of the sound I don't want.
The Gen-Ear filters ALL frequency below 100hz and above 1200hz along with a press to hear button for safety.I would also like to incorporate a VU meter and possibly a portable audio spectrum analyzer to find peak signals along the line.
I graduated from an electronics school about 30yrs ago.Do you even use transistors anymore? I can read a simple schematic, solder like crazy and even fabricate a nice,sturdy container for all this.But, I could really use some help on the particulars of this project.
Thanks,Mark

 

[ronsimpson]

Very interesting product. I could use one. Are you building one or 100?
I think the microphone is very important.
Any audio amplifier could be modified to get the 100-1,000hz effect. We could help with that.
I would add a "limiter" effect, (automatic volume control) so very loud sounds will be reduced. The VU meter should not have the AVC function before it.

 

[Dragon Tamer]

You would be best off having an adjustable band pass filter on the outpu of the mic going into the amp. I would us an active band pass filter becuse it will be more acurate and easier to tune. It sounds like it'll need to be a pretty sensitive mic to detect a water leak. It would really help us if we knew what kind of dB level we are talking about so we know what kind of gain to put on the amplifier.

 

[ronv]

Seems like the magic is in the microphone or pick up. What do you have in mind for this? Then we can come up with an amplifier and bandpass filter.

 

[speeedy6]

I'm planning on using a peizo disc element for the mic.That seems to be what the expensive ,purpose built units use. A 43mm fixed into an aluminum puck,drilled and tapped for a probe rod and surrounded by a thick silicon cup to block above ground noise. A 43mm disc is the largest I've found,I guess bigger is better?
I REALLY appreciate the input but, please keep in mind that while I'm comfortable building such a unit , The terminology can get over my head quickly. I've been to Wikapedia several times already. So, go slow with me.
Thanks again , Mark

 

[ericgibbs]

I'm planning on using a peizo disc element for the mic.That seems to be what the expensive ,purpose built units use. A 43mm fixed into an aluminum puck,drilled and tapped for a probe rod and surrounded by a thick silicon cup to block above ground noise. A 43mm disc is the largest I've found,I guess bigger is better?
I REALLY appreciate the input but, please keep in mind that while I'm comfortable building such a unit , The terminology can get over my head quickly. I've been to Wikapedia several times already. So, go slow with me.
Thanks again , Mark

hi,
Download the free Audacity Beta (Unicode). program
You can record sounds then use the in built spectrum analyser.
Clamp your mike to the house cold water pipe and run the tap....

 

[speeedy6]

Eric,
Are you saying I could use my laptop with this program and see my signal in realtime using the spectrum analyser ? Would it also have a notch filter function or is a notch filter program available also ? It's beginning to look like I can do all I need with my laptop with the right programs !
Thanks, Mark

 

[ericgibbs]

Eric,
Are you saying I could use my laptop with this program and see my signal in realtime using the spectrum analyser ? Would it also have a notch filter function or is a notch filter program available also ? It's beginning to look like I can do all I need with my laptop with the right programs !
Thanks, Mark

hi,
This is the way I would start experimenting with 'leak' noise.
Handy point is that you can record leak noise and at later time play back the leak noise into any hardware you build.

I used continuous loop cassette tapes with recorded 'standard leak noises' you could buy ready made tapes... once.

EDIT.
The method I used for underground water pipes in the highway was 'leak noise correlation' , two noise sensors about 200mtrs apart


EDIT2:
Look here.
**broken link removed**

 

[OldTechie]

I found this link to be very useful a few weeks ago when I was researching the same issue. How to Find Leaks
My brother had a leak in a water supply line behind his house that runs nearly 1000 feet. I built a very high gain audio amp (similar to the "Big Ear" mike preamp; I added an additional 20db of gain before the Big Ear amp), glued an electret mike to the side of a 3-foot piece of steel grounding rod, connected the mike to the amp, put on the headphones and quite easily found the leak. The info on the web page I listed helped greatly.
The first place we dug into the ground was only 4 feet away from the leak. Not bad for an amateur leak finder.

Cheers,
Dave M

 

[speeedy6]

Dave !
You are the one I've been looking for ! Could you please send me a schematic of your amp and possibly include built in filters to only pass 100hz thru 1200hz ?
Thank you, Mark

 

[speeedy6]

Uh,hello ?

 

[ronv]

Leak Amp

Here is one I think will work. It uses the electret mic idea. I think you can get the parts at Radio Shack. It has very high gain so watch your ears.

R6 and R7 represent a pot to adjust the sensitivity. The center (wiper goes to the amp)
You will need to google the data-sheets to get the pins for the ICs.

 

[audioguru]

An LM358 opamp has too much hissss to be used as a mic preamp with high gain. It will sound like a leak all the time.
Use a low noise audio opamp instead.
The LM386 amplifier also has a fairly high hisss level when its gain is 200 so reduce its gain to 20. Then use a second low noise audio opamp for more gain of have the preamp provide more gain.

The inverting opamp has a low input impedance that loads down the electret mic and has gain that is much lower than you think. So use a non-inverting opamp preamp with an input impedance of 20k to 100k ohms.

 

[ronv]

Amplifier

Use a low noise audio opamp instead.

How about a TL071?

Then use a second low noise audio opamp for more gain of have the preamp provide more gain.

What would you recommend?

The inverting opamp has a low input impedance that loads down the electret mic and has gain that is much lower than you think. So use a non-inverting opamp preamp with an input impedance of 20k to 100k ohms.

I think what you mean is the mic has an output impedance of 10K (the supply resistor) so like a 3x reduction?

 

[audioguru]

How about a TL071?

A TL071 is inexpensive so its noise level is not as low as a better more expensive opamp like an OPA134. But it has much less noise than a lousy old LM358.
The minimum supply for a TL071 is 7V but I have powered many from a 9V battery and they worked fine when the battery dropped to 6V.

What would you recommend?

A TL072 is a dual inexpensive low noise opamp. Each one can have a gain of 20 and the LM386 amp can also have a gain of 20. Then the total gain is 8000 which is plenty.

I think what you mean is the mic has an output impedance of 10K (the supply resistor) so like a 3x reduction?

I measured an electret mic output impedance to be 4.7k ohms. In parallel with the 10k resistor that powers it its total impedance is 3.2k ohms. If it feeds an inverting opamp that has a 3.3k ohms input resistor then the level from the mic is reduced to about half and the gain of the opamp is about half of what you think. Then the output level is 1/4 of what is needed.

 

[ronv]

What do you like for a power amp?

Speedy, Are you still looking?

 

[audioguru]

What do you like for a power amp?

An LM386 little amplifier IC with a gain of 20 can drive headphones pretty well.

 

[ronv]

Leak2

Here is another shot with some better amplifiers and some more testing. I'll send you a PM with e-mail if you would like me to send you a parts list you can order from mouser in the US.

 

[audioguru]

Here is another shot with some better amplifiers and some more testing.

R14 is not needed since the OPA2134 has Jfet inputs that have no input bias current.

 

[speeedy6]

Project '' LISTENING DEVICE '' Done !

O-K ,

I know it's been forever , about 10 months since I started but I've done other things,waited on other people and been just downright confused on what to do next . My listening device is complete and works great !
With a lot of help from Ronv and others , I got started with the parts and put together my first attempt. It went together pretty well and I couldn't wait to try it so I twisted together the battery , mic and headphones and turned it on . I taped the mic on my sink faucet and eagerly listened to ........ COUNTRY MUSIC ! I couldn't help but laugh but, the closer I listened the more I could hear the sound of the leak . I kept working on the case and installed the workings in a nice project box with proper connections and jacks installed .
I sent the whole project to a ham radio repair friend of mine that said he could help . He kept it for about 6 weeks with me chomping at the bit . When he sent it back I saw he had wrapped all the wires around their own chokes and was still no change in the outside noise. I did some investigating on my own and replaced the wires with shielded wires and put the works in an aluminum box instead of plastic and that took care of the music . I also tried to include a L.E.D. Vu meter I had bought off ebay. My idea was to display the volume of the signal indicating when I got closer to the leak. It did work but wasn't sensitive enough to work well , thats a project for a later time. I eventually replaced the shielded wires with ordinary stranded wires again as the shielded was very hard to work with and I had an idea they were causing some connection issues. My final solution was shielded mic wire and aluminum box. This first instrument became my '' try everything '' unit and I had wires and connections everywhere and I was quite confused with where I was at. So, I stripped everything off the project board and started fresh and put it all in the box . In the meantime, I had started building a second unit and completed it in a new box also. It works well too but , somehow different . I just attributed that to the fact that on my second component order some of the components were different .Please keep in mind I'm more of a blacksmith than a technition.
So now , I have two complete devices I'm eager to use.My mic I built also with a piezo pickup and it's built so I can probe the ground or sit it ontop of the surface such as a concrete slab.
This is just an update and I'll be asking some more questions to improve these units .
Thank you all for the help I've gotten , It's been a very interesting project. Speeedy.