Differential Coil Design for underground pipe locator?

[mike]

I want to locate some pipes in my yard.
I had the locator people in, but they stick to their own
stuff. City won't help with sewer and water lines on the property.

So, being cheap, I set out to do my own location.
I wound a coil on a ferrite C-core and stuck it into the microphone
input on my Dell Axim X51v PDA.
The coil sensitivity peaks around 5kHz., and the utility was using 8kHz.,
so I figgered, "close enough". Clamped the function generator onto the
pipe and ground rod.
The FFT running on the PDA separates the signal nicely and has huge
dynamic range.

Works, but the detection path is very wide.

Ok, so need two coils in series to null at the pipe.
Problem is that I can't find any info on the actual physical
arrangement of the coils to accomplish this.

Everything I found was about crack detection in metals.

I'd rather not reinvent the wheel.
Anybody have any suggestions on how to construct a practical
differential current probe for pipe location?

Thanks, mike

 

[Jamie]

I want to locate some pipes in my yard.
I had the locator people in, but they stick to their own
stuff. City won't help with sewer and water lines on the property.

So, being cheap, I set out to do my own location.
I wound a coil on a ferrite C-core and stuck it into the microphone
input on my Dell Axim X51v PDA.
The coil sensitivity peaks around 5kHz., and the utility was using 8kHz.,
so I figgered, "close enough". Clamped the function generator onto the
pipe and ground rod.
The FFT running on the PDA separates the signal nicely and has huge
dynamic range.

Works, but the detection path is very wide.

Ok, so need two coils in series to null at the pipe.
Problem is that I can't find any info on the actual physical
arrangement of the coils to accomplish this.

Everything I found was about crack detection in metals.

I'd rather not reinvent the wheel.
Anybody have any suggestions on how to construct a practical
differential current probe for pipe location?

Thanks, mike

http://en.wikipedia.org/wiki/Helmholtz_coil

does that help?

P.S.
Trying to imagine how you induced enough energy in your pipes to
detect that ?

Jamie

 

[mike]

Use I-core, not C-core. A ferrite antenna rod or 1-inch inductor good
enough. Put the core vertically. It would give pretty clean null locate.

The operating frequency is not very important; anything from hundreds Hz
to tens of kHz would work for this simple purpose.

BTW I design electronics for utility location and horizontal driling
machines.

Vladimir Vassilevsky
DSP and Mixed Signal Designs
www.abvolt.com

Thanks for the input.
I have new info.
I borrowed a HP/Delcon 4904A with 18042A buried cable coil.
It did a wonderful job finding my water pipe 12" deep in the ground.
I tried it on the deeper iron sewer pipe and got no useful information
at all.
No peaks or nulls anywhere. Just a general signal everywhere.
I moved the ground rod around. Made little difference.

The test frequency is 990 Hz.

So, I put the coil on the bench.
There's no measurable resistance or inductance, so must be cap coupled.
Measures 0.1uF.

I excited it with a wire. There's a fairly sharp resonance at 17.25 kHz.
Almost no response at the 1kHz. test frequency.

It appears that there are two coils. One at the probe tip.
Another 8" higher up the pole. They're 180 out of phase.
The top one seems to have 10% or so more gain than the bottom one.

I'll start with your suggestion of a single vertical coil and see where
that goes.

Thanks, mike

 

[George Herold]

Thanks for the input.
I have new info.
I borrowed a HP/Delcon 4904A with  18042A buried cable coil.
It did a wonderful job finding my water pipe 12" deep in the ground.
I tried it on the deeper iron sewer pipe and got no useful information
at all.
No peaks or nulls anywhere.  Just a general signal everywhere.
I moved the ground rod around.  Made little difference.

The test frequency is 990 Hz.

So, I put the coil on the bench.
There's no measurable resistance or inductance, so must be cap coupled.
Measures 0.1uF.

I excited it with a wire.  There's a fairly sharp resonance at 17.25 kHz.
Almost no response at the 1kHz. test frequency.

It appears that there are two coils.  One at the probe tip.
Another 8" higher up the pole.  They're 180 out of phase.
The top one seems to have 10% or so more gain than the bottom one.

I'll start with your suggestion of a single vertical coil and see where
that goes.

There's a local interfernce 'noise' canceling scheme used in Earth's
field NMR, where you have two coils, out of phase, with equal turn's
area. (Turns x area) (So a small, tight, many turn coil about the
sample. And a bigger, few turn, one to cancel local pickup.)

George H.

 

[miso]

I have no idea where you live, but did you try "call before you dig"?

http://www.call811.com/

They come out and "tag" your property, as in some sort of biodegradable
spray paint.

Otherwise, you can rent pipe detection gear.

 

[mike]

I have no idea where you live, but did you try "call before you dig"?

They come out and "tag" your property, as in some sort of biodegradable
spray paint.

Nope, the city says they have no pipes on the property and any water/sewer
on the property is MY responsibility, not theirs.

Otherwise, you can rent pipe detection gear.

That's not the issue...I want to improve the one I built.
This is electronics.design, not equipment.rent. ;-)

 

[miso]

Nope, the city says they have no pipes on the property and any water/sewer
on the property is MY responsibility, not theirs.

That's not the issue...I want to improve the one I built.
This is electronics.design, not equipment.rent. ;-)

But "Call before you dig" isn't the city, at least for me.

When I have a job to do, I just do it. I don't build spectrum analyzers,
oscilloscopes, etc.