Hey all, I'm having trouble approaching how to detect a low frequency underground. 512hz. It's the standard transmitter frequency for sewer line cameras. I was looking at field strength meter type applications. Or perhaps electret microphones? I could use a little nudge in the right direction. Should I use an OP amp to ADC?
Post a link to the camera data sheet.
jremington:
Post a link to the camera data sheet.
I can't really find a datasheet, but this is what they look like.
Here is a patent for a detector that has a lot of info on how they work.
The patent describes it as a detecting electromagnetic signals from the transmitter device @ 512hz (Hz, not KHz). I think I need to make an antenna, an amplifier, a low pass filter, and feed it to a ADC and do some sort of spectrum analysis to get the amplitude of the frequency I'm looking for. I've ordered a teensy 3.5 to see if I can use the audio library.
I'm not sure of the antenna/amplifier/filter circuit. Should I use an op-amp? What do you guys think is the best way of amplifying an antenna for low frequencies?
So today I had chance to to put together an LM386 audio amplifier using a simple high gain circuit i found. I used just a piece of wire for an input and put it it near the camera/transmitter. With just an open wire I could not detect any frequency from the transmitter, just a lot of noise. When i coiled the antenna and grounded it out, I could detect the signal very clearly with no noise but from only 3-4 inches away. I assuming I need to look into inductive antennas.
Why not use the detector that is designed for that purpose?
jremington:
Why not use the detector that is designed for that purpose?
Because I get crazy ideas and I have to see if I can do it. Plus they are like $1200.
Firejoe22:
Because I get crazy ideas and I have to see if I can do it. Plus they are like $1200.
No problems trying it but it is not really an arduino question.
They cost 1200 for a reason, and that is a relativly cheap one.
You might be better off using something like an LM567 tone decoder.
Its a very difficult task.
You need a resonant antenna , an extremely low noise amplifier and a extremely narrow band filter, which are not beginner projects.
There is no way you can get an image from a 521Hz camara in anything like real time. You simply do not have any bandwidth to be able to modulate that quickly. So either this frequency is wrong or the image takes a long time to send. You can't even send voice signals over that frequency.
Grumpy_Mike:
There is no way you can get an image from a 521Hz camara in anything like real time. You simply do not have any bandwidth to be able to modulate that quickly. So either this frequency is wrong or the image takes a long time to send. You can't even send voice signals over that frequency.
I think the camera sends video signals back down its cable and the 521Hz signal is so you can find where the camera is underground to start digging a hole.
The patent clearly describes an open magnetic core driven by an oscillator at eg 512 Hz.
ie it emits a magnetic field.
This makes things much easier.
To detect it you need a large diameter resonant coil (a bit like a metal detector) with a low noise amplifier
and a notch filter.
Should be doable
eg a coil of 300mm diameter with 400 turns gives about 100mH . Resonate with about 1mF
A coil with a ferrite (eg a rod ) core could be much smaller and may be effective.
At those frequencies even soft iron might do.... or mumetal/permalloy
regards
Allan
That makes a lot more sense.
For communicating underground you normally use the magnetic component of a signal. Much used in speleology Communications in Caves
Thanks for the responses. I really appreciate the feedback. I know its not something that I can build a well functioning replica of, but I would love to experiment with a proof of concept. Detecting a signal several feet underground is looking impossible right now. I think I'm going to experiment with metal detection circuits/coils and see what sort of signal I can pick up with them. I have been playing with a teensy 3.5 and sending FFT data to my computer to look at the frequency spectrum and waterfall. So far its way too noisy and the detection distance is abysmal. Thanks again. If I make any progress I will post updates.
metal detectors work at a far higher freq than 500 Hz. You need a resonant coil
Allan
allanhurst:
metal detectors work at a far higher freq than 500 Hz. You need a resonant coilAllan
Thanks! I'll give it a shot.
sorry - miss-spelling. For 500Hz resonance a 100mH coil needs to be paralleled with about 1uF, not 1mF!
whoops....
Allan
Allan, thanks a lot for helping me out. I haven't had much time this week to fool with it, but I got my hands on some magnet wire and made the coil like you suggested. Worked fantastic! I was able to easily detect the 512hz using just a simple audio amp circuit from 4+ feet away. I have a lot of 60hz noise, but like you said I will need to make a filter. (I am new to analog signals as you probably guessed and dont own an oscilloscope.) When I get some time I'm going to see what kind of digital signal processing I can do on the arduino.
Firejoe22:
Thanks for the responses. I really appreciate the feedback. I know its not something that I can build a well functioning replica of, but I would love to experiment with a proof of concept. Detecting a signal several feet underground is looking impossible right now. I think I'm going to experiment with metal detection circuits/coils and see what sort of signal I can pick up with them. I have been playing with a teensy 3.5 and sending FFT data t
Sounds like a fun project , but very little practical use unless you do tracing for a living
I have done duct work (drains) and rent required when needed.
Equipment is not cheap.
Expect funny looks walking around with a big yellow handbag, unless you have a BT shirt on.