Circuit Ideas?

Small antennas never work well.

ideally you want several hundered turns of fine enameled wire on a form with the largest cross-section practical, (fit the largest dimensions of the case you plan to use.)

build an 8.5Khz linear oscilator (eg: twin tee) and turn the loop gain down until it stops, then couple the antenna to the input

otoh you could probably connect the antenna to a DAC and do the detection in software.

what sort of coding, and what symbol rate it makes a difference to the max Q you can use if you use a tuned circuit.

It's not clear just what the geometry of your transmitting antenna is. Is it a single conductor 100-300m long, ie. about 80 metres diameter, or a smaller diameter with multiple turns?

A lot depends on the current in your transmitting loop, something you haven't told us.

With a bit more info people here should be able ot provide oyu with better advice.

This smells something like the wireless heart-rate and/or bicycle sensors and associated receiver watch.

Make sure the "antenna" inductor is structured like a tuned circuit around the transmitted carrier to at least attempt to attenuate other unwanted signals.

From there, treat it like audio, though exactly how it would depend on how the signal is carried before you can know how to condition it from there.

I've only observed these signals in the past, I hadn't had to do anything with them, so I can't offer too much past there.

It's a receiver that will be used with existing equipment. The equipment being for domestic animal containment. A length of single core wire is installed around the property perimeter. The animal (cat/dog) wears a collar that is activated when brought within about 300mm of any point of the wire. The wire 'loop antenna' therefore has no fixed regular shape. Unfortunately I have no control over the design of the Tx as that already exists. All I have access to are the pulses (protocol if you like, transmitted by the Tx). The Tx transmits burst of pulses 187 every 15mS. The Tx output is variable. The output comprises a TDA2003 audio amp. The signal measured at the output of the Tx is 3v P-P into a 5ohm load, about 210mA. I have yet to measure the resistance of a typical 300m 'antenna'.

Tried the audio approach with limited success. Tried opamps MCP6002. The other point is power consumption - battery life is important. On the micro side its OK, I obtain 80uA standby by placing the micro into sleep mode.

Le Big Mac!

Thanks for your input.

Agreed, unfortunately these are the parameters in which I have to work, I have no choice. The phsical size of the choke is important to the construction of the design.

I like this osc idea of yours, I think I will give that a try....

Will give this one some thought - not enough signal directly from the antenna, thats why I need to front end to amplify the received signal and use it to provide a clean signal to the micro.

It's a receiver that will be used with existing equipment. The equipment being for domestic animal containment. A length of single core wire is installed around the property perimeter. The animal (cat/dog) wears a collar that is activated when brought within about 300mm of any point of the wire. The wire 'loop antenna' therefore has no fixed regular shape. Unfortunately I have no control over the design of the Tx as that already exists. All I have access to are the pulses (protocol if you like, transmitted by the Tx). The Tx transmits burst of pulses 187 every 15mS. The Tx output is variable. The output comprises a TDA2003 audio amp. The signal measured at the output of the Tx is 3v P-P into a 5ohm load, about 210mA. I have yet to measure the resistance of a typical 300m 'antenna'.

Thank you for you thought and help.

It's a receiver that will be used with existing equipment. The equipment being for domestic animal containment. A length of single core wire is installed around the property perimeter. The animal (cat/dog) wears a collar that is activated when brought within about 300mm of any point of the wire. The wire 'loop antenna' therefore has no fixed regular shape. Unfortunately I have no control over the design of the Tx as that already exists. All I have access to are the pulses (protocol if you like, transmitted by the Tx). The Tx transmits burst of pulses 187 every 15mS. The Tx output is variable. The output comprises a TDA2003 audio amp. The signal measured at the output of the Tx is 3v P-P into a 5ohm load, about 210mA. I have yet to measure the resistance of a typical 300m 'antenna'.

Thanks for you input - looking forward to your comments.

"SFD"

I doubt an inductor sold for use in SMPS is any good at all.

Best wind your own on a ferrite rod with *lots* of very fine wire - then find a cap value that tunes it to 8.5kHz .

Op amps with large GBW products and low input noise are never low current too, so you need to maximise the performance of that coil.

..... Phil

Thanks Phil, youre probably well aware of "The Trials and Tribulations of an Electronics Bod!" Your suggestion for pickup would be the ideal scenario, unfortunately the customer (bless their bank account!) almosts insists on using a small antenna. They have two existing similar devices working using a choke as the pickup. The choke they use is virtuall identical in shape and size as the one supplied by Electus/Jaycar type LF-1108 470uH - having said that, I have yet to remove a choke and measure its inductance. Trying to reverse engineer whats already been done is a no go due to epoxy encapsulation. Thanks for your input.

"SFD"

Hundreds of mHs, if possible.

This one from Element14 is getting closer.

.... Phil

Thanks for that info Phil, I have ordered a few pieces. Other similar chokes I'd seen in the mH range appeared physically too big - this is a good find! The spec indicates its inductance is measured at 79.6kHz, will be interesting to see its final in-circuit performance at 8.5kHz. Will keep you posted once they arrive and put a test cct put together. Cheers.

If you need to save power Phil's tuned circuit idea is better especially if the frequency is stable.

do a tuned curcuit and use the B-E junstion if a transistor as the detector C-E pulling a microcontroller pin down and a 100K pulling it up

that's a 12.5kHz repetition rate

Is that 187 bursts every 15 ms ?

where are you getting 8.5kHz from?

and what's the coding you wrote about?

basically I want to know the bandwidth of the signal

"Jasen Betts"

Jase has actually asked an insightful question.

Will wonders never cease...............

.... Phil

There are plenty of suggestions for receiver sensors in the other posts, I suggest you set up a test rig and try things out. There are many variables involved. But also be aware that when you move to the field, the level of interference you find may be unexpectedly high.

Your main challenge is likely to be achieving a bandwidth across your total signal path that allows you to transmit hte data you require. These magnetically coupled setups often depend on high Q to achieve sensitivity and noise reduction, but that can easily wreck the capability to send digital data.

... quite correct. I should have clarified. The existing Tx is 12.5 kHz. The Tx will be modified to 8.5kHz (187 x 20mS). These devices will be distributed in the US and apparantly have to be under 10kHz - so I am told.

The problem has been solved. Using a higher value sensing device (>10mH choke) into a FET and opamp to stabalize and clean the signal. Thanks for you input, appreciated.

Problem solved. Measured the choke used in existing device and found it to be 10mH. Made up a test cct, choke into FET, into dual copamp to stabalize and clean up the signal - perfect. Sensitivity will no doubt (hopefully) be better with the 27mH device. I was being over optimistic (stupid comes to mind) trying to get a 470uH choke to work, a few simple calculations would have told me otherwise. Oh well, when will we ever learn that shortcuts and assumptions often take longer!