Driving Tank Circuit at 125kHz

[lostengineer]

Alright guys,

I need a bit of help with a project I've been working on for a while. I need to make a circuit which can amplify a 125 kHz sine wave signal (around 1V pk-pk input) and place it on a magnetic loop antenna (big coil of wire). This is basically an RFID amplifier but no data needs to be transmitted; I only want to boost the 125 kHz signal and have it make a large magnetic field so it can excite a receiving circuit several feet away.

I'd like to use a pre-made amplifier IC to avoid designing a custom FET amplifier from scratch. A simple LM386 seems to have a bandwidth of 3MHz so I think it could work? How should I attach the output of the LM386 to my parallel resonant tank circuit? Do I need a matching network as the LM386 output is 8 ohms and I believe the tank has infinite impedance at resonance?

Please see attached image for a better view of what I'm trying to accomplish. Any thoughts are greatly appreciated.

[Benta]

I doubt if that will work. Gain-bandwidth product is one thing, full-power bandwidth is something else.
Looking at the LM386 datasheet, this is not specified, but the significant distortion rise towards 20 kHz makes me suspicious.

I'd look at a power opamp instead. Something like the TLE2301 (Texas).

Cheers.

[oldway]

You can drive it with a square waveform....It is a serie/parallel filter, no problem at all.
Don't forget you need to feed power enough to replace power losses of your tank circuit.

[orolo]

You could try to inject the energy into the resonator via inductive coupling. Theoretically you can accomplish that with a very low energy oscillator, depending on the Q of the resonator.

In the spice example, a 6V amplitude oscillator with less than 1mA current excites a 30V, 110mA current in the resonator via inductive coupling. L2 could be your air coil, and L1 and L3 two windings in an adequate toroid. Never tried this, but in theory it works.

[oldway]

Something is not very clear for me....The strength of the magnetic field must be proportional to the 125Kh signal or not ? Or it must only transmit the information of the presence of the 125Khz signal or not ?

[lostengineer]

The TLE2301 Op-Amp looks extremely promising. I'm definitely going to go this route for the amplifier section. I could likely use inductive coupling but I believe that is just using a transformer to isolate the amplifier from the tank circuit and maybe do some impedance matching? I'd like to avoid big magnetic components (transformers) if possible - this is why I was thinking of using a series inductor as an impedance matching element instead of other methods such as tapping the tank circuit inductor. See this link: http://www.richieburnett.co.uk/indheat.html

I am going to model this all in spice a bit later. again, I'm more concerned about the matching section and connection between the output of the op-amp and the tank circuit itself. Do I need any matching between the output and the tank at all? For this project, I'm basically trying to take an RFID signal carrier (125kHz wave) and amplify / transmit it in order to excite a 125kHz tag several feet away. No data needs to be transmitted but the receiving device must be able to detect the 125kHz field.

[Zero999]

You can drive it with a square waveform....It is a serie/parallel filter, no problem at all.

I agree. A squarewave is the most efficient way to drive this circuit. There's no need for sine wave source and a linear amplifier, which will be inefficient. If you already have a sine wave, then don't worry, you can still use it drive a transistor to inject pulses into the LC circuit.

[bktemp]

I would also go for a square wave driver: Your matching network acts as a low pass filter.
You could use any fast enough half bridge driver, like a cheap mosfet driver. They typically have a mosfet half-bridge with less than <10ohms, therefore can supply a couple of 100mArms even in small SO8 or DIP8 packages. That gives >1W output power at 12V.

[oldway]

You said : "No data needs to be transmitted but the receiving device must be able to detect the 125kHz field."
Why transmitting 125Khz and why transmitting it by a magnetic field ?
Is that used for a metal detector ?

[lostengineer]

Alright thanks for the help guys. I understand I could drive a couple fets or an H-bridge with a square wave to power the tank circuit. I also know I can take my sine wave and apply it directly to some fets. I believe I could also run the sine wave through a series of inverters to make it into the proposed square wave.

My real question is again: how can I properly match the output of the amplifier system (they all seem to have low <10 ohm output impedance) to the tank circuit? And how should I design the tank circuit for maximum magnetic field? Do I actually need the matching circuit at all?

I do have an application for this system. I'm trying to emulate the 125 kHz field produced by a car door keyless entry system for a security research project.

[SeanB]

You can use a 555 to act as a driver, it will drive the coil direct ( though you will have a lot of harmonics) and will drive at quite high current, plus it will act as a buffer for the drive quite well.

Just keep the matching network, and add local decoupling to the 555 right by the supply pins ( 1000uF 25V electrolytic) and the output will be usable for you. The drive only needs to be a square wave with 50% duty cycle, the filter and the LC of the coil will make it a sine wave for you, no need for any sine wave drive, the harmonics will not have too much effect, just like on the original one, which probably uses a square wave drive.

[bktemp]

My real question is again: how can I properly match the output of the amplifier system (they all seem to have low <10 ohm output impedance) to the tank circuit? And how should I design the tank circuit for maximum magnetic field? Do I actually need the matching circuit at all?

The matching circuit transforms the input impedance from a rather low impedance of the amplifier to the high impedance of the LC tank. For maximum magnetic field the voltage at the LC tank should be as high as possible, therefore you need the matching circuit to step up the voltage. Without a matching circuit you can only drive the LC tank with a voltage as high as your supply voltage (unless you got to a series LC tank).
The matching circuit needs to be adjustet to limit the power to not overload the amplifier: Larger L, smaller C -> higher impedance -> lower input power.

[Elf]

You said : "No data needs to be transmitted but the receiving device must be able to detect the 125kHz field."
Why transmitting 125Khz and why transmitting it by a magnetic field ?
Is that used for a metal detector ?

Passive RFID tags are powered from the near field of a carrier (e.g. standard 125kHz, 13.56MHz) and transmit information by backscatter, essentially by modulating their consumption of the carrier signal at a slow data rate.  See: http://www.eetimes.com/document.asp?doc_id=1276306

A keyless car entry system was also mentioned though. I think some active electronics also transmit data at 125kHz.

[mikeselectricstuff]

A few feet range is going to need an antenna coil of similar size to the range you want. Square wave driving a series lc circuit is the normal way. Note that th lc junction will typically see several hundred volts due to resonant rise, so the tuning cap needs to be rated accordingly. For max power you need to keep the coil resistance low, maybe also use litz wire or similar to reduce the impedance at that frequency.

[Ammar]

If the frequency is fixed a Class E amplifier would be an efficient option. http://people.physics.anu.edu.au/~dxt103/class-e/

[innkeeper]

i would try it this way.
no need for a tank unless your coupling it to something...but setting up a band pass filter makes sense to keep the signal clean.
you could probably even forgo that if you wanted to be quick and dirty about it.

i almost breadboarded this for the heck of it.