Hello all,
This is my first post on this forum, but likely not the last. I am a wildlife biologist with a strong interest in the use and development of radiotelemetry devices for use in tracking small animals. Although I have developed a strong interest in RF circuit design, I am still far from proficient. Over the last few years I have been working with and modifying a design of one of the world's smallest VHF radiotags (see attached circuit diagram) for a species of bird that I work with in the Neotropics; a design published by Naef-Daenzer and others (2005).

For a number of years the Naef-Daenzer tag was sufficient for my needs, and I didn't need to understand the circuit itself to use it. However, I am now interested in moving from this simple RC-pulsed design to a similar design, but whose RF pulses are controlled by an MCU. Although I have already gotten a circuit working using an ATtiny20 MCU, I would like to have greater understanding of the different components of the Naef-Daenzer circuit so that I can identify and then remove unnecessary components, and modify others (such as matching and filtering networks). What I have gleaned from communication with Naef-Daenzer himself, his paper, and my own limited understanding of electronics, I have been able to identify components that deal with smoothing input voltage and limiting output amplitude (purple components in diagram), those generating/modifying the pulses (blue), those responsible for RF generation and amplification (yellow), and those involved in RF filtering (red). I have tried to simulate this circuit with PSPICE, but have had little luck getting it to work, although an AC sweep analysis with an AC source in place of Q does reveal strong resonance at ~203 MHz, which is NOT the frequency produced by this tag [i.e. 148.5 MHz]).

My questions, for anyone willing and able to help point me in the correct direction are the following:

1) What is the function of R2, L1, L2, and C3 in this circuit? Do these appear to be involved in impedence matching and filtering, or are they related to the feedback required for running the oscillator? Generally, how are they operating?

2) L3, C4, and C5 are supposedly a "pi-filter-like matching circuit," for the 148.5 MHz frequency of the tag. Can anyone explain how this filter operates, given that it is not exactly a pi filter? How can I calculate the hypothetical cut-off frequency for it? Is it possible to modify the values of C and L to operate on a different frequency (e.g. on the 166.38 MHz band), or would a completely different matching network be required?

Any advice or knowledge you could offer for any of these questions would be greatly appreciated!

Many thanks,
Julian

 

R2 is a bias resistor. It sets the voltage at the base of T1 when the transmitter is not transmitting. Since R1 is very high, a small current flows from the battery through R1, R2, T1 (B to E) and R3 to ground (back to the battery). This current sets Vbe for T1 to some 0.6 volt. The current will not start to flow before the voltage over C1 is greater than 0.6 volt. This voltage level will serve as a reference for the on/off cycling of the transmitter.

Now, the base of the PNP transistor will also be at the same level as the base of T1. When C1 has charged to ca. 1.2 volt (i.e. the sum of the two base emitter junctions of the PNP and T1) , the PNP will conduct and supply (from C1 through emitter to collector of the PNP) a large base current to T1. This triggers the transmitting part of the cycle. When C1 discharges sufficiently below 1.2 volt, the PNP will close, transmitting will stop, and the process repeats. As the circuit is now, R2 is absolutely essential for its function.

L1 is a low pass filter (with C2) that stops the RF pulses generated over L2 and C3 from interfering with the charging circuit R1 and C1. L1 cannot be removed.

L2 and R3 is a tank circuit (search "L/C tank"). The tank is the collector load for T1 and is absolutely essential. You can find its resonance frequency from an online L/C Resonance Frequency Calculator.

I will take a further look at the correctly coloured red components that seem to be filter and/or impedance matching of the output amplifier to the antenna.

 

Thank you for taking the time to provide this detailed and very clear explanation mekongrf!! Very much appreciated!

A point of clarification on your answer in case anyone else reads your explanation: I think you mean "L2 and C3" form the tank circuit, not "L2 and R3." In addition to my remaining questions about the matching circuit I've got one follow-up question if you (or anyone else) can help:

I am unclear about how C1 can supply the burst of current needed for the full duration of the RF pulses of this circuit. When C1 is sufficiently charged (1.2 V) why do T1 and the PNP transistor stay on once C1 starts discharging below this value? Is there feedback I'm not able to 'see' that keeps the transistor(s) in active mode past the point that their bases are seeing less than 1.2 V? Are the RF oscillations responsible for feedback that keeps T1's base current sufficiently high to keep it in active mode? This point isn't too essential for my purposes since I'm going to be replacing the RC control of the pulses by an MCU, but I feel like there's something fundamental I'm not understanding here (again, I'm a biologist with no formal training in EE).

Regarding L3, C4, and C5, my original thought was that this might be an LCC configuration filter, but the topology isn't quite right (for this to be an LCC configuration; C4 would have to be in series with L3, with C5 connected to ground between them). This is an important aspect of this circuit for me to understand since it is clearly an essential component of the "RF portion" of the transmitter, and the values chosen will be frequency dependent. I will have to understand what's going on well enough to modify the values to match the frequency I'll be using (~166 MHz), and since I can't seem to simulate the circuit trial-and-error learning to stumble onto the answer won't quite cut it.

Again many thanks to you and anyone else who can (and is willing to) help!

-Julian

 

Sure - the resonant tank is L2 and C3. It resonates at a bit over 40MHz. Guessing that the 3rd overtone crystal is some 49MHz or so, it is reasonable to assume that (in combination with other effects) the parallel circuit L2//C3 is a very high impedance at the fundamental frequency.

Hence, L1 (as well as L3 and C4) should imo also be regarded as part of the collector load of T1 as in this circuit where the antenna is connected to the top of the LC tank http://www.jbgizmo.com/page32.htm . If L2//C3 is supposed to have infinite impedance, then you might find your LCC network or something similar (but in this link the output is taken from a different point http://electronicdesign.com/communications/back-basics-impedance-matching-part-3 ).

Anyway, the output circuit is too complex for me, so I suggest that you take an empirical approach to it and try to change the components one by one to determine their effects. This is somewhat like they report in http://jeb.biologists.org/content/208/21/4063#T1
Alternatively, you will always find some university teacher / researcher that will be able to analyze it for you.

When it comes to the pulse duration, there is no component than can deliver current as "feedback". R1 is the only additional source to C1, and it is a high impedance delivering only negligible current. This seems to be a desired point (low duty cycle) in the link in the paragraph above. The circuit does not switch momentarily because of the "rounded" characteristics of the transistor "turn on / off point".

 

Looking at L3 in series with C4 gives a resonance frequency of 277 MHz. Adding some ad-hoc slack for practice versus theory, it seems obvious that L2//C3 is related to the crystal fundamental frequency and L3--C4 and C5 are related to the 3rd overtone radiated frequency. This will give you a hint on component change direction when changing crystal frequency.

But I still think you will have to optimize by tinkering - parasitics of a fraction a pF / a few nH will be uncontrollable in the design phase. Hence your imo only way is a rough calculation followed by do and see.