Dick Cappels' Project Pages

HOME                                                                            

      
Circuit of the Digital RF Field Strength indicator with LED Display using Atmel AT90SA2313


The circuit:

The circuit contains a small analog section and LED driver, tied together by the micro controller. A two-transistor circuit provides a stabilized voltage and serves as a detector that serves an A-to-D converter based on the comparator contained in the micro controller. The voltage from the RF detector is measured once every 160 milliseconds and after removing the zero offset, correcting the polarity (higher field strength results in lower voltage from the detector), overflow detection and formatting to BCD, the micro controller provides drive signals to the three digit LED display, which is refreshed once every 16 ms, with a dwell time of about 4 ms per character.This is a little slow for a handheld device as when the meter is moved, a trail of digits is left on the user's retina, so if this bothers you, increase the refresh rate.

ANALOG SECTION

Q1 is a combination shunt regulator and bias source for Q2. As a shunt regulator, it provides a stabilized voltage for the A-D converter circuit. This shunt regulator has low loop gain, thus the display may jitter because this voltage affects the RF detector's output voltage and power supply fluctuations resulting from driving the LEDs may result in feedback that causes the readings to fluctuate. This could be improved by using a band gap reference ahead of the shunt regulator, but it drives cost and complexity up. I have found the circuit to work erratically when using some batteries -these were some very cheap carbon-zinc and (supposedly) Ni-MH cells that were counterfeits of name brands I bought in S.E. Asia. I suspect these cells have a relatively high impedance, thought I did not confirm this experimentally. Name-brand (real ones) alkaline and Ni-Cad batteries don't have these problems.

The base voltage on Q1 also serves to establish the bias voltage for Q2's base. Q2 serves a an RF detector. Q1 is biased so that there is about 600 millivolts between its collector and emitter. A small wire antenna forms a capacitive coupling from the input of the circuit to the RF source being observed while the bulk of the circuit, including the battery holder forms a capacitive coupling from the circuits' "ground" to "earth". When the user holds the meter in his or her hand, this ground coupling is enhanced. RF from the small wire antenna is capacitively coupled to Q1's base and this modulates the collector in Q1. Gain of the amplifier is independent of the value of the collector resistor (39K in this case) because that's how transistors work (to use vacuum tube terms, collector signal gain is Gm*Rl, and Gm is nearly proportional to collector current, so doubling the collector load cuts Gm in half) .The nonlinearity in the base voltage-to-collector voltage transfer funcion results in detection of the RF signal. Collector current is averaged by the low pass filter established by the collector current and the capacitance to ground.

The other input to the AT90S2313s on-chip voltage comparator is connected to ramp capacitor C1. The time it takes R1 to charge C1 to equal the voltage from the RF detector (on pin 13 of the AT90S2313) is measured to determine the detector's output. See Atmels' application note, AVR400, for the theory of operation for the A-to-D converter.

The power switch is SPDT set up to short the +3V line on the circuit to ground to make sure that the decoupling capacitor was completely discharged so the circuit would reset properly if the power was switched off then back on quickly. This was only done as a precaution and is probably not really necessary.

DISPLAY DRIVER

The LED display is a common-cathode multiplexed display. The cathode associated with each digit is driven by a separate digit driver (Q3, Q4, and Q5). Completely non-critical, just make sure you have enough base current to saturate the transistors. I used 2N4401s with 1K base resistors because I have a lot of each of them.

Port D bits 0 through 6 directly drive LED segments 1 though 7 respectively by shunting current from the 220 Ohm resistors for the off segments to ground. This results in slightly increased average current over driving the segments from the positive power supply, but it does reduce the ripple current on the power supply, thus making the display jitter less. If you are using a more efficient LED display, you can use lareger resistors and reduce battery consumption.

The display is run with each digit enabled 25% of the time during normal operation so the power ratings of the resistors can be scaled proportionally. This trick won't work with the I/O pins on the micro controller. If you are careful not to exceed the maximum permissible for the part you are using, you could consider eliminating the digit drivers (Q3,Q4, and Q5). If you do eliminate them, be sure to invert the bits in the code.

 

CONSTRUCTION

Since the collectors and emitters of the detector transistor are both at RF ground, choice of transistors isn't all that critical. A low base-collector capacitance will enhance the VHF and UHF sensitivity. All transistors should be of the same type to minimize thermal drift. The DC gain of the detector is about 20X (estimated by multiplying the voltage drop across the collector load by 38).Assembly is not critical and mine was built on punched fiberglass board without a ground plane. Sensitivity can be adjusted by changing the rate of charge of C1 by changing either R1 or C1. Just be careful that the entire conversion can be completed (that is, that capacitor voltage can charge up to the RF detectors output voltage) within the 4 milliseconds available.

 

HOME

Use of information presented on this page is for personal, nonprofit educational
and noncommercial use only. This material (including object files) is copyrighted
by Richard Cappels and may not be republished or used directly for commercial
purposes without explicit permission For commercial license,
click HERE.

Please see the liability disclaimer on the HOME page.

Contents ©2002,2004  Richard Cappels, updated November 2004 All Rights Reserved . http://projects.cappels.org/