Are you tired of squinting at cross-needle SWR/wattmeters? Do you keep missing the optimal tuning settings when you use an antenna with an SWR of 5 or greater? Are you annoyed by your SWR meter's poor resolution at SWR levels above 3? Do you wish that your MFJ SWR/wattmeter worked at QRP power levels or that your QRP SWR/wattmeter could handle 100W? Are you annoyed when increasing your power increases the SWR reading? Are you causing too much QRM because you need at least 10W to read your SWR meter?
If you answered yes to any of these questions, then the QROP Meter project for 160m to 10m operation is for you. Why?
The QROP Meter uses a directional coupler circuit to sample the forward and reflected RF voltages between the transmitter and antenna. RF transformers sample the load voltage and load current, and the circuit is configured to isolate the forward and reflected voltages. These two RF voltages are fed into 1N34A Germanium diodes for rectification. LM324 operational amplifiers are configured to compensate for the diode drop loss. A PIC16F72 microcontroller drives the output LEDs of the SWR and wattmeter displays.
Current transformer TX1 samples the load current, and voltage transformer TX2 samples the load voltage. In TX1, the load current flows through N1s and induces the load current sample in N1p. Because N1s consists of 1 turn and N1p consists of 10 turns, the current through N1p is 10% of the load current. In TX2, the voltage across N2p is the load voltage and induces the load voltage sample across N2s. Because N2p consists of 10 turns and N2s consists of 1 turn, the voltage across N2s is 10% of the load voltage.
Resistors R1 and R2 configure the circuit so that V_F_0 equals 10% of the forward voltage and V_R_0 equals 10% of the reflected voltage.
FT82-43 ferrite toroids are used for the transformers. Type 43 ferrite material is the optimum material for the HF bands due to its broadband capability in transformers. The inductance per winding decreases with respect to frequency and limits the reactance values of N1s and N2s at high frequencies but allows N1p and N2p to have high reactance values at low frequencies. The FT82 size is larger than that of the toroids used in QRP SWR meter designs due to the need to handle larger amounts of power. Applying excessive power to a toroid results in excessive heating and flux density, either of which may affect toroid properties or even cause damage to the toroid. However, the use of a larger toroid requires a longer length of wire in the primary windings of the transformers. The use of longer wire in the windings results in more stray capacitance between windings and interferes with the proper operation of winding N2p, which is designed to have an impedance of 10,000 ohms.
For the tradeoff between maximizing current-carrying capacity and minimizing stray capacitance between windings, #20 wire is used in the secondary windings that carry the full load current. Because the primary windings need minimal stray capacitance between windings rather than the maximum current-carrying capacity, #30 wire is used.
R1 and R2 both need to be at 50 ohms. In QRP SWR meters, almost any resistors are suitable. Because the QROP Meter must be able to handle more power, I recommend using 51-ohm 2W resistors for R1 and R2. If you prefer the convenience of Radio Shack, you can also use a group of five resistors in parallel to gain the 2W capability: four 220-ohm .5W resistors and one 680-ohm .5W resistor.
A standard 12V power supply or battery connected to a switch mounted on the front panel powers the QROP Meter.
A 1N4001 diode offers reverse polarity protection with only a modest voltage loss (around .5V) for VCC.
Four FB73-101 ferrite beads in series with the 1N4001 diode provide inductance to protect the supply voltages of the op amps from external AC interference that can interfere with their proper operation.
An LED in series with a 1000-ohm current-limiting resistor provides the "idiot light" that shows that the power supply is connected to the rest of the circuit.
The 7805 transistor uses 12V to provide a 5V source for the microcontroller.
The ceramic capacitor for the 5V source should be as close as possible to its respective pins on the microcontroller. The rest of the Power Supply Circuit should share the board with the LM324 amplifier in the Active Rectifier Circuit.
Several decoupling capacitors protect the supply voltages of the op amps from both externally generated and internally generated AC interferenece. Real capacitors have different behaviors at different frequencies. The 100uF, 10uF, and 1uF electrolytic capacitors decouple low frequency interference. The .1uF capacitor decouples high frequency interference and should be placed as close as possible to the leads connecting the VCC and ground pins of the LM324 IC of the Active Rectifier Circuit. Additional .1uF capacitors are used in the SWR Display Circuit and the Wattmeter Display Circuit.
Do NOT add additional lower-value capacitors (such as .01uF or .001uF) with low ESR (internal resistance in series with the capacitance), as this will create more antiresonant frequencies, frequencies at which the impedance between the power supply and ground spikes upward. Thus, the network of decoupling capacitors will fail to work as intended at certain frequencies.
WARNING: Electrolytic and tantalum capacitors are polarized. The positive lead MUST be connected to VCC, and the negative lead MUST be connected to DC ground. Electrolytic capacitors connected with the wrong polarities will pop and make a mess. Tantalum capacitors connected with the wrong polarities will start a fire! Although the 1N4001 protection diode ensures that VCC=0V if the power supply polarity is reversed, the risk of improper polarity still exists in the building stage.
Diodes D_F_1 and D_R_1 should be placed on the same board as the directional coupler to minimize the effects of stray inductance from long wires.
Diode D_F_1 rectifies the RF forward voltage sample V_F_0 into the DC voltage sample V_F_1. Diode D_R_1 rectifies the RF reflected voltage sample V_R_0 into the DC voltage sample V_R_1. Germanium 1N34A diodes are used to minimize losses. However, both RF voltages will be under 1V when transmitting at low power levels, and this causes a substantial diode drop loss.
Both D_F_1 and D_R_1 must be matched in order to enable the SWR meter to provide good accuracy at high SWR values. The diode drop loss varies from one individual diode to another, and the smaller the voltage to be rectified, the greater the difference between the rectified voltages. Instructions on how to match these diodes are provided later in this article.
The next step is to compensate for the diode drop losses in the rectifiers with logarithmic non-inverting op amps. Two noninverting logarithmic LM324 operational amplifiers are configured to provide gain that varies according to the input voltage levels. The resistances of feedback diode D_F_2 and resistor R_F_2 determine the gain of the forward voltage compensating amplifier, and the resistances of feedback diode D_R_2 and resistor R_R_2 determine the gain of the reflected voltage compensating amplifier. When V_F_1 is low, D_F_2 has a high resistance, and the forward voltage compensating amplifier has a high gain. When V_R_1 is low, D_R_2 has a high resistance, and the reflected voltage compensating amplifier has a high gain. When V_F_1 is high, D_F_2 has a low resistance, and the forward voltage compensating amplifier has a gain near unity. When V_R_1 is high, D_R_2 has a low resistance, and the reflected voltage compensating amplifier has a gain near unity.
D_F_2 and D_R_2 must also be matched in order to enable the SWR meter to provide good accuracy at high SWR values. Diode resistance at a given voltage varies from one individual diode to another, and the smaller the voltages involved, the greater the difference between one diode and another.
R_F_2 and R_R_2 must both be the same value, but the optimum value of these resistances CANNOT be known in advance and will vary from one individual QROP meter to another. Instructions on how to determine the optimum resistance value are provided later in this article.
Capacitors C_F_2 and C_R_2 are used to minimize oscillations. When there is no significant reflected or forward voltage sample to amplify, the gain will be high due to the large diode resistances. Adding these capacitors in parallel with the diodes will limit the amplifier gain for AC signals. Omitting these capacitors will allow unwanted AC noise (such as 60 Hz noise) to amplify. The AC noise may cause intermittent unwanted oscillations and cause parts of the rest of the QROP Meter to behave strangely. As is the case with the power supply decoupling capacitors, watch your capacitor polarities.
.01uF decoupling capacitors are needed to minimize RF ripple in V_F_3 and V_R_3.
I use an LM324 op amp in the active rectifier due to its low cost, its widespread availability, its ability to withstand a 10V input in the absence of a power supply, and an input range that includes ground.
|
LED
(SWR display) |
Approximate
SWR |
LED
(wattmeter display) |
Forward Power |
| D1 | 1.1 | D11 | 280mW |
| D2 | 1.4 | D12 | 560mW |
| D3 | 1.8 | D13 | 1.1W |
| D4 | 2.3 | D14 | 2.2W |
| D5 | 3 | D15 | 4.5W |
| D6 | 4 | D16 | 8.9W |
| D7 | 6 | D17 | 18W |
| D8 | 11 | D18 | 35W |
| D9 | >17 | D19 | 71W |
| D10 | infinity | D20 | >100W |
My earlier version of the QROP Meter used an LM3914 linear display driver chip to drive the SWR meter display and an LM3915 linear display driver chip to drive the wattmeter display. The SWR display behaved erratically between transmissions and during very low power transmissions. One or more of the SWR indicator LEDs would light up at random because the forward and reflected voltages feeding the LM3914 inputs had similar values.
If the forward power is less than 200mW, then all of the SWR meter LEDs and the wattmeter LEDs will be dark. SWR readings at these extremely low power levels would be less accurate due to the offset voltages from the op amps and the quantization errors in the A/D conversion process.
It is essential to make sure that both active rectifier circuits provide the same output voltages given the same input voltage, and an oscilloscope and sine wave function generator are necessary tools for this. Both active rectifier circuits should be originally constructed on a breadboard. R_F_2 and R_R_2 must match. (A value of 10,000 ohms is recommended initially.) A constant RF voltage of around 100mV 0-peak should be provided for V_F_0 and the DC voltage V_F_1 measured. The process should be repeated by providing the same RF voltage to V_R_0 and measuring V_R_1. For the same 100mV 0-peak RF voltages at V_F_0 and V_R_0, respectively, both V_F_1 and V_R_1 must have values within 1% of each other. Trial-and-error is needed to find two matching diodes, and it will take several tries to find a matching pair.
Once D_F_1 and D_R_1 are found, a matched pair of diodes for D_F_2 and D_R_2 must be found through a similar trial-and-error procedure. For the same 100mV 0-peak input at V_F_0 and V_R_0, the matched diodes D_F_1 and D_R_1 in place, and the matched resistors R_F_2 and R_R_2 in place, V_F_2 and V_R_2 should have measured DC voltages within 1% of each other. Again, trail-and-error is needed to find two matching diodes for D_F_2 and D_R_2.
Once D_F_2 and D_R_2 are found as well, it will be necessary to determine the optimum resistance value for matched resistors R_F_2 and R_R_2. With the two diode pairs in place, the optimum value for these two resistors can be determined. The matched diodes remain in place, but the RF input voltages are changed. Test the active rectifier circuit for input voltages in steps of a factor of 2 or less from 10mV to 10V. The input 0-peak voltages and the output DC voltages of the active rectifier should match to within about 10% of each other at all values from 100mV or higher and within about 30% of each other at all values from 40mV or higher. In some QROP units, the optimum resistance value will be 10,000 ohms. Other individual units will require a higher or lower resistance value. There is no way to know in advance what the optimal value will be.
WARNING: Watch your capacitor polarities.
I use a metal chassis that is 7 inches wide, 5 inches long, and 3 inches tall. This is the right size for four Radio Shack 276-150 PC boards inside the instrument and the LED displays on the front panel. I use two SO-239 connectors for connecting the directional coupler to the transmitter and receiver in order to be consistent with most transceivers. The back panel contains these two RF connections and a DC power connector. This DC power connector MUST be an insulated one so that the DC and RF grounds remain isolated. The front panel contains 22 holes of 1/4-inch diameter each. Ten of these holes are used for the SWR meter display, ten of these holes are used for the wattmeter display, one hole is used for the "idiot light" that indicates that a power supply has been connected, and one hole is used for a switch that connects the power supply to the circuit.
The rest of the circuit handles only DC, and stray inductances and capacitances are not a significant issue.
I dedicate one board to the LM324 operational amplifier, one board to the SWR meter display, and one board to the wattmeter display. If you use four Radio Shack 276-150 PC boards, I suggest putting the power supply connections (including the protection diode, capacitor, "idiot light" LED, and its current-limiting resistor) on the LM324 board, as the remaining two boards will be crowded with LED connections for the displays. Again, you MUST be aware of the dangers of polarized capacitors, and you may wish to substitute a nonpolarized capacitor or a series of nonpolarized capacitors in parallel for the 10-microfarad capacitance that only a polarized electrolytic or tantalum capacitor can provide.
The DC and RF grounds are connected by a jumper with four FB73-101 ferrite beads. One end is connected to the RF board, and the other end is connected to a DC board. The DC grounds on all three DC boards are connected with a jumper wire.