PN-Junction Tuned VFO Supplemental Page





Introduction

This web page is a supplemental page to the PN-Junction Tuned VFO web page. Presented are two simple 30 meter band receivers and a crystal oscillator schematic.

Thrifty Thirty Receiver

Discussion

The thrifty thirty is a very simple, low-cost receiver that uses a 1N4007 rectifier diode to tune the entire 30 meter band. The Figure 1 VFO tank uses L1, a 2.3 uH inductor, a 51 pf plus a 20 pF NP0 ceramic capacitor and a small trimcap to set the lower band edge. The variable trimmer capacitor was a 5 to 20 pF unit that I purchased from Digi-Key a couple years ago.

The chosen tuning voltage on the IN4007 was 7.65 to 10.03 VDC. The 1N4007 tunes most linearly in the 5 to 10 reverse volts region and the minimum and maximum tuning voltages were chosen to fit within this range. In addition, to limit the tuning range to 10.10 to 10.150 MHz, the 150 K resistor to ground and the 27K series resistor were specifically chosen. A lower value zener diode could have also been used but I had several 11 volt zeners in my zener diode drawer and went with one of these. On Jan 26, 2003, two polystyrene 100 pF caps were added to this VFO to enhance frequency stability. In addition, the cheap, low-Q trimmer capacitor was replaced with an ceramic trimmer that I bought from Dan's Small Parts. The VFO is now much more frequency stable .
Temperature compensation is both easy and essential for many simple VFO designs. Use capacitors with either positive or negative temperature coefficients. I might add replace an NP0 with a polystyrene capacitor (negative temp coefficient) for a VFO that drifts up in frequency and a silver mica (positive coefficient) for a VFO that heads down in frequency as it warms up. Trial and error with different caps may result in better stability for your VFO designs.
The best thing about projects such as this is the learning and fun that they provide. I am pleased with this receiver and enjoyed building it. This is the most inexpensive HAM band receiver I have built that was not a regenerative type. The audio gain of this receiver is tremendous.





The receiver is the same one used in the Funster 40 meter transceiver project and it's operational details are explained in that project. The input filter has an inductive and capacitive reactance of 308 ohms at the frequency of 10.13 MHz. I used a 51 pf NP0 ceramic capacitor to resonate the inductor, however a 47 pF capacitor would also be okay. This filter was sufficient to prevent any local broadcast radio being detected, but I did hear shortwave stations in the headphones at night. This was expected with the unbalanced product detector and simple RF input filter. I listened to this receiver over several nights and the amount of shortwave radio (SW) detected varied greatly with band conditions. On one night, no CW could be heard and tremendously loud SW stations were being detected. Other nights, gave great CW copying with very low levels of SW interference. The time of night also varied the amount and intensity of the SW interference for this receiver. To my surprise 30 meters has more SW interference than 40 meters in my area. Tapping down on the T1 secondary winding may help improve the selectivity of the filter.

The double-tuned bandpass filter from the Thirty meter DC receiver project on this site was built and placed between the antenna and the receiver. This greatly attenuated the SW interference. One could eliminate T1 and just use just the bandpass filter, however the unbalanced product detector is so prone to SW interference, including both the BP filter and T1 is probably prudent. This filter is shown in Figure 3 and is connected to the T1 transformer/filter. L1 and L2 could be wound on T37-6 torroidal cores to reduce its size.
For T1, it is easiest to wind the 31 secondary turns and then wind the 3 primary turns over the secondary windings in the same direction.
















The entire receiver was built in ~5 hours using Ugly Construction. Although it may be hard to see, a strip was isolated at the back of copper clad board which is connected to the B+. This is a normal procedure for me and provides a convenient method to provide B+ to you the various stages. A hobbyist motor-tool was used to grind away the unwanted copper. Note the copper surface under the inductor is also ground away and the coil is sealed down with Polystyrene Q-Dope. I also made a couple of "islands" for the tank NP0 capacitors, variable trimmer capacitor and tuning diode, although you cannot see them. Wires connecting the reverse voltage to the tuning diode are short and stiff so they do not change capacitance and thus tuning with movement. The 10K AF Gain resistor has S1 built into it and costed 90 cents. The linear-taper 50K pot was harvested from a broken HI-FI set. Tuning is very smooth and linear. If you can not locate a 3 pF NP0 ceramic capacitor, put two 5 pF NP0's in series instead. The copper PC board is 3 and 3/4 inches by 1 and 7/8th inches. Number 6-32 bolts and nuts secure the PCB to the chassis.





The Dirty Thirty Receiver

Discussion

Here is another simple, popcorn receiver that uses a NE602 (or a SA612 or NE612) as the product detector. Note the two IC's in Figure 4 are drawn pins facing upward for building with Ugly Construction. This receiver uses a single 6.8 volt zener to provide regulated voltage to the local oscillator, it's buffer, the VVC diode and the NE602. The VFO coil L3 was wound with 20 guage wire to enhance the Q factor. The VFO did drift steadily upward in the first version I built, but I replaced the 5-20 pF trimmer cap with a higher Q variation and this cured the problem. The minimum and maximum tuning voltages on the VVC diode are ~5.03 to 5.57 volts which gives a corresponding minimum and maximum frequency of 10.10 and 10.14 MHz respectively. Tuning is reasonably linear as the VVC tuning voltage change is very small.

I did not have any RF chokes with a greater inductance than 100 uH, therefore the 100 uH chokes we used for this project. Use of greater value RF chokes may be prudent. The voltage from the VFO buffer to pin 6 of the NE602 should be around 0.5 volts peak to peak and I was a bit under but I wanted couple the buffer to the main oscillator stage very lightly and did not want to use a value greater than the 33 pF capacitor specified. The original design did not have a VFO buffer and I learned that this is a critical part for the best frequency stability possible; at least in my simple designs.







A plastic project case was used as I got it for two bucks. The 10K AF gain pot has a long control rod on it and when a knob was put on it, it stuck out too far and looked silly. When the metal pot control rod was touched, AM radio could be heard, so the case of the 10K pot was grounded to eliminate this from occurring. Metal project cases rule!



A strip is cut on the rear edge of the copper surface for the B+. The right sided strip has the 13.8 volt line voltage and the left one has the 6.8 volt regulated supply. The strips are joined by the 470 ohm resistor.



Here are the two sister popcorn radio sets stacked together. The bottom receiver has a more stable VFO and significantly less SW interference.




Crystal Oscillator

Some authors have built BFOs or VXOs using rectifier and VVC diodes as the variable capacitance. VXOs generally tune very non-linearly except in more advanced designs. Figure 5 shows an example design which was built for the PN-junction tuned VFO web page. As shown the total frequency swing was just under 1 KHz. Different crystals tried gave different frequency swings and varying the 22 uH inductor also affects the frequency swing possible for a given crystal. No attempts at linear tuning were made as this is pretty much redundant for a simple designed VXO.







Return to the QRP HomeBuilder Homepage