Rob's web

A.C. power for the noise generator

Home - Techniek - Electronica - Radiotechniek - Radio amateur bladen - QST - A.C. power for the noise generator

Photo 1
The diode noise generator and its power supply fit into a small-size Minibox. The dial is on the variable resistor that controls the noise-diode current.

Regulated D.C. for crystal-diode noise sources.

W7ZFB presents a simple way to solve the battery-replacement and varying-voltage problems inherent in conventional battery-powered crystal-diode noise generators. Application of a.c. power and Zener regulation does the trick. If you haven't built a noise generator, this tells you how. If you have, incorporate a confidence-building stable voltage supply.

After using a conventional battery-powered crystal diode noise generator for several months, it occurred to me that a simple zener-regulated a.c. supply would solve the battery-replacement problem and also remove any possibility that my noise measurements were inconsistent because of varying battery voltage. I let this idea age for several months and then experimented. The result is a practical and flexible piece of test equipment which is invaluable to the v.h.f. man. Cost is nominal, whether you start from scratch or just add the power supply to an existing generator.


The circuit of the noise generator is shown in Fig. 1. Those who have constructed and used the generator shown in recent editions of the ARRL Handbook will immediately recognize the noise generating part of the circuit and typical construction from the photographs.

Fig 1
Fig. 1. Circuit of the a.c.-operated noise generator. Resistances are in ohms (K=1000).
C1 250 µF, 25 V (Sprague TVA-1208).
C2 100 pF, disk ceramic.
CR1 Silicon rectifier, 400 p.r.v., 500 mA
CR2 Zener diode, 5.6 volts, 1 watt; 1N1520 or Barry Electronics No. W40-556 (89e).
CR3 1N21 or 1N23 (see text).
J1 See text.
R1 33 ohms, 5 watts.
R2 5000 ohm control, c.c.w. log-taper.
R3 51 or 75 ohms, ½ watt.
S1 S.p.s.t. toggle.
T1 12.6 volt, 2 amp. filament transformer.

Since this portion is conventional and described in the Handbook, I will concentrate on the power supply.

A 12.6-volt filament transformer feeds a halfwave rectifier and filter (CR1-C1) and produces between -11.0 and -11.5 volts d.c. under load. This voltage, dropped through R1 and maintained constant at -5.6 volts by the Zener diode CR2, causes current to flow through the variable-resistance noise-generating circuit R2CR3R3.

For consistent comparative noise-figure measurements, this voltage must be held constant. To do this, circuit constants must be set so that the dynamic range and power rating of the zener diode will not be exceeded as R2 is varied over its range. Component values specified in the parts list were experimentally chosen to meet this goal.

If we establish this condition, then the rest of the operation is relatively simple. Zener diode action is analogous to operation of a gaseous VR tube. Its dynamic resistance changes, as the R2CR3R3 path resistance varies, in such a way that the voltage at the junction of R2 and CR2 is always constant. AB R2 is increased, more current flows thru the Zener; as R2 is decreased more current flows thru the noise diode. Zener diode current changes automatically to compensate for changes in R2.

Photo 2
Inside view of the noise generator shows the simplicity of construction. The 1N21-type crystal and load resistor are mounted directly to the coax connector. The Zener and power-supply rectifier diodes are the top-hat types on either side of the rectangular-shaped resistor.


Study of the accompanying photographs will show that construction is simple and does not require special tools. The transformer and power switch Si are located as shown in the Minibox (Bud C-3006A). CR1, C1, R1, and CR2 are wired to conveniently placed terminal strips. Components R3, C2 and CR3 are located at one 4 end of the box as close as possible to the output connector. Short leads for R3 and C2 are imperative to insure sufficient output at 144 Mc. In this instance, following Handbook instructions and using aagt d-cap clip and a contact from a miniature socket to mount the crystal diode works nicely. J1 can be any good quality v.h.f. connector. I used a nondescript female panel-mounting type N connector for J1 with ap1 propriate interconnectors, but the specific type I is unimportant so long as connection to the receiver or converter is as short as possible and l does not introduce much mismatch.

Several comments about parts substitution are in order. Critical components are R1 and CR2. If the Zener is changed, R1 may have to be changed so that maximum current (Zener wattage divided by its voltage rating) through the Zener does not exceed its power rating. This can be checked by metering the current through CR2 with R2 disconnected at A-A in Fig. 1. At the sanie time, R1 should not be so large that the voltage drop through it reduces the voltage level at the Zener to less than its rated 5.6 volts, with R2 connected and at minimum resistance.

A c.c.w. log-taper pot should be used at R2 so that the noise level control will be linear in db., for ease in estimating the magnitude of noise figure improvements. Other tapers can be used, but this may make noise-level adjustment somewhat difficult.

Use any of the diodes listed for CR3. I found 1N21C to be the most prolific noise generator, ut you may find one of the others best. If you ve several, choose the diode which gives aximum output.

R3 should be 51 ohms if your receiver or conerter input impedance is 52 ohms. If it is 75 ohms, use this value.

Finally, from what I can determine, there is no particular advantage to either the positive- or egative-polarity approach. I used the negative approach simply because of the mechanics of mounting the crystal diode. As shown, replacement is easy. If you prefer, though, reverse all diodes and C1 and the circuit will work equally well. Just be sure to back-bias the Zener and to forward-bias the crystal diode.


Adjustment of the unit is simple. First, check the voltage at the junction of R2 and CR2. It shouldn't vary more than 0.1 volt as R2 is varied. Second, meter the voltage across R3. This should vary from 0 to about 3 volts as R2 is varied, if you use 51 ohms for R3; proportionately less, of course, if you use 75 ohms.


While a battery-powered noise generator is essential for in-place adjustment of antenna-mounted preamps, this unit is more practical for the majority of us who do our work in the shack. Output is good at 2 meters and should be usable above this frequency. The upper limit, of course, is a function of both the generator and the noise figure of the device under test, and is reached when maximum generator output can no longer mask receiver noise.

John T. Conley, W7ZFB/4.