October 1947 QST
The advent of selenium rectifiers in the 1940s was a very welcome new option to circuit designers, consumers, and servicemen. Before that, vacuum tubes did the job (with some use of copper oxide rectifiers). Selenium rectifiers have the advantage of ruggedness and reliability over tubes (~85% vs. 60%, respectively). Not requiring a heater voltage eliminates needing to create heat in excess of that dissipated due to the innate inefficiency. Voltage and power handling is adjusted by stacking appropriate layers and adjusting the physical size, respectively. A failed selenium rectifier reportedly often emitted a very foul odor, which although offensive to the user, proved to be a nice bonus for the serviceman since it immediately gave a clue as to what went wrong.
S.R. Circuits and Their Applications
By Ralph Berkman, * W7HWY
To many, the new selenium rectifiers are just a substitute for the rectifier tube of the more common a.c.-d.c. sets, but to those who have investigated their possibilities, they represent the heart of an ideal power supply for many of the pieces of gear that the ham, for one reason or another, is continually acquiring. Using a rectifier of this type, the power supply will require very little space as compared to the more conventional type, and components may be placed wherever convenient. If operated within their limits, rectifiers of this type will give off very little heat. No warm-up period is required, as in the case of the filaments of tube rectifiers. In VFOs, converters and receivers, heat has always been a problem and in these applications the dry rectifier with its low heat radiation merits special consideration.
Construction of the selenium cell varies with the various manufacturers, but all are basically the same in that the selenium is placed on one electrode and then crystallized. The plate is then formed and the barrier film forms on the surface of the selenium. Against this the other electrode is pressed, making up the cell. Cells are stacked to produce the desired current and voltage rating, making up what is known as the selenium rectifier. Rectification action takes place in much the same manner as in any other rectifier. In the forward direction there is good conductivity but in the back direction there is not perfect cut-off, and, as a result, there will be some a.c. ripple content.
These units come in the 100-, 150- and 200-ma. sizes and the additional cost of the larger sizes is very little. The efficiency of rectifiers of this type run high. The voltage drop across a unit averages about 5 volts. The critical temperature of these units is very close to 155° F. Therefore, protective resistors should be installed in series with them to limit the condenser charging current to a safe value. Temperatures may be kept down by not trying to push the unit to its full capacity, and then some. Give it a little safety margin and you will get longer life and have a trouble-free power supply.
All selenium rectifiers mount with a No. 6 screw and the unit is insulated from the mounting hole. This permits mounting directly to any metal surface, which will help to dissipate some of the heat. It will be found that mounting the unit in a vertical position will provide better air circulation. If these few precautions are observed, up to 450 volts at 200 ma. is available with standard component parts and you won't have a supply that is several times heavier than the rig it powers.
Although selenium rectifiers may, of course, be used in any of the standard transformer-rectifier-filter systems, it is natural to associate their advantage of compactness and light weight with transformerless supplies. Fig. 1 is a straightforward half-wave rectifier circuit which may be used in applications where 115 to 130 volts d.c. is desired. It makes an ideal bias supply, for instance. In this, as well as other circuits, it will be observed that the negative side of the output is common with one side of the a.c. line and it is suggested that this side be fused with a 1/2-ampere fuse.
Fig. 2 shows several voltage-doubler circuits. Of the three, the one shown at B is the most desirable since there is no series condenser. It is a full-wave circuit and there will be very little ripple voltage appearing at the output. On the other hand, the circuit of C has one very desirable feature in that point X is common to both condensers in the rectifier and also to the first condenser in the filter. This means that a single-unit three-section condenser may be used, saving space. If less than 100 ma. is being used, this, in the author's opinion, is the best circuit. The ripple content under these conditions, and the leakage between sections, will not be excessive. These three circuits will find ready application in communications receivers, converters, VFOs, test equipment, etc., and especially in cases where heat has been a problem.
Fig. 3-A and -B shows voltage-tripler and quadrupler circuits respectively, for use where higher voltages are desired. They are ideal for powering the small portable or fixed-station rig and the compactness and light weight will be appreciated. The writer uses the tripler circuit for powering a small 'phone-c.w. rig where the weight of a comparable conventional power supply would make portable operation prohibitive.
All components are standard. C1 in all circuits is for" hash" filtering and its value is not critical. A 0.05-μfd. 600-volt-working condenser should serve. All other condensers should be 40-μfd. 200-volt units, except those in the tripler and quadrupler circuits. Those in the circuit of Fig. 3 should have a rating of 450 volts working. In the voltage multipliers and in other circuits where a condenser is passing the full current, good condensers should be used because the a.c. ripple mentioned above appears across the condenser and increases as the load increases. If the current is allowed to become too high, it will cause heating and deterioration of the condenser. This can be kept to a minimum by using a capacitor of high value and making sure it is of good make. Aside from this, no particular difficulties should be encountered even when using voltage-tripler and quadrupler circuits. R1 should be 25 ohms, but if it is found that the rectifier units are running a little too warm, this value may be increased to as high as 100 ohms, with a corresponding drop in output voltage, of course.
A single-section filter, as shown in Fig. 2-C, will provide sufficient smoothing for most applications.
At W7HWY the circuit of Fig. 3-A is being used for powering a BC-474 and the circuit in Fig. 1 for the receiver portion. The transmitter is running 15 watts on 75 'phone and 18 watts on 80 c.w., while on the road, and in my estimation it's tops.
Summing it all up, for the guy who is looking for a lot out of little things, this is economical power in a mighty small package.
Posted July 21, 2016