No.35
Amateur Radio
4CX1600B

As we enter a new sunspot cycle, Dean's 4CX1600B amplifier should pique the interest of most DX and contest station operators. But, please note, this is a conceptual piece, not a construction article. While experienced amplifier builders would likely ferret out the missing details, we caution novice builders not to use this material as a first attempt at amplifier construction. For example, in the power supply, several voltage divider resistor values are given, but are lacking the wattage data. These values would have to be calculated with some margin for safety. Many of the smaller supplies for the filament, bias, screen, and relay voltages were originally shown as 120-volt AC designs, using the 220 neutral as a common return. Since this raises some safety issues, the editors have taken the liberty to redo the design using 220-volt components throughout. Meter shunts may require adjustment depending on the meters you select, and likewise the components associtaed with the RY2 relay time-delay circuit may also require some tweaking. -Editor

The Svetlana 4CX1600B tetrode, now available in the United States, provides an ideal vehicle for the design of new, state-of-the-art tetrode power amplifiers.* The tube has exceptionally high power gain and a transconductance of 50000 micromhos. As such, the 4CX1600B requires care to obtain the expected performance. It is relatively rugged; there are only two areas that require watching. They are the two grids: screen and control. That isn't to say that they are particularly frail; they aren't. However, these grids are practically the only areas where a "heavy-handed" builder can challenge the tube design. Although the circuit diagram of a highly successful amplifier built around the tube is included here (see Figures 1 and 2), most of its design details are not. This article was written with the intent to describe good engineering practice with respect to the '1600, not as a "how to" article. However, there are some sophisticated design features which will be discussed in principle.

The tube

Photo A shows the 4CX1600B. The most important tetrode element (which influences the very high efficiency of the tube) is the anode and its interior. It is machined to present four annular cavities facing the center of the tube. They are numbered 27 (one number for four cavities). Their function represents the most important difference between the '1600 and other external anode tetrodes. All tetrodes share a disadvantage. At the point where the instantaneous plate voltage drops near or below the screen potential, the screen attracts more of the cathode current. The cathode current is then reflected in a higher screen current reading. At this same point, the plate current is at maximum and more secondary electrons are produced. These electrons are attracted to the screen. If there is no design provision to circumvent or decrease the effects of secondary emission, the instantaneous low plate voltage always will produce heavier screen current. That is where the design of the 4CX1600B shines. The aforementioned annular recesses in the anode are deep enough so that the bulk of the secondary electrons kicked out of the plate cannot escape the recess (the "trap") and are forced back into the plate. Those electrons normally would have to buck the incoming current, which may be on the order of one ampere. The cathode current stream carries with it a large negative charge commensurate with that current; hence it provides the field gradient (within the trap) to return the secondaries to the plate. Thus, it is possible for the plate voltage to dip below the screen voltage without producing inordinately high screen current. The most important benefit of this behavior is that the coincidental increase in plate voltage swing will produce both higher efficiency and higher power output. Actual confirming performance data will be presented at the conclusion of this article. With a transconductance of 50,000 micromhos, the tube cannot be operated with a tuned-grid circuit without careful neutralization. It is designed for operation with a swamping resistor shunted directly between the grid and ground. The driver power is therefore dependent on the drive voltage (peak) and that depends upon the grid bias, also a function of the plate supply voltage. So the drive power for the tube usually is between, say, 30 and 50 watts, all of which is dissipated in the resistor. The recommended value for the shunt resistor is typically 50 ohms, non-inductive. It should be sized for the maximum continuous drive power, with some reserve power capability, say, 60 watts or so. It should be in the cooling air flow path; easy to do, because it will be near the base of the socket. The input capacitance of the '1600B is about 86 pF. That shouldn't be a problem on the lower bands, although a conscientious designer would certainly compensate for it. The amplifier shown here (Figures 1 and 2) uses a series compensating inductance, calculated to have equal and opposite reactance at 21 MHz. It was compensated only at 15 meters, and that seemed to be adequate. The main advantage seems to be in presenting a resistive load to the exciter. It would be slightly more serious at VHF; however, it proved to be no problem at 160 to 15 meters with the proper parasitic circuit and the compensated load reactance. The inductance to compensate at 15 meters is 0.162 µH. The precision of an ordinary grid-dipper to evaluate the inductance is adequate. The maximum full-power rating is 250 MHz, so 15-meter operation presents no problems at all. Svetlana recommends using the tube as a class AB1 amplifier. This recommendation proves to be correct, and it operates very well with the no-grid current characteristic of AB1. This particular tube wouldn't quite deliver 1500 watts at 3250 (loaded) plate volts (un-loaded Ep is 3550). Because one objective was to deliver the full 1500 watts, it was decided to "push" it a little (very little) into class AB2. No one will notice the difference in transmitter power (at the receiving end) between, say, 1450 and 1500 watts. The exercise was performed just to see if it was possible, and to gain some additional experience. I don't recommend that anyone duplicate it. It's better to raise the screen voltage a little. The justification rests largely on the published maximum grid power dissipation. It is 2 watts, maximum. Svetlana recommends limiting operating grid dissipation to 0.1 watt. Let's re-emphasize that it is possible to push the tube in class AB2, if you are comfortable with this type of work and if you can measure grid current with the necessary sensitivity and precision. The amplifier shown here has a "multimeter" type switch with a difference. The grid current is measured on a scale position that is 1 miliampere full scale. The necessary grid current to produce the desired 1500 watts is nominally 110 microamperes. Except for a small ego-trip, the increase in power and the delicateness with which it should be approached make such an adjustment not worth the trouble. Even so, the grid dissipation is quite adequate for this small excursion into class AB2.

Important data

Some of the design numbers of use are listed in Table 1(bottom of page). More in-depth information is provided in the manufacturer's data sheet, which is available from Svetlana. However, Table 1 lists most of the necessary factors emphasized in the design of the W5LAJ amplifier. Though the absolute maximum grid dissipation is listed as 2 watts, Svetlana recommends no more than 0.1 watt as a matter of practice.

Cooling

This amplifier was constructed and operated at an elevation of 6,000 feet above sea level. Table 2, copied from the Svetlana '1600 data sheet, reveals that in order to operate at full plate dissipation ratings at the elevation of Silver City, New Mexico, and at an inlet air temperature of 25 degrees centigrade, 0.25-inch change in pressure (w.g.) is required across the tube. The air flow is listed as 44 cfm. A suitable blower, obtained either at a ham flea market or a supply house, may or may not provide successful results. The information on the nameplate may not be sufficiently accurate. Flow characteristics depend on the restrictions in the flow-path, which, for example, include even elbows and the like. It's always better to measure it yourself. Very few hams can measure flow, say in cfm, on their own. They can measure DP, the pressure drop around the tube anode. It can be done by improvising a temporary manometer of TYGON tubing and water. Pressure drops around the air paths will be different, and the blowers also will no doubt have different flow rates, for example. Air cooled tubes universally require centrifugal blowers; blade fans are high-volume, low-pressure drop devices. Always use the centrifugal blowers. The worst enemy of most electronic devices usually is heat, so these considerations are well worthwhile.

Screen voltage

The screen current may reverse occasionally; this is true of most tetrodes. The screen supply should be designed with this in mind. Svetlana recommends that a screen-to-cathode current path must be provided, and the source impedance of the supply should not exceed 3000 ohms. Linear operation requires a steady screen supply. This amplifier uses an unregulated supply, loaded so that an occasional "blip" of negative current can be absorbed by the bleed current. The screen supply source impedance should be at least 50 ohms in order to protect the screen in case of an arc. On the occasion of incidental or accidental removal of plate load or bias voltage, the screen voltage should be removed--the quicker the better. A screen protective circuit is designed into this amplifier.

Plate operation

The large current emitting surface of the cathode can supply amperes. The tube and associated circuits should be protected by including a series current-limiting resistor of 25 ohms or more. It should be capable of withstanding the high surge-current of an arc or short, and should not be used as a fuse. Svetlana recommends that you test your protection circuit by shorting the plate supply source to ground through a 0.09-mm diameter copper wire, at least 50-mm in length. The copper wire must remain intact.

The tube in operation

This amplifier operates very well (its efficiency is 70 percent). Other characteristics are listed in Table 3. The amplifier is unequivocally stable and has no bad characteristics. It is also very linear. No spurious radiation was detected, nor was any parasitic problem found. Photos B and C show the front panel of the W5LAJ Svetlana linear and its interior, respectively.

A working amplifier

Full schematics of the amplifier, power supplies, and protective circuits appear in Figures 1 and 2. I want to call out certain provisions (also specified by Svetlana) to accommodate the tube requirements. The cathode protective fuse is shown in Figure 1, as are some of the auxiliary circuits (PTT, etc.). The changeover relay (T/R) is shown as constructed. Should two relays be used, the output changeover relay should be the faster of the two. Ideally, the output relay should be a vacuum type, because its operation times are in the millisecond range. The screen supply is somewhat special; it calls on techniques that have long since lost popularity. It operates as choke-input supply (not uncommon), which uses a resonating capacitor (uncommon) across the choke. This has the effect of stiffening the supply. It's an excellent application of a method that was used in the past for high-voltage plate supplies, except this one is for a screen supply. Linear amplifiers do require steady screen voltages. All in all, the 4CX1600B is a very good tube, especially for this service and for other classes, too. It is linear, stable, easily tamed from parasitics, and is trouble-free. It is a delight to operate.

Acknowledgments

I want to thank Dick Linari, WØYXM, for his careful explanations of the vagaries of amplifier design. He is one of the most knowledgeable engineers I've ever encountered. His frequent sessions have been of inestimable value not only to me, but to many silent listeners as well. I would not have been able to use this tube to its full potential, or even understand fully how it works, without his help.

Table 1. Design Numbers from the Data Sheets

Table 2. 4CX1600 Air-Flow Requirements

Table 3. Amplifier Characteristics

**The information provided in this application note is intended for general design guidance only. The user assumes all responsibility for correct and safe usage of this information. Svetlana Electron Devices does not guarantee the usefulness or marketability of products based on this material.

Back to Technical Bulletins Listing