No.25
Amateur Radio
4CX1600B

Introduction

I returned to the air in 1993, after about 20 years of very little activity. Many things have happened during those years which have considerable significance to the radio amateur, such as: the development of reliable, high - performance, solid-state, all - band, all - mode HF transceivers at reasonable prices; affordable home computers; increase of the maximum power limitation for the amateur service from 1000 watts input to 1500 watts output, and the end of the "cold war."

In the past I have built QRP rigs and operated from hilltops with Wes Hayward, W7ZOI, but I have always secretly wanted a "cool California kilowatt" amplifier. (Please don't let Wes know!) Years ago I worked my way up through progressively bigger tubes (4-65A, 4X150A, 4CX250B, and more recently, 4CX800A) to enhance my signal. Reading recent advertisements in QST and versions of the ARRL Radio Amateur's Handbook just about had me sold on the 8877/3CX1500A7 power triode as the tube of choice for such a project. This tube design has very good intermodulation characteristics, and, unlike some of the older tube designs, it doesn't require hundreds of watts just to light the filament! It seemed to be pretty nearly the ultimate final for amateur service.

Then I read about the power tetrodes which are manufactured in Russia, and due to the new relationships between our countries, are now available in the United States. I used one of these tubes to bring new life to an old DenTron MLA-2500B amplifier1. The tubes in this family offer certain advantages of their own, not the least of which are favorable price and superb warranties. The tubes are guaranteed for two years ... and the warranty clock doesn't start running on the invoice date, but instead starts when you first put them into service.

Design Philosophy

As might be guessed from the tubes in my design experience enumerated above, I have favored power tetrodes, mainly because of their high power gain. Nowadays, nearly any of the popular transceivers is capable of about 100 watts output into 50 ohms, with suitably low intermodulation products and harmonic output. (In fact, such transceivers require a 50 ohm non - reactive load and a low output SWR to realize these highly desirable output characteristics.) Therefore, plenty of drive power should be available to excite a legal-limit amplifier, so long as 12dB of gain can be realized.

One satisfactory design for a linear amplifier (and a very successful one!) is the use of cathode-driven (or "grounded-grid", if you prefer that term) triodes. Under normal operating conditions for maximum legal amateur output power, these amplifiers have an input impedance close to 50 ohms. The directly-heated or indirectly-heated cathodes of these tubes can be driven directly, with the cathode being isolated from RF ground by a suitable RF choke. However, it has been shown that linearity is improved substantially by using a low-Q tuned input circuit for each band. This sort of design is used in some commercially-available high-power amplifiers for the HF bands.

Another approach is to use a tetrode with its high power gain, but dissipate all of the driver's power output in a non-inductive grid resistor. This simple method, which has been called "passive - grid - driven," uses the driving power to produce the required grid voltage swing, and it also removes the requirement for any tuned input circuits. Further, it presents an ideal almost constant resistive "dummy load" to the driver. This passive - grid - driven design is used in recent commercially-available, high-performance, maximum-legal-output amplifiers using a pair of 4CX800A tetrodes. Tetrode amplifiers do, however, require a suitable power source for the screen grid.

With this as background, and my success with this circuit in my DenTron conversion, I embarked on the design and construction of the "Sunnyvale/Saint Petersburg Kilowatt - Plus."

Realization

There are some important constraints in designing an amplifier with the 4CX1600B. The first is that linearity of tetrode amplifiers is known to fall off rapidly if the control grid draws excessive current, because the grid is exquisitely effective in intercepting the current emitted by the cathode; a positive control grid can draw a significant portion of the cathode current emitted, even when the grid is only slightly positive. The 4CX1600B is designed with a "striped - cathode," where emission takes place only between grid wires, thus reducing the fraction of the emitted electrons which can be intercepted by the control grid. (The 4CX1500B is similarly designed.) The second reason that the control grid in this type of tube must not be allowed to draw appreciable current is because the rated control grid dissipation is low, only two watts. (The first versions of the data sheet for the 4CX1600B specified the grid dissipation as 100 milliwatts! The engineers at the factory really think you shouldn't draw much grid current! As described above, the design of the 4CX1600B tetrode allows full output power to be achieved with little or no grid current. Compare the control grid dissipation figures of the 4CX1500B [one watt] and the 4CX1600B [two watts] with the 25 watt rating of another tetrode, the once - popular 4 - 1000A!) Any circumstance where measurable control grid current flows is certain to result in nonlinear operation with its attendant splatter and harmonic generation, and might easily exceed the power dissipation limits of the control grid.

This constraint made it important to provide some sort of "grid current prevention" scheme, or, at the very least, a "grid current warning" alarm for the practical tetrode amplifier using the 4CX1600B. To accomplish these objectives in this amplifier design, first, the grid of the 4CX1600B is tapped down on the input resistor, at a point which, with 100 watts of drive, will provide approximately the maximum grid voltage swing without resulting in grid current. A further aid toward preventing grid current flow is to provide some cathode degeneration, that is, letting the cathode potential rise as the input (drive) voltage increases. This is done by the expedient of placing a (noninductive) resistor between the cathode and ground. (Another benefit of this cathode resistor is that the standby cathode current provides additional bias which results in reduced plate dissipation. A further characteristic of this circuit configuration is that the driving power requirement is increased. At full output power, the input voltage swing necessary is larger than if the cathode were grounded. This, in turn, requires a greater driving power.) In addition, a sensitive grid current meter is provided, reading 1.3 milliamperes at full - scale deflection. Finally, a very simple grid - current - activated warning light is also included in the design, where a transistor base - emitter junction is used to detect the occurrence of grid current, and the transistor's collector current (which is ß times as great as the base current) flows through a red LED grid - current warning indicator.

In "receive" mode, the screen grid of the 4CX1600B has a 100 ohm resistor to chassis ground, which keeps the tube cut off to avoid any noise generation. In "transmit" mode, there is a 17.5 kilohm, 15 watt resistor to ground which serves dual purposes. First, it keeps a constant load of 20 milliamperes on the regulator, which allows the regulator to function properly even with up to -20 ma of (negative) screen current. Second, this 20 ma constant load is indicated on the meter as "zero" on the (screen current) meter scale, so that the meter reads actual screen current from -20 ma to + 80 ma.

The heart of the amplifier, consisting of the RF deck proper, control and metering circuitry, and cooling system, are all mounted in a surplus 19" rack-mount cabinet of the sort which can be picked up at surplus stores and hamfests.

Figure 1 shows the schematic diagram of the RF deck. The 4CX1600B is mounted in the Svetlana SK-3A socket, modified as described below (to allow the cathode to operate away from ground potential), and Svetlana's CH-1600B chimney routes the cooling airflow through the anode cooling fins. An additional CH-1600B acts as a chimney extension, discharging the air through the top of the RF deck's cabinet. The cooling fan is a squirrel-cage type of blower which, according to the 4CX1600B's data sheet, should deliver at least 36 cfm. (cubic feet per minute) of cooling air at 25 oC at a back pressure of 0.4 inches of water for 1600 watts dissipation.

The low cost, stock filament transformer specified actually produces 13.5 VAC (with my mains voltage), so two 0.1 ohm 5 watt resistors were added to the circuit to drop the voltage at the filament terminals of the 4CX1600B to the 12.6 VAC recommended by the tube manufacturer.

The input grid resistor is a 51.6 ohm unit having a dissipation capability exceeding 100 watts. It consists of 3 (three) Caddock MP850 resistors; two 71.2 ohm units in parallel, in series with a 15 ohm unit, all mounted on a surplus heat sink (5.0" x 5.5" x 0.75" or 12.7 x 14.0 x 2.0 cm). This "dummy load" is mounted below the chassis, near the SK-3A socket. (The grid resistor has its own small cooling "biscuit" fan. The air below the chassis is pressurized by the main blower, to provide cooling of the tube, but this auxiliary fan, which runs continuously, serves to keep the air "stirred up" to prevent any stagnant hot air below the chassis.) The grid of the 4CX1600B is tapped at the 35.6 ohm point of the input resistive divider. As an aid to stability, a 10 ohm 2 watt composition resistor is placed in series with the grid lead. This connection resulted in an input SWR of about 1.0 at 1.9 MHz, which increased to just over 1.6 at 29.6 MHz, mainly due to the reactance of the 86 pf input capacitance of the 4CX1600B. No frequency compensation was deemed necessary! This is done without the requirement for tuned circuits for each band or input bandswitching!

The cathode resistor is made up of four 15 ohm 3 watt noninductive metal oxide film resistors, from the cathode terminal ring on the socket to each of the 4 socket mounting screws.

The plate tank circuit components include a heavy-duty bandswitch, a silver-plated inductor for the high bands, ferrite toroidal inductors for the low bands, and a plate choke wound on a delrin form. These components are those used in a Command Technologies HF-2500 amplifier, but other suitable components could be utilized. (As it is currently configured, the plate tank cannot be tuned to 30 meters. Operation at full power on this band would require another position on the bandswitch and another tap on the tank coil, or compromises on other bands. These are options which the author considered to be unnecessary and undesirable, because for U.S. hams the power limit is 200 watts on this band.) The anode connector is a Svetlana AC-2, and the plate parasitic choke is two turns of tinned copper strap (0.032" thick x 0.24" wide or 0.8 mm x 6.0 mm) over three 91 ohm, 2 watt composition resistors in parallel. (Any value from 47 ohms to 100 ohms is likely to be satisfactory.) The antenna change - over relay has a 115 VAC coil (12 VDC would be fine!) I used this one because it had nice, wide, gold - plated contacts.

First digression: The power supply

[Remember that almost every voltage inside a power supply for a high-power linear amplifier is lethal! Turn it off, unplug it, and short it out before you touch anything! I always apply the "one hand in the pocket" principle when working on anything above 24 volts! ]

In my junk box I had what I thought was an enviable assortment of high power transformers. As it turned out, they were not very useful for the modern tetrodes, which operate efficiently at lower anode voltages than those of the last generation. I could easily produce 4000-5000 volts at several hundred milliamperes, just right for a 4-1000A, but I couldnÕt come up with 2000-3000 volts at an ampere.

I convinced myself (and the XYL) that I had to bite the bullet and buy a proper transformer. I ordered a Peter W. Dahl ARRL-002. The day it arrived, it stayed on the front porch until I got home from work, because my wife couldnÕt move it. It weighs 46 pounds, but with modern silicon-steel and two tape-wound core sections, that is only about one third of what my old 3KW (9KV CT) transformer weighs. The Dahl unit is much smaller, and is a work of art!

As shown in Figure 2, there is a simple "step-start" circuit to limit the current surge charging the filter capacitors. The transformer's output is rectified by a bridge of K2AW's "Silicon Alley" 10KV diodes, and the filter capacitor is made up of a string of ten 470 µf, 400 volt electrolytics taken from a laser power supply board which was available at a local surplus store for $14.95! (Alltronics, Santa Clara, CA) The voltage is divided equally across the capacitor string by 25 kilohm, 25 watt resistors which also serve as the power supply bleeder. (This divider results in a considerably higher bleeder current than the typical 100 kilohm resistors often seen. The result is a "stiffer" power supply, but more heat is generated.)

The junk box produced a transformer which had output windings of 275 volts @ 60 milliamperes, 6 volts @ 2 amperes, and 35 volts at 150 milliamperes. These windings were dedicated to a regulated 350 volt screen supply, a regulated 12 volt DC supply for relay and indicator lamps (by means of a full-wave doubler and a three - terminal IC regulator), and the control grid bias supply. The circuitry for these supplies is very straightforward, and is shown in Figure 3. These supplies were built in the same cabinet as the plate supply. All power supplies are cooled by a muffin fan on the rear panel of the cabinet. The fan probably isn't necessary, but I prefer cool components; they're sure to last longer. The major source of heat in this cabinet is the bleeder - resistor chain, which dissipates about 36 watts when the plate voltage is 3000 volts. High voltage is monitored with a 200 µA surplus movement (see "metering," below) through a Caddock MX430 20 megohm multiplier resistor.

All power to the RF deck is supplied from the power supply cabinet. There is a standard IEC 120 volt AC cable for the 4CX1600B's filament transformer and the antenna changeover relay, an "auxiliary" power cable, and a high tension line for the anode voltage. The shielded auxiliary power cable carries the screen and control grid bias voltages and the 12VDC, and the ground. The high voltage line is a beautiful 40KV #18 AWG wire (from a local surplus store) with Millen 37001 connectors at each end.

In this design, because no control grid current flows, the control grid bias voltage (nominally -56V) is provided by the simple expedient of a half - wave voltage doubler, low - power zener diodes, and a potentiometer to allow grid bias adjustment to achieve the desired no - signal cathode current. The (rather common) practice of using a zener diode in the cathode circuit to provide operating bias was rejected because of the requirement for actual resistance between the cathode and ground.

The screen supply provides a DC voltage of 350 volts by means of a series electronic regulator. The regulator has a current limiting feature, wherein the output voltage falls if the screen draws more than 60 milliamperes. This prevents the screen grid dissipation from exceeding its maximum rating of 20 watts.

Second digression: The SK3A socket

Because the SK-3A socket has the cathode effectively tied to the chassis ground (via the socket's mounting plate) and an internal bypass capacitor for the screen grid appears between the screen grid and the cathode, the socket, as supplied, is not suitable for this application, where it is desired to provide some degenerative feedback via the cathode. The following procedure can be used to modify the socket for such applications, however. You will need to acquire some additional parts for this procedure: four (4) insulating shoulder washers for 4 - 40 screws (made of teflonâor another insulating material.)

1. Drill out the four rivets holding the screen ring to the screen contactors at the very top of the socket.

2. Now working at the bottom of the socket, remove the four nuts from the machine screws holding the socket assembly together.

3. Disassemble the socket; a) first remove the cathode contact ring, marking its position relative to the underlying bakelite layer b) remove the bakelite socket layer which has the factory markings and serial number, marking its position relative to the socket mounting plate. (This is the 0.060" [1.5mm] silver-plated brass plate.) c) carefully remove the screen contactor assembly, freeing the contactor "ears" by springing them outward (Don't drop the capacitor! It is the ceramic annulus with the silver plating on each side. It is very brittle!) d) remove the spring plate, the capacitor, and the other spring plate if they didn't come with the screen contactor assembly in "c" above e) remove the mounting plate assembly, marking its position relative to the remaining socket assembly

4. Drill out the four holes in the mounting plate assembly using a #14 drill (0.180"). These are the second set of holes in from the outer edge, through which the socket assembly screws pass. (The screws should still be in the top layer of the socket, with heater, grid, and cathode contactors.) 5. Put the teflon shoulder washers on the screws. Now, when the socket is reassembled, the cathode will be isolated from the main mounting plate and the screen bypass capacitor.

6. Replace the capacitor assembly in the order: spring, capacitor, and spring. Now replace the screen contactor assembly and the bakelite bottom section, taking care to align this section with your previous mark, and carefully guide the socket solder tabs through the bakelite bottom without bending them.

7. Cut the outer tabs off the cathode ring contact. After all of this work, you don't want this ring (the cathode terminal) to be grounded when you mount the socket in the chassis! Place the modified cathode contact ring over the screws.

8. Replace the washers and nuts on the socket assembly machine screws and tighten each a little at a time, until the assembly is snug.

This completes the socket conversion. Characteristics are now as follows: a) The screen ring on the 4CX1600B is contacted exactly as before. The upper ring of the screen contact assembly can be re-riveted, but this requires some very small rivets. This metal ring could also be soldered to the contactor where the rivet joint previously was. The internal screen bypass capacitor still appears between the screen grid and ground (via the socket mounting plate.) b) The cathode annulus on the 4CX1600B is contacted exactly as before, but the electrical connection for the cathode is now isolated from the chassis. The cathode contact on the socket is now made through the thin cathode ring on the bottom of the socket. (The ring is silver - plated and easily soldered, convenient for an application like the present one, which requires multiple contacts.)

The heater, control grid, and screen contacts function exactly as original.

Control circuitry

The control circuitry is quite simple and is shown in Figure 4. The entire unit is turned on with the main switch/breaker on the power supply cabinet; when the switch is thrown, all voltages are ready (after the "step-start" delay, that is.) If the AC cable from the RF deck is plugged in to the power supply, the 4CX1600B's filament begins to heat, the cooling fans go on, the time delay starts, and anode voltage is applied to the 4CX1600B. After the mandatory three minutes, the +12 VDC (nominal) control voltage is enabled by the time delay relay. At this time, the control circuitry (consisting of transistors Q1 - Q5) decides whether it is acceptable to apply screen voltage to the 4CX1600B and to activate the antenna changeover relay. Q5 is the main switch which activates the T/R relay K2 whenever 12 volts is available (i.e. after the 3-minute warmup period) and its base is high. The high - base condition can occur only when all of the following additional conditions are met: 1. The anode voltage for the 4CX1600B is available. This condition is sensed by the resistive divider R4 / R5 on Figure 1. If the "HV sense" line is low, then the amplifier/switch combination made up of Q1 and Q2 holds the base of Q5 to a very low level, preventing application of screen voltage.

2. The negative control - grid bias is present. If this voltage is near zero, transistor Q3 is saturated, and again Q5 is turned off. 3. The T/R switch from the exciter has pulled the base of Q4 down, allowing its collector to rise.

To summarize, Q2 , Q3 , and Q4 must all be cut off to turn Q5 on, thereby putting the amplifier in "Transmit" mode. Until these conditions are met, the tube is cut off; the screen grid is essentially at ground potential, (and the control grid should have -56 volts applied.) Under no other conditions is the tube drawing any current (except for the filament, of course.)

[The functioning of this circuitry was tested at full operating voltage with no drive applied. When RF was applied, however, it was apparent that I had forgotten that the control circuitry could also respond to RF potentials! The result was a rather dazzling audio - visual presentation of flashing indicator lights and clicking relays! I spent the next few hours putting ferrite beads and bypass capacitors on all leads in and out of the control circuit board. That tamed the beast!]

Metering

At a local surplus store, I obtained some attractive meters with 200 µA movements. Their internal resistance is 2000 Ohms. (This information is all that is needed to use them as indicators for almost any desired full - scale current or voltage. See the "Measurements" section of the ARRL handbook for how to do this.) I used one movement as a voltmeter on the power supply (0-4KV, as described above), one as a triple-purpose multimeter to measure anode current (0-1.3A), screen grid current (-20 to +80 mA), and control grid current (0 - 1.3 mA), and a third movement to indicate forward (0 - 1500 W) and reflected power (0 - 150 W) at the output coax connector. After D.C. calibration against a digital multimeter, I carefully remove the cover and face of each movement, and attach a home-made laser-printed scale, thereby coming up with "customized" meters at low cost. (Yes, it does take time!)

Grid Current Warning

The circuitry for this indicator light is very simple, and is included in figure 4. When control grid current flows, it develops a potential across R11 (on figure 4), and the collector current of Q6 lights a red LED indicator brightly when grid current is about 1.0 milliampere. (Although the battery is always connected to the circuit of transistor Q6, the current drain is negligible, being just the collector - emitter leakage current, so battery life should be very long. If you don't like the floating 9 volt battery, a small DC power supply could be included, or the smallest "wall-wart" type of DC supply could be utilized, and built right into the cabinet. It must however, be capable of "floating," at the grid potential, about 100 volts away from the chassis ground potential.) During tune - up, when the LED flickers on voice peaks, it's time to back off on the microphone gain control slightly, (or on the transceiver's RF output control, if you're using CW; that is, reduce the drive. Note that, in CW mode, many transceivers will put out a high-power spike upon initial key closure, even when the RF output is set to quite low values. This undesirable transceiver "bug" can be turned into a "feature," because it gives, via a brief blink of the LED, the premonitory warning "you are approaching the positive grid - current area.") The circuitry for the grid - current warning indicator is built into a small aluminum minibox, and RF energy on the leads in and out is prevented by feed - through capacitors and RF chokes.

Results

The zero-signal plate current is about 280 mA, resulting in a zero-signal plate dissipation of 840 watts. On 40 meters, for example, at the full 1.5 kW output, the plate current is about 0.8 amperes, and the anode dissipation is still under 1000 watts. (Of course, until the TR switch is activated, the screen voltage is zero, and the tube is effectively cut off, so there is no plate dissipation except during "transmit" periods.) After a period of operation, I run the fan for a few minutes (i.e. run the amplifier in "standby" mode) to ensure a cool tube before turning the amplifier off.

More complete performance figures for the amplifier are presented in Table 1. Factors other than the realized performance which make the building of the "Sunnyvale/Saint Petersburg Kilowatt-Plus" an economical project with attainable goals are summarized below:

- First, the screen and control grid supply requirements (for those wishing to build their own version of the amplifier described) can be met by a small power transformer like that used by the author, or by connecting multiple inexpensive surplus low voltage transformers connected "backwards" with their low voltage windings in parallel across the filament transformer's 12.6 VAC winding, and their 110 volt AC windings in series to provide the required voltages.

- Second, the "passive grid driven" configuration completely circumvents the need for separate input tuned circuits for each band, together with the accompanying bandswitching, which are major but necessary complications with a cathode - driven amplifier.

- Finally, the low idling plate and screen dissipation will have the direct result of extended tube life and operating economy.

Acknowledgments The author would like to thank George Badger, W6TC, Pat Stein, N8BRA, Pete Dahl, K0BIT, and Bob Alper, W4OIW, for their support and encouragement in this project.

References Cited 1. Daughters, GT. New life for Dentron MLA2500s. QST, May, 1996: pp45 - 48.

TABLE 1.

4CX1600B, Class AB1, Passive Grid-Driven Service

Figure 1
Figure 2
Figure 3
Figure 4

Legends for Figures

**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