The Heathkit SB-200/201 Vintage Linear Amplifier
A Great Buy at Today’s Prices
Tom Sowden (K0GKD)
4450 West 188th Street
Stilwell, KS 66085
Sooner or later it seems that every amateur that operates on the HF bands wants to try to boost power with a tube type linear amplifier. Getting through the QRM can often make the difference in enjoying a good solid QSO in today’s crowded bands. Also it just feels great to get good signal reports. With prices for new commercial linear amplifiers right up there with new solid state transceivers, considering used vintage equipment presents an option that works a lot better for those of us with limited budgets. I suspect that many amateurs are fearful that older amplifiers will not work, or if they do, not for long. If you understand the circuits and get comfortable with the nuances of the older vintage radios you can keep them going for years with a minimum of effort, and they will perform with the best of the commercial units watt for watt.
One of the best buys on the market today is the Heathkit SB 200 and SB201. These old workhorses were built 20 plus years ago, and there are still a whole bunch of them out there doing yeoman’s duty on a daily basis. These amplifiers often sell for around $300 or less. When you consider that a new high voltage transformer alone would cost you more than this it seems unreal to me. At this price about the only way you could lose would be if the power transformer were inoperative. Further, the SB 200 uses a pair of relatively inexpensive 572B tubes, which can put out a nice solid signal approaching 600 watts (output).
I recently purchased a SB201 on Ebay for $225. The added freight brought the total to $260. Naturally I was concerned about the power transformer so I e-mailed the seller before bidding to ask about its condition. He would only certify that when the amplifier was turned on the power meter indicated about 2,500 volts. That was enough to convince me that the power unit was okay. When the Heathkit SB201 arrived I took it apart and started a visual examination. A good rule to follow with used equipment is to never plug it in without first inspecting the components thoroughly, and running whatever tests seem necessary. When working with high voltage equipment safety should be in your thoughts at all times. I follow rules due to the dangerous potentials involved. They are:
· Never work on the unit when it is connected in to a source of electricity.
· Use a “chicken stick” (QST September 2003 see graphic) to make sure every contact that you intend to touch with your hands is not “alive”. This is especially important around the high voltage section where electrolytic capacitors could be still charged if the bleeding resistors have not dissipated the stored voltage. Often the “bleeder” resistors are non functional or the solder joints have come undone leaving the capacitor charged with very dangerous voltages. (Tim Reynolds WA9EEH suggested that a 10 ohm 20 watt resistor should be placed in series with the shorting wire to avoid damaging charged electrolytics.)
· I wear rubber soled shoes so that my feet are not grounded. I never stand on a wet surface for obvious reasons.
· If possible, I only use one hand keeping the other more or less behind my back. If this is not possible I do a double grounding check with the “chicken stick” to once again make sure there are no voltages present on the areas that I may contact.
· I never work on amplifiers if I am tired.
· When possible I have a friend help.
It helps me to better understand problems that may arise if I know the basic flow of the schematic. The only way I can sort out understanding circuits is to divide them into segments. While a lot of what follows is fairly basic to a lot of hams, thoroughly understanding the circuit helps with potential problems and repairs.
Lets start with the high voltage power supply, and segment this from the overall schematic.
Figure 1 High Voltage Transformer
A look at the transformer connections indicates the manner in which one can use either 110VAC or 220VAC because of the dual primary windings. If 220VAC is the choice the primary windings are simply connected at the mid-point (2 & 3) in series so that they appear and act like a single winding. Conversely, the midpoints can be wired in parallel for 110VAC (1 & 2, 3 & 4). You can see what the engineer’s were doing by lowering the ratio in half between the primary and secondary for 220VAC, and doubling it for 110VAC. This, of course, results in the same voltage across the secondary of the transformer windings.
Since the amplifier demands a lot of amperage I prefer to use 220VAC in the shack. Regardless of which voltage you decide on a separate dedicated supply should be installed. Place a circuit breaker nearby that is wired with a feed line to a female plug to accommodate the linear. The amplifier should have an appropriate male plug. This will allow you to disconnect the power leads by simply throwing out the circuit breaker, and unplugging the unit. Most hardware stores carry 15 amp disconnects which work well with the 220VAC option. A plug designed for 220VAC should be purchased if you plan on connecting to this voltage. Keep in mind that each leg of the 220VAC measures 110-120VAC to ground, and 220-240VAC across the two elements. When wiring the plug, normally the green wire is connected to ground, and the white and black wires to the respective terminals. For 110VAC make sure you have a plug with a ground connection. The black wire is usually connected to the copper plated terminal. It is also a good idea to follow the leads to their termination in the linear for confirmation. If you are not sure about the wiring have an electrician help you out.
The high voltage power supply circuit theory is fairly conventional. The secondary windings have about a 8:1 or 4:1 ratio to the primary windings depending on how you wired the primary. The root mean square voltages across the secondary is going to be about the same at 880 (8 x 120 or 4 x 220), or higher if the line voltage is higher. Most amplifiers use voltage-doubling circuits to avoid very high voltage transformers, which are expensive. The SB 201 is no exception. To fully understand how these work one only has to follow the flow of electricity during the duty cycle of the AC. Lets examine this for a better comprehension of the factors involved.
Figure 1A AC Duty Cycle
We know that household electricity is alternating current at 60 cycles per second. During the positive phase the voltage starts at zero and quickly rises to its peak of about 158 volts (note: when we measure AC voltages we are looking at the average or root mean square of the voltage, not its peak which is about 1.44 times greater). As the voltage rises and falls it sets up a magnetic field, which induces voltage in the secondary of the transformer in relationship to the ratio of the windings. In the case of the SB200 the peak AC voltage will equal about 8 times 156 (110VAC x 1.414) or 1,244 volts, or 4 times 311 (220 x 1.414), also 1,244 volts. As the voltage increases on the positive portion of the cycle, the rectifier diodes wired for the positive voltage, as shown on the schematic below, begin to conduct. Current flows through the electrolytic capacitors to the secondary return winding (shown in between the two sets of three capacitors). Since the capacitors are hooked up in series with equalizing resistors (30,000 ohms at 7 watts) the voltage across each capacitor equals about 1/3 of the peak positive rectified voltage. If one could measure the voltage instantaneously across the three capacitors (DO NOT) they would find a total voltage of about 1,244 volts DC. (Note: the equalizing resistors also act as “bleeder” resistors after the power is turned off dissipating the voltage stored in the capacitors.)
Figure 2 Rectifier Voltage Doubling Circuit
When the AC cycle goes negative the positive directed diodes stop conducting, and the negative directed diodes start up allowing current to flow into the negative terminals of the second set of three electrolytic capacitors. This current terminates like the positive phase through the return secondary winding. Voltage measurements across these three capacitors will be the same as the first three, or about 1,267 volts DC. When you measure across all six capacitors (do not as this would be dangerous) the sum of the two sets of three capacitors will measure 2,534 volts (1,267 plus 1,267) from the top of the first set to the bottom of the second set. The effect is to double the voltage of the power transformer secondary. A pretty slick trick! Of course, if the line voltage is higher than 110 or 220 than the total voltage across the capacitors will be higher by the multiples involved.
This type of circuit is used in many different applications to double, triple, or quadruple voltages. The main disadvantage of these circuits is voltage regulation where relatively high current demands can draw down the voltages. This is understandable if you think of the capacitors as if they were batteries. Of course, the more capacity they have the more they can deliver without dropping the voltage too much. Heathkit engineers tended to use electrolytic filter capacitors with only about 100 microfarads, probably because they were readily available at reasonable prices. Since there are a total of six capacitors, effectively in series, the net capacity is about 16 microfarads (100uf/6). This tends to allow the high voltage to drop 5-10% on key down or on SSB voice peaks. Although this voltage drop is not acutely noticeable at the receiving end, it does impact the linearity of the amplifier nominally, and to some degree the audio quality. (Tim says: Sagging 5% or 10% regulation on the power supply usually will not cause a change in audio quality. Unless the amp is under loaded to begin with. When loading a amp, load to maximum output key down then increase the loading about 5%. This will help linearity and prevent flat topping.)
Open Amplifier View (Picture 3)
With computer grade electrolytic capacitors available at reasonable prices I like to replace the older capacitors with brand new ones that have considerably higher capacitance. Replacing these older parts is a good idea anyway as more than likely they are the same age as the linear. There are usually telltale indications of age with signs of leakage or heat stress where the containers have expanded. I recently purchased six new capacitors that were rated at 100 degrees C, 450 volts at 400 microfarads for about $5.00 each. You can readily see that these new capacitors have over four times the capacitance of the original ones. Therefore they can store much more energy and even out current demands, especially on SSB voice peaks. Unfortunately, one consequence of the higher net capacity is the surge of current it takes to charge the capacitors when you first turn the amplifier on. When I first power up my amplifier it infrequently kicks out the circuit breaker. This is not a big problem. I simply turn the linear off, and reset the circuit breaker, and than turn the amp on again. Since the capacitors are partially charged up, the second time I flip the on switch the surge in current is much less and the unit stays on. This is probably not a good practice and there is a much better solution, which better protects the components. You can install a “surge” circuit to negate this affect. Harbach Electronics (www.harbachelectronics.com) offers a kit, which works great by having power resistors in series with the load for a short time period to restrict the initial flow to about six amps. Alternatively, your electrician may have a solution with a time delay relay that shorts out wire wound resistors placed in series with the load to initially drop the voltage on power-up.
If you decide to replace the capacitors in the circuit make sure you connect the correct polarity positive or negative. The electrolytic capacitors are marked with either a “-“ or “+” sign by the respective terminals. If you are not sure of the circuit board polarity just double-check the schematic with the capacitor placement. It is also a good idea to go over the rectifier diodes. You can do this while they are in the circuit by simply using a ohm meter set to the 1,000 ohm range. (Be sure capacitors have been discharged with the “chicken stick”). By placing the positive (red) lead on the negative side of the diode (away from the arrow marking), and the black or negative lead on the positive marking you should read a very low resistance. Reversing the leads of the ohmmeter should show a much higher resistance. By checking each diode this way you can insure that they are all functional. Diodes that have broken down due to voltage or current excesses almost always show a direct short. If they need replacing do so with higher values than the original diodes, and replace all of them. I like to use the ones that have at least a three-amp current rating at over 1,000 volts. You can buy them for about $0.35 each. Keep in mind they are hooked up in series so they can take the high voltage across the capacitor bank of approximately 2,500 DC. Four 1,000 PIV rated diodes give you more than adequate margin (4 x 1,000) to handle the 2,500 volts. (Tim says: Power supply diodes should by shunted with 470K 1/2 watt resistors to equahze reverse breakdown voltage. The exception is possibly when you use enough of them - two or three times the rms voltage of the transformer. Also a .01 uf cap 1KV across each one should be added to equalize the switching times and suppress to the rf hash. Many amp manufactures use controlled avalanche diodes that have equal switching times and sometimes omit these parts. )
An understanding of the bias supply will help you identify possible problems in this area. This part of the circuit takes off a 120VAC from a separate set of windings on the power transformer, and rectifies this voltage for the bias supply. Note the direction of the rectifier diode with the positive side facing the secondary thereby only passing the negative voltage. This is confirmed by the electrolytic capacitor with the negative side up and the positive side grounded. The 10,000-ohm resistor is used for voltage regulation and as a “bleeder” to bleed off the capacitor and its stored voltage when the unit is powered off. Even though this bias supply is markedly less voltage than the high power part of the supply it can still be dangerous if touched.
After the power is off (and the unit unplugged) use the “chicken stick” to make sure the capacitor is grounded out before touching any part of this circuit. Look this area over carefully. It is probably a good idea to change out the bias filter capacitor too. Like the high-powered rectifier ones, it is probably old and may eventually fail due to age mostly they tend to dry out over time. You can usually find a substitute in the parts box that will handle the voltage (150VDC). Make sure the negative side is connected as indicated. Also check the rectifier with the positive/negative leads of the ohmmeter to make sure it has not shorted out due to excess current or voltage.
Figure 3 Bias Supply
The negative bias supply is used to shut the 572B’s down during the receive cycle, and to limit their current at idle when in the transmit mode. You can follow the bias voltage in the schematic through the rough pencil marks on the schematic.
Visualize the bias current flowing around through the relay activation coil, the 2,000-ohm resistor in parallel with it, and to the grids of the tubes via a 1.5-ohm resistor. The resistor is in series with a choke (RFC-2), which blocks the “RF” from getting back into the bias supply. The bias is also connected to the external antenna relay terminal through the 33-ohm resistor as shown in Figure 4. In the “receive mode” the () 135VDC is present at the grids of the tubes in more or less full force since there is virtually no current flowing. When the push to talk switch on the exciter is activated the external transceiver relay or circuit grounds the bias through its various resistances the relay coil in parallel with the 2,000-ohm resistor, and the 33-ohm resistor in series with the lead to the antenna relay terminal. The effect is to drop the bias voltage significantly so that the tubes begin to draw about 80 milliamps without drive, or at idle. Once the drive is fed to the amplifier the 572B’s will draw about 600 or so milliamps of plate current.
One of the problems grounding the bias circuit through newer transceivers is the relatively high voltage present. My Ten-Tec Jupiter will not take 135 volts! The method used to close the circuit is through a transistor switch within the Jupiter. The high voltage negative bias would likely damage the circuit so a modification is necessary. I have found the easiest approach is to use a small 9 VDC relay, readily available at Radio Shack, where the terminals of the relay can provide the grounding function. By connecting the bias voltage feed line that goes to the back of the unit to one of the contacts on the relay, and the other relay contact to ground, you keep the voltage isolated to the relay. Use a nine volt battery with the negative lead grounded and the positive lead wired in series with the relay coil with the opposite lead connected to the “antenna relay” terminal on the back of the amplifier. When the 9 VDC is grounded by the external transceiver the relay is activated grounding the 135VDC bias through the relay contacts and not the transceiver. See the diagram below (receive mode).
The battery should last at least a year under normal conditions. Alternatively, the Jupiter, as do most modern transceivers, has its supply voltage of 12-13.5VDC available at the back of the unit. You could use a 12-volt relay by providing the “transceiver” voltage to activate the relay. The return lead for the 13.5VDC could be routed to ground through the transceiver’s circuit. The terminal provided at the back of the amplifier for the ALC circuit provides a good access point to bring in the 12-13.5VDC from the transceiver source. The ALC circuit does not seem necessary with solid-state transceivers due to their relatively low output levels. I use shielded cable for both the voltage and the antenna grounding wires to avoid the possibility of picking up stray RF. Of course, if your transceiver can handle the higher bias voltage than this modification is not necessary. If you are not sure than use the additional relay.
In further assessment of the unit I examine the parasitic resistors used on the plate feed for the 572B tubes. The circuit appears below:
These resistors are in parallel with the small RFC chokes where they connect to the plates of the tubes. Normally they are carbon 2-watt 33-ohm resistors. More often than not these are burned out because of runaway currents carelessly caused when the amplifier was out of resonance. If so, they need to be replaced.
A look at the two resistors in the grid circuit feeding the bias voltage is often productive, as these tend to burn out too. Note the detail.
Heathkit did a good job with heavy-duty wire wound resistors in place of lower wattage carbon resistors, so they should be okay. Check them anyway. Finally, checking the continuity of the plate choke to make sure it has not severed, and a reading of the plate feed point at the tubes to ground just takes a quick minute. The latter should read around 100,000 ohms or so (if there is an interlock switch the reading will be much less or near the value of the resistor in series with the switch. By pushing down on the interlock as if you had replaced the final tank shield you should be able to get the nominal 100,000 ohm reading to ground.)
Once you have examined the unit for these types of potential problems, and possibly replaced the filter electrolytic capacitors, and made all of the final checks, you can fire up the amplifier without the tubes. I like to disconnect one of the high voltage leads from the power transformer for starters. This prevents the filter capacitors from charging, and allows you to check the bias voltage without the high DC voltage being present. Of course, the high AC voltage is still “hot” and dangerous that is present on the transformer terminals.
Before going further make sure the amplifier is grounded separately in addition to the electrical ground (through the power leads). Connect the jumpers for 110VAC as per the schematic. Use a 110V three wire plug (ground the green wire). Screw the shield back on top of the amplifier final tank circuit (as noted earlier, in many units, there is a safety interlock switch that ground out the high voltage supply if this shield lid is not screwed on). Place the amplifier on your workbench in a manner that will allow you to safely check the bias voltage.
Make sure the power switch is off and plug in the unit. Turn on the power switch. Than carefully with one hand check the negative bias voltage with your voltage meter. It should read about 135 VDC or so.
It is also a good idea to check the filament voltage windings. Remember you are checking AC voltages on the filament windings, and DC on the bias supply, so change the voltmeter settings accordingly. Thanks to Mike Hardtke (N7LEQ) he noticed an error in the original writeup on filament voltages. You should get a three plus volt reading on each side of the grounded center tap. The filament wires coming off of the transformer will read six plus votls ac across the two green wires (note: the voltages combine in phase on the tube filaments to equal the six plus volts needed and only three plus from green to ground). Figure 7 in the manual is somewhat of a misrepresentation as the center tap voltages are three plus volts and not six plus.
I have found some filament voltage measurements as low as 4.5 volts, and the tubes still function okay. The “Handbook” indicates that +/- 10% is acceptable, but my experience is it can be quite a bit less without damaging the tubes. If the bias and filament voltage check out, turn off the power and unplug the unit. Once again use the “chicken stick” to ground all areas where you will likely come in contact before proceeding.
The SB201 Filament Choke
Keep in mind that because the tubes are “triodes” the filaments act as the “cathode” so they must be “raised” above RF ground. This is done with a heavy wire wound choke on a metal core. The choke feeds the filaments with the current they need, and at the same time provides a inductive reactance to the RF drive voltage being fed directly to the filament, and keeps it from grounding out through the center tap of the transformer windings.
Somewhat the same situation exits at the grids of the two triodes. You cannot physically “ground” the grids to achieve “grounded-grid” operation because you would also be grounding the bias voltage. Accordingly, the engineers use capacitors to ground the RF but still isolate the DC bias. I prefer replacing the smaller value Heathkit silver mica capacitors (200uuf in this case) with .001 bypass capacitors. This reduces the capacitive reactance to virtually zero throughout the HF range. In my experience it tends to stabilize the amplifier at the expense of requiring a slight increase in drive wattage. I have found that the silver mica capacitors are often damaged and need to be replaced anyway. You should take the time to check the standing wave ratio between the transmitter or transceiver and the amplifier. While this step is probably not necessary, it will likely permit you to use much less drive by reducing the reflected power between the exciter and the amplifier. A lot of the older linear amplifiers were designed to be used with pi-network tube-type exciters. Accordingly, there can be a fairly sizable mismatch with newer equipment.
With one of the high voltage leads still disconnected, the unit unplugged, and the “chicken stick” test completed, put the tubes in their sockets. Make sure the pins line up, as it is easy to plug the tubes into their sockets improperly. You can connect the plate leads with the parasitic resistor and choke. Screw the lid back on to the tank circuit. Once again, do not connect the high voltage lead that you took off earlier. Put a SWR meter in series with the driving transmitter or transceiver. Connect the output of the transceiver to the input terminal of the amplifier through the SWR meter, and a dummy load to the output coax terminal. Prepare an insulated lead with two alligator clips and ground one of the leads to the chassis. Connect the other end to the relay terminal on the back of the exciter. Plug in the amplifier and turn it on. The relay should engage. Turn on the exciter and reduce the drive to a minimum about 10 percent or so. Note the SWR for 80, 40, 20, and 15-meter bands. Adjust the coils for each band for minimum reflected power using low power setting of the driver transmitter or transceiver.
L1-L4 Left Filament Choke Right With Grid Resistors (Picture 6)
Sometimes the coils (L1 L4) are glued shut, and they will need to be worked loose. If they cannot be adjusted they may have to be replaced providing you cannot get sufficient drive. Turn off the amplifier, remove the jumper across the relay terminal, and unplug the unit. Once again use the “chicken stick” to ground out all potential sources of voltage. Now you can hook up the high voltage secondary. Everything is now ready for the final test. Also you may want to first view Tom Boza’s web page - NE7X (www.ne7x.com). Tom ran an analysis of the input circuit and found the higher band’s considerably off resonance.
With the unit grounded, and a adequate dummy load hooked to the output of the amplifier (I use a “Cantenna” Heathkit load resistor), route the coax from the transceiver to the “input’ connector, and terminate the “output” to the dummy load. Note carefully the markings on the terminals on the back to make sure you get the input and the output in their respective places. Connect the relay terminal to the exciter or transceiver as noted in your manual using shielded cable note: the circuit should be open in the receive mode. The shield can act as the return path for ground. Plug in the amplifier. Hook up a microphone to the transceiver with a “push to talk” feature. After double-checking all of the connections turn on the power of both units. Make sure that the transceiver and amplifier are set to the same band. The amplifier should show the high voltage in the “HV” setting of around 2,500 VDC, and the plate current should be “0”. (Always keep in mind that you are dealing with very high voltages that can be lethal if touched. Do not try to measure these voltages with a standard volt-ohm meter. Most off-the-shelf meters are not designed to assess these very high voltages, and you clearly do not want to expose yourself to possible arcing and other unpleasant experiences. You will also damage your voltmeter irreparably. The SB200 like most amplifiers has built in voltage meters that tell you the high voltages across the capacitor bank).
Make sure the drive level of the transceiver/transmitter is at the lowest power setting and in the CW mode. Key the transceiver on. The amplifier relay should engage and the plate current should now indicate about 80 to 90 milliamps. Very slowly increase the drive level until the plate current of the amplifier starts to rise. Immediately dip the tuning capacitor of the amplifier for minimum current, and the load capacitor for maximum current. At this stage follow the load up instructions in the operating manual, which probably suggest you tune to maximum RF output. The amplifier should load to 600 or so milliamps of plate current. Do not continue with this level of output for an extended period, as the amplifier will possibly overheat. Note the color of the tubes, as they should start to glow a pinkish hue. If either tube does not have color the same color as its counterpart it may need to be replaced so that a better match between the two tubes will exist. If the tubes need to be replaced you might consider used ones. They tend to have less gain and are therefore more stable. They can usually be found on eham.net for about $25 from reputable amateurs. New Chinese matched tubes work well as they seem to have less gain. They are normally priced at around $80 for a matched pair.
Place the transceiver in SSB mode and reduce the mike level to the minimum. Key the microphone and begin speaking and advancing the mike level or drive until voice peaks move the plate current to 250-300 milliamps. If there are no problems you are set to go. Unplug both the amplifier and your exciter, and allow the high voltage to dissipate do not disconnect anything until the “HV” indicates zero. If you are going to use 220VAC you will have to change the primary connections within the amplifier, and the male and female plugs. In operation you will normally have more than enough drive with a 100-watt exciter to overdrive the linear. Running the amplifier at a reduced level will extend the life of the tubes for years, and make very little difference in the output of the unit. You should get 800 to 1000 watts input consistently, and your contacts will note the clean signal and good audio that results from properly driven linear application.
If you experience problems they will likely be in the areas of the amplifier discussed herein the power and bias supply, the parasitic suppressing circuits, the new relay setup, and the tube grid circuit. The final tubes can sometimes short out.
If at any time during the entire process you are not sure that you understand what you are doing get help from a fellow “ham”. It is a good safety idea to have someone run the tests with you. And always remember to be aware of the high voltage and follow good safety practices. Safety should be firmly set in your mind so that you will not become careless. This awareness will allow you to work with confidence.
Now you will have that added punch that will get you a 5/9 plus report, and the bigger thrill to some degree of “re-building your own”!