SB-220 160 METER CONVERSION NOTES

Greg Chartrand – W7MY

1/12/2014

 

Forward

The purpose of this article is to provide information for the ham who wants to add 160 meters to an SB-220. This is not a “how to convert manual” but a collection of information I used to convert and modify my SB-220’s.

 

It’s important to keep in mind that the SB-220 is a 1500 watt amplifier with cooling adequate for 1000 watts and a power supply and bandswitch for 500 watts. No modifications other than replacing major parts like the power transformer will change the calculus of having a power compromised amplifier. Please refer to the modified schematic and pictures at the end of this document as you read this article.

History

I purchased my first SB-220 in 1976 as a kit that I assembled. I needed an amplifier for 160 so I made modifications to it as it was assembled so that modification would be easier to complete at a later date. This amp was rebuilt for 6 meters many years later and a new SB-220 160-10 meter amp was recently completed (this modification).

 

Including 160 was first accommodated by adding additional switch contacts that would allow switching extra capacitors for the 160 PI network and wiring the band switch to include a 160 position while losing 10 meters. This gave me a 160 through 15 meter capability.

 

The input circuit of my first 160 modification utilized one T-200 core for the input PI coil and the output PI coil, I wound wire on two T-200’s stacked to provide the additional inductance. Extra capacitance was switched in the plate tuning and loading with the modified bandswitch.  The modification worked… sort of. The plate tuning drifted despite my best efforts to find a temperature stable padding plate tuning  capacitor. In addition, the meter lights would glow brighter when I would transmit, a clear sign that the filament choke wasn’t large enough. Ultimately I replaced the plate tuning padding capacitor with a 465 PF vacuum variable which solved the drifting problem but other problems remained.

 

I used this SB-220 as my principal amplifier for about 25 years and worked most of my 160 DX with it. In trying to use it for contesting, I blew up more parts than I can remember including HV and filament transformers, zeners, HV diodes and capacitors. I ultimately purchased an AL-1200 which replaced the old 220 which was retired to the scrap heap.

 

As my radio interests later changed to VHF, I needed an amp for 6 meters; the SB-220 looked like it could fill that gap so it came off of the scrap heap. After looking it over, I decided that I’d be better off starting with a clean chassis and rebuild the old 220 as a 6 meter only amplifier from scratch. I purchased a bare SB-220 chassis and other parts from Ebay and rebuilt it into a really great 6 meter amplifier using most of the parts from my old 220. Once finished, I still had leftover parts from my Ebay acquisitions including the original chassis, bandswitch, and spare tubes.

The Rebirth of My HF SB-220

The leftovers of the old 220 were calling me to put them back together as a complete amplifier again. Over the course of three years, I collected SB-220 and TL-922 parts and other assemblies from Ebay and other internet sources. By 2012, I had everything I needed to make another SB-220. I decided to take a different approach with this amp and make it a 160 through 10 meter and address the failings of my previous conversion so that 160 worked as well as the other bands.

 

In addition to the 160 conversion (detailed below), modifications to the basic SB-220 were identified during my 6 meter conversion from Internet articles I found on the Web and used on this amp as well.

 

Note: Several of my modifications were pre-assembled on PC boards; this made it much easier to incorporate into the tight confines of the amplifier.

 

The basic (non-160 conversion) modifications I chose to incorporate were:

 

  1. Grounding the grids[i].
  2. Replacing the zener with diodes and the high voltage diodes (installed a Harbach diode board)[ii][iii].
  3. Replacing the HV filter capacitors and bleeder resistors to more appropriate values[iv].
  4. Replacing the 110 VDC relay.
  5. Providing more air flow for the tubes and reduce the noise when receiving.
  6. Optimizing the input circuit matching network to include WARC bands.
  7. Adding a HV fuse.

 

All of the above modifications are well documented on the Internet (see the endnotes), except for:

4. Replacing the 110 VDC Relay

Most relay modifications add an additional relay or transistor to operate the 110 VDC in the amplifier (so you don’t blow up your rigs relay driver circuit). I chose to replace the relay with a pair of Magnecraft W78RESX-7 “ice cube” relays (RL-1 and RL-2). They are DPDT 10 amp 240 VAC relays and have a 12 VDC coil. I have modified the antenna switching relay by cutting the common relay contact wires and routing them outside the plastic case so to minimize the lead lengths for switching the antenna connections. I used these modified relay for my 6 meter SB-220 and decided to use it on this one as well.

 

Figure 1.0 RL-1Replacement Relay

 

The 12 volt relays need a 12 volt DC source. I salvaged a 12 VAC transformer from an old “wall wart” 12 volt power supply. I incorporated a voltage doubler circuit. This provides both 12 and 24 volts for relays (See parts related to T3 in the schematic).

5. Providing More Airflow - Quietly

I acquired a Harbach fan for the amp and installed it. It provides considerably more air (turns much faster) than the original Heathkit fan. The bad news is that it’s a bit noisy. The relay modification (above) gave me an extra set of contacts that close during transmit (on RL-2). I decided to slow the fan down during

 

receive by putting a 4 MF 200V capacitor in series with the fan. This slowed down the fan considerably along with the noise but kept a decent airflow across the tubes. The extra set of contacts shorts the capacitor so that the fan runs at full speed during transmit. To assure that the tubes get enough cooling after a long transmission, an Amico KSD301 Series KO-50 thermostat (50C/122F) temperature switch (normally open) is connected across the relay contacts so that if the RF compartment is still hot, the fan will continue to run at full speed on receive. The KO-45 resets at 35C (95F) and drops the fan to the slower speed. The thermostat is mounted with nylon and ceramic hardware.

temp sens

Figure 2.0 Location of Temperature Sensor

7.0 Optimizing the Input Circuit

I decided to do my own design of the input matching circuit using the existing coils. I designed the values for a Q of 2 and incorporated a trimmer capacitor for the input side of the PI.

 

BAND

C1 (PF)

C2 (PF)

L1 (MH)

160

1500+180 trimmer

1500+180 trimmer

5

80

700+180 trimmer

820

2.3

40

350+80 trimmer

420

1.2

20

150+80 trimmer

220

0.6

15

80+80 trimmer

140

0.4

10

47+80 trimmer

100

0.3

Table 1.0 Input PI Network Values

 

 I purchased 82 PF 150 v trimmers for 40-10 meters and 160 PF for160 and 80 meters. The trimmers are installed so that they can be adjusted from the top of the amp. I used an MFJ-259 to preset the inductance to the calculated values. The MFJ-259 is an absolute necessity for this amp project for adjustments and measuring component values. I adjust the trimmers and cores on each band using the 259 connected to the amp input coax connector and activate the relays. I temporarily solder a 56 ohm resistor from pin 5 of a tube to ground. Each band is adjusted using the trimmer and core. The WARC

bands can be covered by making the adjustments to compromise between two bands, for example, 15 and 17 meters. The 259 makes it easy to find the best compromise settings.

 

Figure 3.0 80-10 Meter Input Circuits and Trimmers

7.0 High Voltage Fuse

A high voltage fuse is useful for minimizing damage from failure conditions the amplifier might experience. Mine is a “homebrew” W7IUV fuse consisting of a single strand of wire from a lamp cord.

 

Figure 4.0 HV Fuse

160 Meter Modifications

I was unable to find a 160 modification that offered a “no compromise” solution including my first conversion 30 years ago. I wanted to achieve performance on 160 that was equal to the other bands so the design had to meet this goal. There is not much room in the amplifier for additional circuitry so the components chosen had to be relatively small. Other component types can be substituted but the difficulty will always be fitting appropriate parts in the space available. There are three additional relays added to the amp for 160 although two could have been used (described later). Some existing components were upgraded to handle the lower frequency coverage and improve performance.

Output PI Components

The addition of 160 is essentially adds components to the 80 meter band using three additional relays. Because the Jennings vacuum variable (USLS-465, 465 PF) is sufficient for 160, no additional switching is needed for the plate side of the output PI network. The extra PI inductance for 160 is provided by a pair of T-200 cores stacked together and #16 wire wound on them to provide 11 MH (L10). This inductor is in

 

 

series with the existing coil and shorted out (for 80 meters) by using a Jennings vacuum relay RF1D-26S (RL-3). The relay is physically small enough to fit into the RF compartment and was able to be positioned

 

so that the additional lead lengths were minimal. It would be possible to include the additional load capacitance with the same relay but I decided to do it with a separate relay (RL-4) so I could more easily change the capacitance value.

OUTPUT PI FIGURE.jpg

Figure 5.0 RL-3 and L10 – New PI network components

160 Meter Input Circuit

The 160 input circuit is physically remote (chassis bottom), and independent from the 80-10 meter circuits. It is switched in place of the 80-10 meter circuits with a DPDT relay KS2E-M-DC12 (RL-5). The inductor for the circuit is made from two stacked T-68-6 torrid cores wound with #24 wire to provide 5 MH. The input and output capacitors of the PI for the matching network incorporate 180 PF variables so that an exact match can be made. The associated relay and matching parts were made as a sub-assembly located to minimize any cabling lengths.

 

Figure 6.0 160 Meter Input Circuit Sub-assembly

 
Improved Components

The Heathkit plate choke marginally has enough inductance for 160. I replaced mine with a TL-922 choke purchased on Ebay. Ameritron sells the plate choke used in the AL-1200 and 1500 at a reasonable price. This would also be a good choice for a 160 conversion.

 

The Heathkit plate blocking capacitor is only 1000 PF. I replaced mine with a pair of 1000 PF doorknob capacitors in parallel. The resulting 2000 PF is a necessary improvement.

 

Figure 7.0 New Plate Blocking Capacitors and Vacuum Variable

 

One of the more difficult improvements is adding inductance to the filament choke. I chose to add a second choke in series; I think mine was from a TL-922. The two in series provide enough inductance such that the meter lights do not glow brighter when transmitting. The filament voltage is still within specification but on the low side.

Figure 8.0 Additional Filament Choke

 

 
 
160 Meter Switching

The switching between 160 and 80-10 meters is accomplished by activating the three 160 meter related relays. I used a SPDT toggle switch that had a very small mounting diameter. I made a small increase of the upper left hand corner screw hole that mounts the front panel to the chassis. The toggle switch fits perfectly in the hole and looks like it belongs there so no additional holes needed to be added to the front panel. Because the vacuum relay is 24 volts and the other 160 relays are 12 volts, each relay has an isolation diode in series with the positive voltage connection. The negative connections of the three relays tie together and are switched to ground with the toggle switch. For 160 operation, the bandswitch must be in the 80 meter position and the switch activated. Conversely for 80-10 operation, the switch must be de-activated. Having the switch in the wrong position will result in improper operation. To help negate confusion, I installed two 1W red LED’s over the meters that light when the 160 relay switch is activated. This serves as a reminder that the amp can only work on 160 meters when the meters glow red.

 

RELAYS LABELED.jpg

Figure 9.0 New and Additional Relays (chassis bottom)
Conclusion

The modifications to the basic amplifier upgrade its performance and provide additional “safety” upgrades to address previous component failures. The addition of 160 meters to the amplifier met all of my expectations and unlike my 1976 modification, does not compromise operation on any band.  The modified SB-220 works flawlessly on all bands 160 through 10 including 17 and 12 meters. The input circuits provide a good match on all bands without the need of a tuner.  The 30+ year old 3-500’s and amplifier can still put out 1400+ watts power output for casual low duty cycle operation.

SB-220 V9

TOPCHASSIS

 

TOP CHASSIS-2.jpg

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

FRONT PANEL

 

 

 

 

 

 

 

 

 

 



[i] AN SB-220 COMPENDIUM http://www.orcadxcc.org/content/sb220_compendium_va7jw_rev1.pdf

[ii] SB-220 Mods - KE5YA https://sites.google.com/a/hamtx.com/dx-club/sb-220-mods---ke5ya

[iii] Harbach Electronics http://harbachelectronics.com/product-category/heathkit-sb220-sb221/

[iv] SB-220 Capacitor Bank http://forums.qrz.com/showthread.php?323597-SB-220-Capacitor-Bank