SideBand Engineers Model 34
SBE "SB-34"
Bilateral Amateur Radio Transceiver
(1966) |
SideBand Engineers SB-34 is a HF SSB transceiver, arrived on market in 1966.
Considerable numbers manufactured, this radio could be classified as "interesting radio" for most of the amateurs;
although quite charm and was indeed popular,
it is not well regarded in comparison to the big stars such as Collins, Hallicrafters or Drake.
However, if you become familiar with its design and construction, you will be amazed with this advanced, ambitious radio. The configuration of SB-34 is almost identical to its predecessor, SB-33, but its style is much more sophisticated. SB-34 is fully transistorized except transmitter driver and final amplifier. It has a built-in power supply which is capable to run both on AC117V and DC12V (SB-33 required external inverter in order to run from DC12V). Compact body archived by the transistorization is suitable for the mobile operation. The transceiver only handles SSB, and covers 3.8/7/14/21 MHz bands. Output transmit power is claimed to be 80W. SB-34 employs a Collins mechanical filter, and has a "bilateral" circuit configuration. Block Diagram shows how the bilateral works - signal flows in opposite direction when transmitting, than when receiving. It made possible to reduce total number of components, thus the compact, lightweight and lower priced rig. |
|
|
|
Block diagram here shows its unique bilateral configuration.
Yellow blocks are used both in reception and transmission.
There are 2 keying control buses. While receiving, the voltage of one bus is high and the other is low. When transmitting, voltage level of buses are swapped. One bus is used to disable the Tx only blocks while receiving, the other disables the Rx only blocks while transmitting. Keying is done in such manner, successfully eliminated the mechanical relay. |
|
|
I acquired this SB-34 at the Livermore swapmeet, and it was the first day on my way traveling to Canada,
a week long vacation I and my wife had been waiting for.
When we arrived at a motel in Oregon, I could not help starting the test of the new radio.
The AC power cable came with the rig, and I turned it on.
The radio skipped the most exciting moment, 15 seconds filled with great anxiety.
Yes, this is solid state... the radio immediately came back to life with warm color of the dial illumination. Since I did not have an HF antenna, I connected a power cable for my laptop computer to the antenna terminal using a paper clip. It worked as a short wire antenna. Receiver seemed to be working. The radio picked up several stations on 7MHz calling CQ contest. Returned to Silicon Valley I fired up the radio on my workbench with a random wire antenna and my antenna tuner. Although the audio volume seemed weak, the receiver picked up South America and Australia. Then, after a while, sensitivity suddenly dropped, and finally, even the background noise faded into silence. I was puzzled and tuned the dial. The receiver still worked at the lower frequency in the same band. I continued to monitor CW stations but eventually the receiver died completely. The radio was obviously in trouble. In the next evening I found the radio revived. Wondering what happened last night and I enjoyed monitoring DX stations. 30 minutes later, however, the sensitivity again suddenly dropped and finally it died again. Looks like the operation time is limited to 30 minutes. How can such thing happen? |
Tried many things, the symptom was as follows.
Now it was the time to open the case. The metal sleeve case does not have ventilation holes except the vicinity of the final power amplifier. I guessed the overall ventilation should not be enough. I removed the case, and observed the internal configuration. No modification or fixing attempt was found. The receiver kept operation much longer if operated without the case, but eventually it lost its consciousness. The receiver revived temporary, if I use a Japanese "Uchiwa", or hand fan, to apply air flow to the receiver. It confirmed the heat as the primary cause of the trouble. I placed small electric blower fan (used for PC power supply). The receiver maintained its conscious after the several hours of operation. |
I soon suspected the VFO ceased its oscillation.
Somehow the VFO stopped operation when it got hot, therefore the intermediate frequency could not be generated.
This hypothesis clearly explained the phenomenon observed.
As a matter of fact, the sensitivity immediately came back if slight airflow was applied to the surface of the VFO board. The VFO circuit generates frequency between 5.4569MHz and 5.7069MHz, regardless of the band. The circuit consists of 2 transistors; one is the oscillator and the other is the buffer amplifier. I checked the VFO output in a way suggested by the service manual. Placed another shortwave receiver beside the rig, and connected a test lead wire to the VFO output. By winding the vinyl test wire to the whip antenna of the potable shortwave radio, it could then be used as a signal strength meter. The test proved my hypothesis - VFO output rapidly dropped when the circuit got hot, and eventually no signal was generated by the oscillator. |
In a weekend afternoon I kept the garage door closed.
Californian summer afternoon sunlight turned my lab into a nice temperature chamber.
Now the problem is easily reproducible and I measured voltages at the various points of the VFO circuit.
What I found was that emitter voltage of the VFO oscillator transistor was very temperature sensitive.
When this voltage exceeded the certain threshold, it caused the oscillator halt. In order to maintain the voltage within the operational range, I added a resistor to the bias circuit. It made the oscillator stable in operation even if constantly used under the high temperature. The problem was solved. However, the true reason is still unknown. I guess the property of the transistor or resistors was changed due to aging. |
Now the reception is satisfactory stable- also stable in weak volume.
Although my antenna is mere long wire swinging in the backyard,
MFJ preamplifier-tuner should provide enough RF signal to the receiver.
The operating manual states that volume control at 50% position gives enough volume for the indoor use.
In a reality the volume control must be full in order to archive the usable audio. Insufficient gain of the audio stage was suspected, however, the speaker blasted me when a local San Jose HF station began calling CQ. The audio output stage itself seems to be functioning normally. By injecting CD player output into the audio driver, it was confirmed that driver stage also functioning as expected. The sound is very dry, communication equipment taste. It's not a receiver for listening entertainment program in any way so this is not a problem. |
The first suspect was the AGC and volume control circuit which is in very unique design.
If nothing is done, this radio will remain useless so I decided to start tweaking. The volume knob controls not only the audio gain but also the IF gain and RF mixer gain. AGC line controls RF mixer gain too. The RF amplifier and 1st frequency converter (RF mixer) is wired on a "RF board" located under the chassis. So I disconnected the AGC line and volume control line from the board, and replaced them with fixed voltage so that the RF board would always work with full gain. With this configuration, strong signal became heavily overloaded, seemed like the board was actually working with full gain. However the audio from speaker was still not loud. |
I mounted an extra rug terminal board to the RF board to provide an easy hookup for the multimeter.
AGC line from IF board was then connected here.
The time constant capacitor of the AGC circuit might loose its capacity,
so I added the new capacitor on this rug terminal board and removed the original one. During the modification, ZAP!!!!! I saw a spark from the tip of the soldering iron, and something flew away and hit something at the corner of the garage shop. I soon realized that one of the transistors was exposing its gem - plastic mold was blown up. The receiver was completely dead when powered up. The power cable had been disconnected at the moment of the incident, but the chassis had been connected to my old oscilloscope. It seemed like my soldering iron leaked, electric current flowed from the iron to the scope, damaged a RF board. Oh my... Damaged component was a 2N2495, a transistor for the Rx RF mixer. Although the type is different, the receiver worked weakly, when I placed the Tx RF mixer transistor on the Rx RF mixer socket. I was in need of a 2N2495 or its compatible transistor. No exact part found at Halted Specialties. I purchased several similar PNP Germanium transistors, but none of them gave the life to the receiver as before. Transistor substitution manual suggested there is no direct substitution. Searching web found nothing. NTE would be the best destination in such a case as this. It sells many replacement semiconductors with NTE original number. The NTE replacement for 2N2495 was NTE160. Halted Specialties carried a stock so I purchased one. Cost $4.00, pricey but worth to pay. Although NTE160 was a replacement, its packaging was different and could not be inserted into the socket on board, so I had to solder it directly. The receiver came back to life. But pay attention from now. Better to disconnect all the wiring for the measurement, in addition to the power connection. Of course the best solution would be to buy a good soldering iron, though... |
Now, let's return to the troubleshooting, the weak volume.
I measured voltages at various points and monitored waveforms, but no true find.
Could it be in the RF stage or the IF, or may be a bad balanced mixer.
Could it possible the Collins mechanical filter has something wrong, or is it the true performance of SBE? Spending several nights, the mystery was still unsolved. No more idea came up in my mind. My exhausted biochemical computer then discovered an ultimate and brilliant solution... Compare to another working rig! But, hey, there's nobody having SB-34 nearby.... I surfed the web seeking the answer and what I found was... a working SB-34 for sale! Without getting my wife's approval, my finger clicked a button of the mouse. Two weeks later, two SB-34 sit side by side on the workbench of my lab. |
The SB-34 #2, which cost me more than #1, had a genuine SBE hand mike and service manual,
and was cosmetically original.
It played beautifully with plenty of volume. By opening the cover, modification was immediately apparent. Most obvious thing was the crystal calibrator board which seemed to be a general purpose kit, not a SBE genuine product. The genuine option calibrator unit was designed to be mounted on the back panel of the SB-34. This unidentified board had an identical connection to the genuine option, so shouldn't be a problem. All of the electrolytic capacitors in the receiver circuit had been replaced. Desoldering and resoldering work was observed throughout. Especially at the AGC circuit, virtually every point was reworked. The previous owner should have already paid efforts I was considering. And most important thing is that this unit is working perfectly. No more better example could be expected. I had discovered many differences of component values used, between circuit diagram and the #1 radio. The manual which came along with #2 radio was newer than the first one. The component values of #1 radio matched with #2 manual; i.e. #1 and #2 radio are the same revision, and the #1 manual was older. |
Side by Side, was indeed a powerful method.
The answer was lying in front of me.
All I had to do was to check and compare, and all I need was the effort and patience. As a brute force method, I tried to swap the circuit board. Not a physically swapping, but disconnect the wiring between the boards and interconnect the #1 and #2 radio. As a result, the problem was tracked down to the IF board. When I connected the #2 VFO board output to the #1 IF board, reception was weak. On the other hand, #1 VFO and #2 IF board worked well. Swapping the mechanical filters did not show any difference, therefore they were okay. This was a good news; if the mechanical filter was broken, the cost of the replacement would be, if ever obtained, more than a half of the original purchase price of the radio. As far as the price was concerned, a SB-34 could be a mechanical filter with all surrounding circuit attached. Comparing the ring modulator, which employed four diodes, did not show a big difference. Audio drivers and power amplifiers seemed to be working identically. From these observations, I became pretty sure that the 1st audio amplifier right after the ring modulator had the trouble. |
The 1st audio amplifier, is just a single transistor.
Almost there! Nothing should be tough anymore. Nevertheless, I could not determine the true cause.
Although it was basically a simple amplifier,
the circuit was connected to the microphone amplifier and keying control which switched Rx and Tx.
I disconnected those circuit in order to separate the problem, couldn't find the cause.
Tried another transistor, replacing capacitors,
and even built a new amplifier circuit and replaced with it.... still no luck. Honestly speaking I haven't gotten a formal education of analog electronics in a school or an university. Strongly realized the necessity of basic knowledge, started to read electronics textbooks including ARRL handbook, refereed my dictionary for unfamiliar English words, back to the radio and tried something different, again returned to the textbooks.... Tired with a deep frustration of not being able to fix a mere single transistor amplifier, I picked up some capacitors removed from the radio. My analog circuit tester reacted when quality-looking black plastic mold electrolytic capacitors were connected, as if the capacitors were laughing at me. However, I noticed that one of those black small caps did not catch the instrument's attention. The marking read 10uF, the tester should respond for its charging current. What was this capacitor all about? Quickly checked the radio, and I found that I had removed from the radio but not replaced with a new one. Placed a new old stock capacitor, the amplifier started to amplify! It was an audio frequency bypass in the emitter circuit. Two capacitors were connected in parallel, another one was a 0.1uF. If this 10uF was missed, audio bypass was far less than ideal, thus the low gain of the amplifier. What a simple cause! Ashamed with my poor troubleshooting skill, but I felt quite happy with very loud voice someone shouting the "CQ contest!". |
Restored the tentative test wiring to the original, readjusted the RF oscillator.
The fixing was completed, at least for the obvious receiver problem.
The #1 is much much powerful than before, but a little weaker than the #2.
Replaced RF mixer transistor might be the reason, while the entire re-alignment would be necessary. AGC circuit should be the next task. Frequently adjusting the volume control was necessary, which was annoying especially monitoring the net operation. The #2 showed almost the same behavior, so I guess this was the original characteristic of the SB-34. So, in order to improve, the AGC circuit arrangement should be modified. But how could it be made? Perhaps its unique volume control concept will be abandoned and regular AGC method will be used. If such modification can be done, adding a S meter functionality would also be possible. Since I did not have a proper antenna capable for transmission, Tx part was untested for both units. I pressed the PTT button for less than a second and I could hear a squeak from other unit to which the antenna was not connected. Therefore it looked transmitting somehow, but not knowing how much power it could produce. Some SB-34 owners (or restorers) reported on the Internet that they suffered weak transmitter, seems it is a common problem to the 34. Reportedly the frequency stability is not sufficient while transmitting, due to the marginal ventilation. These points must be carefully checked someday. Although not active right now, SB-34 back-on-the-air project is still in progress; I purchased a new dummy load and a SWR/power meter solely for this project! |
|