The SW-717 is Heathkit's solid state successor to a series of receiver kits that started with the AR-2 followed by the AR-3 and then the GR-91 and the GR-64. All of these were single conversion tube-based superhet shortwave sets covering 550 KHz to 30 MHz in four bands. These were considered entry level shortwave sets, often built by teenagers as a first major kit building project and commonly used as a first receiver by amateur radio novices. All have a BFO and bandspread. The AR-1 was of course the great grand daddy, but it only covered three bands and had no BFO or bandspread. (see catalog ad for the AR-1 on bottom of page at this link)
The color scheme and outward design of the set show the family resemblance to the GR-64 and GR-91. Like all of the sets, the SW-717 is single conversion but replaces the usual 4 or 5 tubes plus rectifier of its predecessors with 9 discrete transistors and 11 diodes. It also has 3 ceramic filters in place of the usual 455 KHz IF transformers. Transistors are MPF105 oscillator, 40673 RF amplifier/ mixer, 2N5308 Darlington IF amp, two 2N5323A IF amp and AGC amp, 2N3393 audio preamp, X29A829 audio driver, and MPSU05 and MPSU55 for audio output. (S2090 and S2991 for audio output in some versions. Other transistors for audio and possibly for IF may also have been used during the ten year run of the set.)
The radio has a ferrite rod antenna attached to the cardboard back for reception of the broadcast band and the usual terminals for an external antenna for the three shortwave bands.
Repairs
This set arrived with a power cord with bare wires showing a short circuit at the point of entry at the chassis clamp. One knob was missing and another was split. It also had an antenna modification that connected the external antenna terminal directly to the AM broadcast rod. The bandspread variable cap did not move the full distance. I removed the power cord from the chassis, cut off the worn piece and then recrimped and resoldered the cord. On power-up the set worked. I lubed the bandspread cap bearing which then was able to fully close and open. After studying the antenna terminal modification, I decided to undo the mod. I tightened all the ground screws and inspected all the solder joints, reheating some of the obvious cold-solder joints. Deoxit was applied to the controls and bandswitch. I located spare knobs making sure not to overtighten the set screws in the plastic knobs.
Alignment of the receiver
The accuracy of tuning was poor. The set needed an alignment. I had the schematic but not a manual. That was not a problem since alignment is relatively straight-forward. With no IF transformers, only the antenna and oscillator adjustments need to be aligned for each band. I set the bandspread cap in the middle of its travel with the indicator at zero. As in all similar alignments, the capacitor trimmers are for the high end of each band and the coil adjustments are for the low end. The ferrite rod for the broadcast band is of course not adjustable. The trimmer for the antenna side of the broadcast band is on the main tuning cap (section closest to front panel). I have added some detail here since the manuals for the SW-717 are not readily available. If you have a manual, please check my alignment information for accuracy. I set all alignment points at the exact low and high dial indications for each band. In some instances, factory alignment information for the antenna side specifies a 20% and 80% position on the dial for a compromise.
Oscillator Alignment adjustments
The oscillator adjustment points are shown in the picture. The coil adjustments for the low end of each band are L-1 for broadcast band A (shown with the diddle stick in the coil) and L2, L3, and L4 which are for Bands B, C, and D. In the picture, just above those four coils are four trimmer capacitors for the high end of each band. Those caps correspond to the coils. They are left to right for the high end of band A, B, C, and D.
When aligning each band for tuning dial accuracy, there is some interaction between the coil setting at the low end and the trimmer cap at the high end so repeat the adjustments several times for each band until there is no further improvement. With a bit of care, the dial setting for the SW-717 can be aligned quite accurately despite what some on the net have said. Be sure to keep the bandspread cap and control at the center zero setting during all alignment.
Antenna Alignment adjustments
Like the oscillator adjustments, the antenna peaking adjustments for the low end of each band are the coils near the front panel. There is no low end adjustment for band A since the coil is the ferrite rod itself. Band B is L-5, Band C is L-6 and Band D is L-7. Each is adjusted to maximize the sensitivity at the low end of the band. There are three trimmer caps for sensitivity at the high end of the bands. Band A (broadcast) uses the trimmer on the main tuning cap itself. It must be adjusted first since it affects the others.
There are two adjustment trimmer caps below the chassis but the adjustment screw is reached from the top through a hole in the chassis. Those two screws are visible on the extreme right side of the chassis picture above. The screw closest to the front panel is for Band B. The one behind it for Band C. There is no antenna trimmer adjustment for the high end of Band D. Again, there is a bit of interaction between each high end trimmer and low end coil adjustment, but a bit of back and forth adjustment will maximize sensitivity on the band.
Tackling the hum
The SW-717 has a reputation for hum at low volume levels. That was also the case with this example. While not objectionable, a low level persistent hum was there that was not changed by the volume control setting. Using headphones accentuates such a hum level because of the greater sensitivity of headphones.
I assumed the hum was due to insufficient filtering of the power supply and decided to test that theory by powering the set with pure DC from two 12 volt gel cells in series. That totally eliminated all hum. I checked ripple voltage at the input electrolytic with a scope and found about 0.2 volt of ripple. Thinking the electrolytic might be the problem, I bridged it with another large electrolytic. No change in hum level. Therefore the input electrolytic was in good condition. I assumed that power supply ripple could be reduced greatly with an added filter stage. After some experimentation, I decided on a 50 ohm resistor between the diode bridge output and the existing first electrolytic. I then added a 470 MFD 200 volt electrolytic recycled from a computer power supply at the output of the diode bridge. As confirmed by the scope, adding that extra filter stage reduced the ripple voltage by a factor of more than 10, eliminating the hum.
If you are using sensitive low-impedance headphones with this set or the typical tube receiver, I suggest adding an attenuator such as typically found in modern hi-fi receivers. A simple attenuator can be made with two resistors, one in parallel with the headphone feed such as 8 to 10 ohms and another in series with the headphones such as 100 ohms. For high impedance headphones, a third resistor in parallel such as another 10 ohms may be needed. Those resistors provide a roughly 10 to 1 attenuation. The volume control must be increased a bit but the underlying hum is attenuated. Experiment with the resistors for the best values.
Performance and observations
The SW-717 can hold its own as compared to others in the entry-level four band Heathkit series. Its broadcast performance is very good. Complaints about poor dial accuracy and a bit of hum can be readily overcome. I suggest tightening all grounding screws and checking all solder joints since these sets were kits.
Complaints about wimpy knobs that break if set screws are over-tightened are justified, but that is true for the previous two generations as well. Complaints about the SW-717's wider passband compared with similar sets with IF transformers are also justified although some have said that the wider passband makes for better sound on AM.
I was surprised by the stick-on speaker mounting pads (no screws to hold the speaker in place). While it may have avoided possible cabinet vibration problems, it looks cheap.
The tubes of its forbears are readily available but a couple of the transistors in the SW-717 are getting quite scarce especially the dual gate mosfet.
Lastly, the $90 price tag shown in the 1979 catalog pictured above seems high for the era and likely hurt sales.
In comparison with other sets in the series, the ferrite rod antenna allows for ease of use if the set is moved around a bit (true also of the GR-64 and GR-91). The solid state design means no drifting due to heating issues. The SW-717 makes a nice shop radio. Mistakenly leaving it on all day is not a problem.
Suggestions from the net to improve the set include taping or glueing a ferrite rod extension onto the existing rod to increase its length. I did not try that, but I did wrap a loop or two of hookup wire around the bare portion of the rod and connect that to the coax from an outside antenna. That obviously improved AM broadcast sensitivity greatly without a direct connection or overloading.
Let me know your experiences with Heathkit's last SWL four bander.
This link from Hans Gatu shows his Heathkit SW-717 along with the schematic.
Try this link from Bytesmiths' infoark for schematics and a couple of pictorials.
date 1-10-12
A Meissner AM-FM tuner model 9-1091C was the previous item on the bench.