| |
| |
 | ||
| | |
 | ||
| | |
Without question, the HW-8 is by far the most popular QRP transceiver every produced in kit form. Heath sold 'em by the boxcar! Even today, the HW-8 has a cult following. The HW-8 is a CW only QRP transceiver that covers the lower CW segments of 80, 40, 20 and 15 meters. The receiver is direct conversion meaning there is no IF stages or amplifiers. The transmitter will produce up to 2 watts of RF into a 50-ohm load. The HW-8 is a surprising nice radio to work QRP with and even though it only produced two watts, you can easily work the world with a HW-8 transceiver. The HW-8 is built on one major PC board with a small audio amplifier PC board mounted on the side panel. Some things to look for first! Because the HW-8 is a direct conversion receiver, the VFO plays an important part of the operation of the rig. No VFO, nothing works! The VFO operates from 8.645 MHz to 8.895 MHz. The active devices in the VFO are Q2 and Q3. The output of the VFO is applied to the balanced mixer along with the HFO oscillators The HFO oscillator operates at 12.395 MHz for 80 meters, 15.895 MHz for 40 meters, 22.895 MHz for 20 meters and 29.895 MHz for 15 meters. Transistors Q4 and Q5 form a mixer amplifier with its output going the transmitter driver, Q8 and to the receiver product detector IC1. So, as you can see, if the VFO, HFO and mixer are not working, the HW-8 will be quite dead. The check for proper operation of the HFO, couple a frequency counter to TP1. This is R94 located near the row of diodes in the center of the board. As you monitor this test point, you should see the HFO frequencies: 12.895 MHz=80 meters If one or more frequencies are not present, remove the counter from TP1 and connect your VTVM. If you have a scope, use it in place of the VTVM. The scope will allow you to see what the waveform looks like, too. Depending on what band is not working try adjusting the HFO coil for the band. 80 meter is the bottom of coil L17 When the HFO is operating, you'll see a voltage appear at TP1. Adjust the coils for maximum voltage. Check for proper operation by selecting and deselecting the bands making sure the HFO oscillator starts all the time. If you find one band that is slow, adjust the proper coil until the oscillator starts reliably. The HW-8 uses diode switching for changing bands. Along with the push button switches to route the high power (two watts!) RF to the output filters. If you have a HW-8 that is dead on one or more bands, start checking the switching diodes. For example, 80 meters requires D22 to be on. A Case History I picked up a broken HW-8 at a hamfest. The seller told me it only worked on 20 meters. There was no transmit at all on any of the other bands. The receiver was also quite on 80, 40, and 15 meters. All I could hear was the audio hiss in the earphones. As I looked around at the PC board, I found that out of twelve of the compression trimmers, only three had their adjustment screws. The rest were missing. Now, why on earth would anyone want to remove the adjusting screws from the trimmers? I've still not been able to figure that one out. Also missing, or rather destroyed, is the slug in the L19/L18 coil. This is one of the heterodyne oscillator coils. If this coil is out of adjustment, the circuit it controls will not function. In this case, L19 controls the operation of the 20-meter oscillator. The other half of this coil controls the 40-meter oscillator. Since the slug for the 40-meter band was in crumbs, it told me that someone someplace was working on the heterodyne oscillator. It also told me that they could not find the problem by adjusting the slug in L18/L19. Checking the heterodyne oscillator Heath wants you to use a RF probe to check for the proper operation of the heterodyne oscillator. The test point is TP1 and is located on the diode side (as looking down onto the PC board) of R94. I have found in the past it is best to couple an oscilloscope to this test point and forgo the RF probe. The scope will not only show you the output level of the oscillator but also what the waveform looks like. Sometime more is not better! So, couple your scope probe to R94. As you select the different bands via the front push buttons, you should see a nice waveform on the scope. There should be at least 50-mV pp on the scope. On this radio, none of the crystals would fire except on 20 meters. Since Q7 and Q6 are the only active devices and they both work on 20 meters, there had to be another problem. The HW-8 is full of switching diodes. They control the various tuned circuits. We've talked about these diodes in the past, and I won't go into great detail here again. Enough to say that if one or more of the switching diodes is kaput, and then the section that is controlled by those diodes will not function. On this radio, the switching diodes controlling the heterodyne oscillator were all working. This test is simple. Use your VOM and check to see if plus 12 volts is routed via the diode to the crystal. The front panel push buttons do much more than direct the 12-volt switching voltage to the various diodes. They do in fact route the antenna to the front-end amplifier. They also route the output of the PA to the various output filters. These push button have wires on the top and connections via the PC board on the bottom. I mentioned this because I found while checking the switching diodes I had 12 volts on the green wire leading from the 20-meter switch even though the 80-meter button was in! The color codes are from the Heathkit manual and allow you to trace the various switched circuits. The color code is: Black for 80 meters While we're at it, looking into the front of the radio, the left most push button is: 80 meters SW1 Each SECTION of any of the push button PINS are numbers like this. The sections are lettered. Again, front panel, RIGHT most switch (the 15-meter button.) The section closest to the front panel is section A. Middle section is B, and the last section is C. This sequence is repeated on all the switched. So, 80 meters can have section D, E and F with pins 1,2,3 and 10, 11, 12 Now having said all of the above, a quick touch of the soldering iron detached the green wire from the 20 meter switch. With this wire no longer connected to the heterodyne oscillator, the other crystals fired right up. No matter what switch was pressed; the wire going to the 20-meter band was always hot with 12 volts. The problem was finally traced to a shorted switching diode in the front end. As it turned out, diode D7 was at fault. Here's what happened. When any other band switch was press in, the switching voltage was applied to all stages. In this case, we'll select the 80-meter band. With the 80-meter button in, 12 volts is applied to select L5 via D5. , C2 and L1 in the front-end circuit are now active thanks to diode D1. At the same time, D22 is turned on and places Y1, 12.395 MHz, in the heterodyne oscillator. At last by not least, diodes D31 and D35 are forward biased and select the L27/C77-C78 combination for the final transistor. What happened is kinda simple once you look at it. With D7 shorted, no matter what band you selected you always had several other stages turned on as well. That's way the heterodyne oscillator did not work. When 80 meters was selected, the 20-meter crystal was still selected as well. That's why the heterodyne oscillator did not run. Two crystals were in parallel! The fix is simple. Replace D7. After the new diode was in place, the 12 volt switching voltage was only present when 20 meters was selected by the push button. The only other fixes were to the L18/L19 coil and the trimmer capacitors. The trimmers were kinda easy to fix. The screw size is 5-56 by 1/4 inch long. Radio Shack sells a pack for about two bucks. But, the head of these screws are too small to hold the top half of the trimmer together. The fix? I used nylon shoulder washers. These are the same kind you would find in a TO-220 mounting kit. The shoulder part of the nylon goes the trimmer first. Then the screw is dropped down in the trimmer. The trimmer has the same thread size, so all you need to do is tighten the screw! Don't over do it as you can easily run the screw head through the nylon washer. The L18/L19 slug was a bit trickier. Before recycling became fashionable, Heathkit was already into it. Many of the same components are used in a variety of Heath products. Case in point is the heterodyne crystals used in the HW-8. They are the same as in the Heathkit HW-101! So, to replace the broken slug in the L18/L19 coil, a slug was removed from the heterodyne oscillator coil from a dead HW-101. Worked like a champ! Transmitter fixes I've come upon some information about the final output transistor that you may find interesting. Heathkit used a house numbered part. But from the printing on the device, it looks like Motorola made the transistor. I can't find a cross-reference for the Heath number anyplace. However, the instruction manual for the HW-8 references the part to a 2N4427. Mouser electronics stocks this transistor. It's about $3 a pop. In my "HW-8 Handbook" a high power modification for the HW-8 is as simple as subbing in an ECG 488 for the final. I can't for the life of me find anyone that stocks the ECG line. But, NTE replacement parts cross the ECG488 to an NTE488. I have not yet tried this in my HW-8s. I did notice that NTE crosses the 2N4427 to a 15 watt RF device in a TO-5 case. The NTE488 comes in at a 5 watt RF device in a TO-5 case. I don't know why the difference between the two. I guess I will have to experiment with some in one of my radios.
I have had luck with the 2N3553 and the 2SC799. Both will produce a good watt and a half on 80 and 40 meters. On 20 and 15, the max power seems to be about one watt. In some of the HW-8s I have come across, I have put in MSPA20 in Q8 and Q5 locations. They seem to have bit more bite that the original Heathkit devices. Although I have not tried them, a METAL case 2N2222A might work also as Q5 and Q8. There are two other locations you should look at if you are having trouble with the HW-8's output. Check the Zener diode at location ZD2. If this diode has become leaky, it maybe the cause of low output. This diode is there to protect the final, Q9, from excessive high collector RF voltages. This happens if you key the transmitter into an open antenna. Any RF voltage over 36 volts will be clamped to ground via the Zener diode. If your HW-8 has been repaired in the past, especially with final PA troubles, look to see if a ferrite bead is on the base lead of Q9. This bead is there to keep Q9 stable. If your transmitter seems to take on a life of its own, check for this bead on Q9's base. The manual calls for 3.5 watts of RF into a 50-ohm load, I’ve found these values to be much closer to real life. Transmit power @ 13.8 volts into 50-ohm load 80 meters 1.5 watts Of course I've seen more power in some HW-8s on 80 meters than 1.5 watts, but the values shown are very typical. On receive: 80 meters .5uV On 20 and 15 meters, you'll often find the trimmer capacitors in the tank circuit require to be tighten all the way down. There's not enough capacitance from the trimmer to get the stage to resonate. Take a 20-pF capacitor and tack solder it across the trimmer that is giving you trouble. Then try the adjustment one more time. With the extra capacitance across the trimmer, you should now find the circuit peaks without twisting the screw out of the trimmer threads. By the way, this fix works for the front-end trimmers as well. Setting the VFO There's one scheme that Heathkit used in almost all of their analog VFO designs; you had to balance the ends of the VFO to calibrate it. Here's how Heath wanted you do to it. You used a receiver to listen to the output of the VFO. You then turned the receiver to the low end of the VFO. You adjusted the VFO trimmer so you can hear its signal. Then you reset the receiver to the high end of the VFO's output and adjusted the VFO's coil so you can hear the signal. You did this over and over again until the two ends were balanced. What you ended up with is a VFO that tracked from the high end to the low end. Now if you have ever tried this using the method described above, you'll go batty! Here's an easier and quicker method. You'll need a good frequency counter. Allow both the HW-8 and your counter to warm up for at least 30 minutes. Couple your counter to test point two. This is resistor R49. Go to the end of the resistor that is closest to L9, the large VFO coil. The VFO in the HW-8 runs from 8.895 MHz to 8.645 MHz. The idea is to set the VFO so it will track from one end to the other between the two frequencies of 8.8895 and 8.645 MHz. You adjust the trimmer located on the VFO tuning capacitor (C302B) and the slug in L9. Again, the idea is to set the trimmer and slug so the VFO will run from 8.8895 to 8.645 MHz. It's not easy to set, so do the best you can. I find the trimmer on the VFO tuning capacitor is very touchy! Hint: Remove the screw from the trimmer on the side of the VFO capacitor and replace it with a 0-20 pF piston trimmer. This will make it much easier to adjust the VFO.
| ||
| | |