The Heathkit HW-101
by Greg Latta, AA8V

Restoration

Click on the image for a super detailed view.

Important Safety Note: Working on or testing equipment such as the Heathkit HW-101 transceiver and the HP-23 power supply is extremely dangerous since very high voltages are present when the equipment is turned on, and may even be present when the equipment is turned off and unplugged. If at all possible, do all work with the equipment off and unpluggedand be sure that the capacitors are properly discharged before working on the equipment. The operator assumes all risk and liability in such matters! Do not work on this type of equipment unless you are experienced with working around very high voltages!

Restoration:

Scroll Down or Select A Link Below:

Introduction	Driver Preselector Belt Replacement
Tube Testing	Shunting of Resistor R202
Initial Turn-On	Chassis Cleaning
Electrolytic Capacitor Replacement	Circuit Board Cleaning
HW-101 Electrolytic Capacitor List	High Voltage Cage Cleaning
Some General Comments On Replacing Electrolytic Capacitors	Cabinet Cleaning And Waxing
Crystal Calibrator Repair	Front Panel And Knob Cleaning
Carrier Oscillator Repair	Replacement of the Crystal for the 29.5 MHz Band:
Preselector Tuning Drum Repair	Coil Cover Modification

Introduction:
After the HP-23 power supply was restored I began work on the HW-101 transceiver. The tubes were removed, noting which particular tube came from which particular socket, since the HW-101 has multiple 6EA8, 6AU6, 6HS6, and 12AT7 tubes. This guarantees that each tube goes back into the same socket it came from. Alignment sometimes depends on a particular tube, and making sure each tube goes back into the same socket guarantees that the alignment is not affected. The transceiver was then thoroughly checked for any stray material on the circuits boards. A vacuum cleaner and clean camel hair chip brush were used to draw out any dirt and foreign material. Gentle compressed air was also used to blow out any foreign materials that might still remain. Everything was carefully inspected for bad solder joints or broken connections. None were found.

All of the controls were rather stiff and were carefully cleaned with Caig DeoxIT. This loosened up the RF Gain, AF Gain, and MIC/CW controls. The Function and Meter switches were quite tight. These were sprayed with DeoxIT and then carefully worked free. After treatment they performed as they normally should. Since the bandswitch was covered with an aluminum panel, it was left for later cleaning.

Tube Testing: All of the tubes were tested on my Hickok 800A tube tester. This tester actually loads down the tube under test and measures its amplifying ability. All of the tubes tested fine except for the 6BN8, which was a little on the weak side but will probably work fine. The RCA 6146 finals tested fine, but the only real test for a final tube is to place it in actual service. In this picture you can also see the top cover on the right. It is in good shape. It just needs a really thorough cleaning.

Click on the image for a super detailed view.

Initial Turn-On:
In April of 2022, after a thorough inspection and initial cleaning, the HW-101 was hooked up to the restored HP-23 power supply, and a speaker and crude antenna were connected. The power supply was plugged into a VARIAC and then the voltage was brought up very slowly while carefully observing the transceiver for any overheating, smells, or other problems. None were observed. The voltage was slowly brought up to 100%, and all was fine. Noise was heard from the speaker, and the receiver appeared to be operating. Several strong signals could be tuned in on 40m. After many, many years the HW-101 was coming alive!

Electrolytic Capacitor Replacement:
There are surprisingly few electrolytic capacitors in the HW-101. They are the following:

C2 - 10uf/15V cathode bias capacitor in the speech amplifier
C12 - 20uf/350V decoupling capacitor in the speech amplifier
C212 - 10uf/15V - cathode bias capacitor in the VOX amplifier
C304 - 20uf/350V decoupling capacitor in the audio power amplifier
C906 - 10uf/150V bias filter capacitor

These were all checked out with a capacitor tester and measured OK. However, for a piece of equipment that is 50 years old it is standard procedure to replace all electrolytic capacitors. They may initially be OK, but when placed back in service the heat can cause them to dry out and fail. I ordered replacement parts from Mouser Electronics at the same time I ordered new parts for the HP-23 power supply. A list of the capacitors I ordered and their part numbers is shown below.

Some General Comments On Replacing Electrolytic Capacitors:
Electrolytic capacitors contain an electrolyte that can dry out over the years. They can still test and function fine if they contain only a fraction of the electrolyte. When a vintage radio is first brought back on line the electrolytic capacitors may function fine at first. However, after some use the heat from the equipment may dry out what is left of the electrolyte and then they fail.

In the long run, it is best to eventually replace all of the electrolytic capacitors (called "recapping") in a piece of vintage equipment. The needed capacitors can all be purchased and installed at the same time. This saves shipping costs and avoids having to open up the equipment multiple times.

To obtain a high enough voltage rating power supply capacitors may have to be replaced with multiple identical units in series with appropriate equalizing resistors across each. For power supply capacitors a higher capacitance and/or voltage rating is OK.

For electrolytic capacitors in speech amplifiers and audio circuits the capacitance should not be changed, since this could alter the frequency response. However, a higher voltage rating is OK.

"Can" capacitors and other units containing several capacitors are very hard to find (or are very expensive) and can be replaced with individual units. However, for can capacitors, disconnect the bad unit but leave it in place for aesthetic purposes.

Crystal Calibrator Repair:
After the initial turn-on it was found that the crystal calibrator did not work properly. Turning on the calibrator caused the receiver noise to rise to a very high level and the transceiver would go into transmit mode unless the ANTI-VOX control was turned all the way down. One lead of diode CR201 was unsoldered to disconnect the crystal calibrator output from the receiver input. This did not solve the problem. All of the components in the calibrator circuit, except for the crystal, were tested and found to be fine. Finally, a crystal from another HW-101 that was used for parts was installed. This cured the problem.

The crystal was bad, generating so much noise in the circuit that the noise travelled down the B+ line to the heterodyne oscillator circuit, which was very close by, where it got into the receiver. The strong noise overloaded the ANTI-VOX circuit, causing it to do the exact opposite of what it should do, putting the transceiver in transmit mode!

Carrier Oscillator Repair:
While using the receiver, it was found that the receiver had much lower gain/sensitivity in the USB/CW mode than in the LSB mode. I suspected that there was something wrong in the carrier oscillator circuit. Testing the output of the carrier oscillator with an RF voltmeter probe connected to pin 3 of V16 showed about 3V of RF in LSB but only 0.5V in USB/CW. Also, the plate voltage on V16A was found to be only 35V in USB/CW. Resistance checks of the V16A circuit showed that plate resistor R6 had increased in value from 33k to 330k. It was also found that the other plate resistor, R7, had increased in value from 33k to 42k. Both resistors were replaced, and the receiver really came alive. Problem solved!

Preselector Tuning Drum Repair:
One of the weaknesses of the HW-101 is the method used to attach the metal pulleys to the preselector shaft. These were to be soldered to the preselector shaft, but most builders did not have a large enough soldering iron to do the job properly. As a result, the drums often come loose.

As shown in the sequence below, a better solution is to use epoxy glue to hold the drums in place.

Preselector Tuning Drum Repair

Click on the image for a super detailed view.	Click on the image for a super detailed view.	Click on the image for a super detailed view.
Unrepaired Tuning Drum: In this picture the original soldering job can be seen. The drum is loose on the shaft, but still turns with the shaft due to the solder built up on the shaft.	Preparation Before Gluing: A Dremel tool is used to grind off any solder and to free the drum, which is then slid down the shaft to expose the mounting area. The mounting area on the shaft is then cleaned with the Dremel tool so that the glue will properly adhere. (Be sure to cover the circuit boards around the drum with a paper towel to prevent any contamination during grinding.)	Finished Repair: The shaft is coated with epoxy glue and the drum slid into position. A bead of glue is then applied to the drum as shown for extra security.

Driver Preselector Belt Replacement:
The Driver Preselector control shaft is connected to variable capacitors C421 and C422 via a pair of rubber drive belts. These belts stretch and wear with time and are broken or stretched in many HW-101s. Mine were still unbroken, but had stretched to the point that they no longer functioned. A variety of replacements were found on e-bay. (Note that the SB-100, SB101, SB-102, HW-100, and HW-101 all use the same belts.) A seller in South Korea offered red silicon drive belts for a higher price, and I decided to go with those, since silicon is more durable than rubber. The new belts were installed and work just fine, as can be seen in the photo below.

Before installing the belts I carefully measured their sizes. Since they came from South Korea, I assumed they would have metric dimensions. I carefully measured them and found them to have an inner diameter of 50mm and an outer diameter of 56mm.

Driver Preselector Belt Replacement

Click on the image for a super detailed view.

The rubber Driver Preselector belts were replaced with silicon belts obtained on ebay. The new belts have an inner diameter of 50mm and an outer diameter of 56mm. They are 3mm thick.

Shunting of Resistor R202:
R202 was a 10k ohm 1/2W resistor that originally connected the output of the first IF amplifier (V3) on the IF amplifier board to the grid of the first transmit mixer (V5A) on the bandpass filter board. It was only present in early run HW-101s, and was removed by Heathkit on later runs. My copy of the 1970 manual and the schematic diagram do not show the resistor, though it was present in my HW-101, indicating that my unit was a very early HW-101.

Rather than remove the resistor and replace it with a wire, I shunted the resistor with a piece of wire and left it in place. This effectively removes the resistor from the circuit but also alerts a possible future owner that the resistor was originally present, and that my unit was an early run unit. See the picture below:

R202 Shunt

Click on the image for a super detailed view.

Chassis Cleaning:
My standard procedure when restoring gear is to first clean as much of the chassis as possible with a very small rag or piece of paper towel and Fantastic cleaner. Never spray the cleaner directly onto the work! Instead, very lightly wet the paper towel/rag with the cleaner and wipe the surface. You must be careful not to let excess liquid wick itself into undesirable places! When finished, replace the Fantastic cleaner with water and rinse the surface. Again, never spray the work directly.

In some special cases I will use Ronsonol lighter fluid (naptha). (Over the years I have found that lighter fluid is relatively benign and does a great job of cutting through certain grease and other deep grime.) If you are in doubt as to whether to use the lighter fluid on a surface, test a small inconspicuous area first.

After as much as possible is done with the rag/paper towel, I then switch to using Q-tips to get into the tight places. Don't use any brand other than Q-tips! (On cheaper brands the cotton isn't fastened well and often just falls off.) I mount my Q-tips in a special knurled brass handle that I made to hold them, as shown in the photo below:

Brass Holder For Q-tips.
Click on the image for a larger view.

The handle is made of 7/32" brass 6" long (aluminum would be fine too). A 1/2" deep hole is drilled in one end with a #39 drill bit (0.099") that provides a snug fit for the Q-tip. The Q-tips are cut off with a pair of side cutting pliers and then inserted into the end of the handle, as shown in the photo. The handle allows me to get into all of the tight places on the chassis that cannot otherwise be reached. You can control the flex of the Q-tip by how much of the shaft you leave on the Q-tip. I usually cut my Q-tips in half, and do a whole bunch at once to save time.

To use the Q-tip, touch it to the outlet of the cleaner bottle and barely pull the handle on the cleaner bottle. If you get it too wet, touch it to a small rag or piece of paper towel to remove the excess. When the Q-tip gets dirty, take it out and insert another. You will go through a lot of Q-tips, and it is much like cleaning the floor of your kitchen with a toothbrush, but you can get to all of those tight places and get them all clean.

I even clean the components (variable capacitors, especially the steatite/ceramic insulation, fixed capacitors, resistors, and coils (after carefully testing on a small part of the coil)) this way as well. Yep, the process is time consuming, but it is about the only way to really get the chassis clean.

After everything is cleaned, the entire process is repeated using Pledge furniture polish to resist dust build-up and give everything a nice shine. As before, don't directly spray the work. When using the Q-tips with the wax, touch the Q-tip with the outlet of the Pledge bottle and barely press the spray button. Some wax will wet the Q-tip and usually some excess will remain on the can. You can mop the excess up wax up with other Q-tips. If you get the Q-tip too wet, touch it to a small rag or piece of paper towel to remove the excess.

Circuit Board Cleaning:
The circuit boards are cleaned in exactly the same way as the chassis, except that only Q-tips are used. I clean the components as well as the circuit board itself. Coil shields and IF transformers are cleaned along with everything else. You must be very careful not to let excess fluid get anywhere, but with a little practice you will find that this isn't a problem. With the Q-tips you can get down between components and really get the dirt out. The photo below shows what the circuit boards will look like after they have been cleaned and waxed. The wax isn't just for looks. It helps to prevent dust from adhering in the future, so that a simple vacuuming with a camel hair brush will be enough to bring things right back.

It all takes a lot of time, but in the end it is worth it!

Interior Before And After Cleaning And Polishing

Click on the image for a super detailed view.	Click on the image for a super detailed view.
Interior Before Cleaning Before cleaning the circuit boards and tubes are covered with dirt. The relay, IF cans, and chassis are covered with a yellow haze. The top of the high voltage cage is covered with a couple of large dirt patches.	Interior After Cleaning After cleaning the circuit boards are clean and bright, and the tubes gleam in the light. The yellow haze on the relay, IF cans, and chassis is gone. The top of the high voltage cage shines without any dirt.

High Voltage Cage Cleaning:
The high voltage cage is cleaned in the same manner as the chassis, but it is much more difficult to clean within the confines of the cage. The cage is also much dirtier than the rest of the chassis because the heat from the final tubes causes more air to circulate, bringing in more dirt. The final tubes are carefully removed and then Q-tips are used with Fantastic to scrub away the dirt. With the Q-tip tool, the Q-tips can be bent into a gentle curve, allowing one to clean under the coil and other areas otherwise inaccessible. With heavy dirt, the Q-tip is wet with cleaner a little more than normal, and the dirt is scrubbed into a slurry. Before the slurry dries, a fresh dry Q-tip is used to "mop up" the slurry.

The coils, relay, fixed capacitors, parasitic suppressors, and variable capacitors are also cleaned, but with these you must be careful not to let any fluid wick its way into places it shouldn't be, such as the variable capacitor bearings. A pipe cleaner is used to clean the space between the plates of the upper tuning capacitor. You must be careful around the red RF choke, as the wire is delicate, and you don't want to break it off at the top.

After everything is cleaned, the entire process is repeated using Pledge furniture polish to resist dust build-up and give everything a nice shine. The picture below shows how the high voltage cage looks after it has been cleaned and polished.

High Voltage Cage After Cleaning And Polishing

Click on the image for a super detailed view.

Cabinet Cleaning and Waxing:
The cabinet was thoroughly scrubbed down with Soft Scrub cleaner. This is a very fine abrasive cleaner and it was very effective in removing dirt that had worked its way down into the finish. The cabinet was then flushed with hot water and a sponge was used to remove any Soft Scrub that might have remained. Finally, the cabinet was waxed with Pledge furniture polish. This brought back the original shine and the cabinet looked brand new.

Cabinet Before And After Cleaning And Waxing

Click on the image for a super detailed view.	Click on the image for a super detailed view.
Cabinet Before Cleaning The cabinet was initially filthy, as can be seen in this photo. Some areas had a black "grime" on them, others just looked very worn.	Cabinet After Cleaning The cabinet shines as if brand new.

Front Panel And Knob Cleaning:
At first I considered removing the front panel for cleaning but decided that it would be easier to clean it in place.

The first step in cleaning the front panel was to carefully remove the knobs. Next, the set screws were removed from the knobs, and the brass reducing sleeves in the Final/Pate Tune and RF Gain controls were carefully removed. The knobs were then scrubbed down with a toothbrush and SoftScrub. The toothbrush dug into the nooks and crannies on the knobs and removed any dirt that was there. The toothbrush also removed any patina that had built up on the knobs over the years. After washing, compressed air was used to blow out any residual water and the knobs were left to fully dry. The knobs were then waxed with Pledge furniture polish, which brought back the original shine. Some of the knobs had bent skirts. These were carefully straightened with pliers. The set screws and bushings were then reinstalled.

After the knobs were cleaned, the retaining nuts and washers were removed from all of the controls to expose as much of the panel as possible. The panel was then broken down into five sections: upper left, lower left, upper right, lower right, and center.

To clean a section of the panel, a rotary Oral B electric toothbrush and Softscrub were used. A rotary electric toothbrush was used because of the random structure of the wrinkled finish. Some SoftScrub was applied to the toothbrush and then manually brushed onto a section of the panel without power to the tooth brush. This prevented spattering of the SoftScrub. After application, and after the toothbrush was in contact with the panel, the toothbrush was turned on and the the panel section scrubbed down. Screws, the headphone jack, and the mic jack were scrubbed as well, but no abrasive was allowed into sensitive interior areas. To avoid spattering, the toothbrush was never powered unless it was in contact with the panel. After a section was scrubbed, it was carefully rinsed with a small sponge. Then the toothbrush was used again with pure water only. This scrubbed out any residual abrasive. The water was absorbed with a paper towel. This was continued until all of the abrasive was removed. The sections were done in the order listed above to provide some overlap.

When cleaning the center section, great care was used to avoid getting any abrasive into or onto the frequency window or the front of the meter. To do so would dull the plastic. Instead, Fantastic cleaner and Q-tips were used to clean the front and back of the frequency window and meter. The front of the round rotary frequency scale was also very carefully cleaned with Q-tips and Fantastic cleaner.

After cleaning, the panel was waxed with Pledge furniture polish, and the retaining washers, nuts, and knobs were reinstalled. The result was a front panel that looked brand new, as can be seen in the pictures below:

Front Panel And Knobs Before And After Cleaning And Polishing

Click on the image for a super detailed view.	Click on the image for a super detailed view.
Front Panel Before Cleaning Before cleaning the front panel is filthy, with some kind of black "grime" everywhere. The slide switches, meter, and frequency scale are dirty.	Front Panel After Cleaning After cleaning the front panel is clean and bright, and the knobs gleam in the light. The frequency scale and meter are easy to read and the the transceiver looks brand new.

Replacement of the Crystal for the 29.5 MHz Band:
When I first got my HW-101 I found that the 29.5 MHz band was dead. All other bands were fine. Connecting a frequency counter through a small capacitor to pin 3 of V19B, the cathode of the heterodyne oscillator cathode follower, showed that there was no output on the 29.5 MHz band. Since there was output in all other positions of the band switch the problem had to be with the crystal, with L608 ,with the adjustment of  L608, or with the connections to them.

If you have a crystal that doesn't seem to be operating, don't replace it with another until you have:
1. verified the connections to it and its associated coil via the bandswitch
2. tried readjusting the coil associated with it

One way to test the coil is to use a grid dip meter next to the coil to see if it resonates at the crystal frequency. Note that the crystal frequency is 8.895 MHz higher than the frequency marked on the band switch. For example, in this case, 29.5 MHz + 8.895 MHz = 38.395 MHz.

In my case, an ohmmeter verified that the connections to the crystal and L608 were OK. My grid dip meter confirmed that L608 did resonate around the crystal frequency of 38.395 MHz. Adjustment of L608 through its entire range did not fix the problem. The problem had to therefore be the crystal.

I was fortunate to have a spare HW-101 for parts. I used Solder-Wick and carefully unsoldered the defective crystal from my HW-101 and the "new" crystal from the spare HW-101. I then soldered the "new" crystal into my HW-101. When soldering and unsoldering crystals use as little heat as possible. Crystals can be ruined with too much heat.

I had no way of knowing the condition of the "new" crystal until it was soldered back into my HW-101. It did not initially work, but it came to life after I readjusted L608. Problem solved!

Lacking a spare HW-101 for parts, it is possible to find replacement crystals on the internet from ebay and other sources. Remember that the HW-101 and SB-102 use the same crystals. This may improve your chances of finding a replacement crystal.

Coil Cover Modification:
The cover on the driver plate, driver grid, and heterodyne oscillator coils should be in place when these coils are adjusted. However, the driver grid and heterodyne oscillator coils in my HW-101 were positioned such that when the coil cover was in place they could not be adjusted. Rather than using a drill to enlarge the holes, which would remove too much metal, I carefully used a Dremel tool with a 1/8" high speed steel cylindrical bit (Dremel #194) to enlarge the holes until the coils could be adjusted with the cover in place. You can see the modified coil cover in the picture below.

Coil Cover Modification

Click on the image for a super detailed view.

The coil cover after modification. Some coils were originally completely covered and could not be adjusted with the cover in place.

Back to Dr. Greg Latta's Electrical Engineering and Amateur Radio Pages

If you have any questions or comments, you can send E-Mail to Dr. Greg Latta at glatta@frostburg.edu