Monday, February 27, 2017

Still digging through HW-100 transceiver and debugging


One of those mornings, I guess. I fired up the HW-100 and scope to continue checking signals when I discovered a complete lack of the carrier oscillator frequency where I expected it. I walked back through stages with no success. 

Back to basics, I guess. Stuck the scope probe on the carrier oscillator tube V16 grid where the carrier oscillation is developed, Good there. Followed it to the balanced modulator. Good there. Found it nulled out on LSB and USB, otherwise developed across transformer T1 on CW and Tune. All as it should be.

Hooked up to the cathode of the isolation amplifier V2 and found no signal. Transformer T1 output is an internal LC network with the coil tunable, coupled through C22 to the tube cathode. Zero signal on the far side of C22, so time to investigate that component and get myself hooked to the transformer output side.
Where carrier oscillator signal disappears
Right out of the transformer, the carrier signal is strong and clean, peaked by tuning the coil, thus it is even more puzzling why I see nothing on the other side of C22 that links the output over to the V2 cathode. The only thing on the far side of the capacitor is a 470 ohm resistor to ground and the tube cathode. 

I yanked the tube out just to confirm that this is not a problem within the tube. No change. I pulled the capacitor C22 and validated its value of 24 pf. I measured the resistor and it is almost exactly 470 ohm. Nothing makes sense here. 

I wonder if I am loading down the coil and circuit, changing the resonant frequency of T1 and blocking the signal?  Time to clip the probe near but not on the line, resulting in much less capacitive coupling. Unfortunately, not enough signal to detect this way. 

I did discover that the carrier null potentiometer was dirty and noisy, easily switching its resistance. Putting some Deoxit inside and working it a bit lead to a much more stable control, that allowed me to null the carrier and keep it that way. 

The output of the isolation amplifier V2 was then monitored, which should give me the same signal at higher amplitude, but it was in fact weaker. This circuit is a grounded grid amplifier (see pin 2 above, the suppressor grid), with the input signal on the cathode and the varying potentials on the control and screen grids determining the degree of amplification. 

The control grid is fed from the ALC (automatic level control) circuit and the screen grid is controlled by the CW/Mic gain circuit. The B+ to the screen grid is switched on only during transmission (including Tune mode), and read about 150V. The control grid potential was -15 to 0 depending on the MIC/CW gain control while running in CW or Tune mode. 

Something is still wrong, I get almost no carrier oscillator amplitude into and out of the isolation amplifier V2. Still haven't figured out the issue but will keep searching. Perhaps I should check the four diodes in the modulator, as these have a history of going bad in similar transceivers. Just in case I ordered some spare diodes from Digikey. 

Meanwhile, I decided to go through the 'voltage charts' provided by Heath, for the unit in both transmit and receive modes. I couldn't get into the one section because all the driver coils and crystals are covered by a RF shield plate, and it wasn't worth the problem to check there yet. 

Everything matched up except for one voltage, which turned out to be the crystal calibrator oscillator plate. That is not switched on unless the PTT-VOX-CAL switch is placed in the CAL mode, which it was not. If the chart had identified that dependency I wouldn't have spent the five minutes researching the 'missing' voltage in the schematics. 

I went over all the modifications I made to the audio, modulator, if and bandpass boards, carefully checking the work for errors, bridges or loss of continuity. Good to know that they were all correctly installed. A few were right at the carrier oscillator, but those changes lowered the plate resistors slightly to increase signal strength. Nothing should be causing the symptoms I face. 

The ALC system lowers amplification on several different tubes and is probably the reason why I have almost no signal strength. I will spend the remainder of the evening studying the ALC circuits and determining what I should be seeing at various tube grids, for when I test again tomorrow.

Sunday, February 26, 2017

Zeroing in on issue with transmit operation of HW-100

A pretty busy day as I had forecast, so only a bit of time I could devote to the debugging. 


I set up the scope with probes on the first transmitter mixer V5A, cathode and control grid. The cathode is the VFO signal and the control grid has the carrier oscillator signal from the balanced modulator by way of the first IF amplifier V3 . I can see the VFO signal running about 700 mV peak to peak. The carrier oscillator with the gain all the way up runs at about 2V peak to peak. 

Assuming the two signals should be about the same amplitude for mixing, this means I need to add an amplifier stage inside my DDS box, to boost its output by about 3X before injection into the VFO cathode follower V20, so that its output is now 2V PP. This is a tiny design and construction task, with the resulting board mounted inside my DDS VFO box.

Missing signal

At this point, I should see a good output signal at the output of the bandpass filter between first transmitter mixer V5A and second transmitter mixer V6. However, all I see on that side is some 12395 KHz signal which is the output of the Heterodyne Oscillator for the 80 meter band setting. There should be a 8395 to 8895 KHz signal from the first transmitter mixer.

I set up a probe on the plate of the first mixer, V5A, to look for the sum and difference frequencies that are input to the bandpass filter. Nothing, not an oscillation to be seen. The tube is not amplifying at all. Time to look at the bias voltages and components surrounding it.

When I look into the circuitry, I see that the tube V5A is biased to cut off while the transceiver is in receive mode, but that should cut off when the unit switches to transmit. This occurs by relay contacts that ground out diode D301. The line to the control grids pass through that diode to a resistor to -130V bias voltage. The relay will ground the diode, which allows the tube to get to a small positive bias and begin conducting.

Grid bias to cut off V5A, V6 and V7 during receive operation
I need to use my VTVM to watch the voltage on the control grid of V5A, I should see it heavily negative when in receive and jump up to a small positive value in transmit mode. What I saw instead was the large negative voltage drop to a low negative voltage, not to a somewhat positive voltage. 

I should see almost ground level, except for the voltage drop across D301. What I find interesting is that there is a user modification I made to this circuit, dropping out a 10K resistor between coil L101 and the control grid of V5A. I will replace the resistor and see what happens tomorrow.

Saturday, February 25, 2017

Fixed microphone, working way through the transmit side of HW-100 debugging and tuning


I put the scope probes on the two ends of the transformer at the output of the balanced modulator, set up the the display to sum the two signals, then tuned it to minimum level while in LSB and USB modes. 

The balanced modulator is designed to block the carrier frequency, by driving equal and opposite voltages to the transformer. Any voice will unbalance the modulator, causing a frequency shift that produces a signal above or below, depending on sideband selected, and passes through because it is unbalanced by the degree of audio input signal.

In CW or Tune modes, the modulator is unbalanced by shorting on side of the transformer, producing the carrier oscillator signal on the output of the transformer. When balanced, the output of the transformer is only those frequencies that shift off the balance point, i.e. speech. 

I will debug the voice input circuit to be sure I am getting the voice signal into the balanced modulator and that it produces the shifted frequency I expect. Nothing, so moved upstream to mic amplifier. Nothing. Moved on to the Astatic mic itself which appeared to have the right connector on it.

The mic stand is mis-wired! The microphone head plugs into the stand with a three pin connector. Pin 3 is ground, hooked to the case and it should be hooked to the shield of the mic cable and thus to the radio ground. Pins 1 and 2 should be wired to two of the four pins in the mic connector at the HW-100, but in fact pins 2 and 3 at the mic stand are shorted together deliberately.

Time to rewire this as it should be, validating each of the four pins on the HW-100/mic connector and through connectivity to the microphone head itself. With that done, I get voice on the audio amplifier output. Further, I get imbalance on the balanced modulator when voice is present.

I then retuned the zero balance on the balanced modulator to absolutely minimize RF output in LSB or USB with no input from the microphone amplifier. In CW and Tune, I get solid RF. I moved on to the transformer T1 that couples the modulator with the isolation amplifier V2 and confirmed the signal gets to the cathode of that tube. I peaked coil T1 to produce the highest amplitude and nicest looking version of the fixed RF signal. 

I can't absolutely confirm that the balanced modulator produces sum and difference waveforms between the carrier oscillator and the voice frequencies because my scope gets confused in its frequency identification logic. Thus, I can't verify the frequency it displays but I do see varying RF signals as I speak. 

Next up, I watched the signal at the output of the isolation amplifier V2, before it enters the crystal filter. The crystal filter has a very narrow passband centered at 3395 Khz, thus allowing only the upper or lower sideband signal to pass, or the CW signal if in CW or Tune mode. The output of the filter will go to the 1st IF amplifier and further. 

I don't see any signal coming out of the crystal filter, but I might be loading it down with the scope probe I hooked up a jumper to use as a morse key substitute, triggering the amplifier when I hit the key. I can here a tone being produced. Also, the ALC goes up and down with the CW gain control. 

Moving on to the first IF amplifier V3, I see the frequencies present at the output, which would be coupled to the first transmitter mixer V5A where it has the VFO mixed with it to produce signals in the range of of 8395 to 8895 KHz as the sum of the VFO (5000 to 5500) and the carrier oscillator (3395 CW/Tune, with additional modulation from voice if in LSB or in LSB). 

The second transmitter mix will combine the 8395-8895 signal with the output of the heterodyne oscillator - the oscillator which is switched to different frequencies by the band switch - to yield a sum or difference that is passed through bandpass filters selected by the band switch. The sum or difference that was selected then is coupled to the driver V7 and ultimately to the final RF amplifiers V8 and V9. 

What appears odd, since the ALC level goes up or down with mic or CW mode transmission, based on the voice level or the CW gain control, is that the relative power reading is always zero. I will look at this next because it is pretty essential to tuning and alignment. It also serves to tell me if the meter circuit is bad or if the transmitter is not operating. 

The way the relative power is measured is a very simple circuit that rectifies and smooths signals from the final tank circuit, consisting of two resistors, a diode and a capacitor. I am pretty sure I am not energizing the transmitter with much power, since the dummy load would get hot if I were. Also, cathode current to the final tubes doesn't change noticeably. 

I have a dinner party to attend tonight, thus have to stop work right now. I also have a pretty busy day tomorrow, which means I won't make much more progress on this over the weekend. 


My replacement capacitors arrived late last night, ready for installation today. Fired up the unit and found it working plausibly, having tested an 8uf 450V electrolytic on the bridge and leakage functions. My 20K test resistor gave a slightly high answer on the Rx100 scale and a odd reading on Rx1. 

No interest in using it for resistance measurement, will not investigate further. The magic cye is a bit dim, but still usable. Back in case and ready to use testing/reforming electrolytics in the future.


We have a session planned for Friday where I will attempt to archive as many of the cartridges on hand as I can. After that, we will test the tool on the Alto's Diablo drive and also debug the booting problem we are facing with cartridges I wrote.

The session will begin early because we have a long-ish interruption at midday, where we are panelists at a special lunch at Xerox Parc, talking with the staff about our restoration experiences. Afterwards we hope to get a few more hours of work in.

Friday, February 24, 2017

DDS working in enclosure, now debugging the transmit issues of the HW-100


This morning I tested the wiring of the DDS enclosure and for correct operation. The unit tunes up and down, the 'up' and 'down' bandswitching buttons work, and the RIT (Receive Incremental Tuning) button works properly. RIT allows me to hold the button and tune an offset so that transmit and receive occur at different frequencies, using the button to control which frequency is active.

The enclosure is just about done. I still need to mount the Heath style feet on the bottom and to put in the sheet metal screws to hold the box closed. It is operational, however.

DDS VFO wired up in its enclosure and passed all tests
I hooked the DDS and the microphone up and attempted to begin the transmitter alignment. My first try didn't seem to work, but I need to work methodically through the steps since the very first act is to set the bias; that could keep the final amplifier tubes in cutoff if not adjusted properly. 

After adjusting the bias potentiometer, the transmit tubes do draw 50ma cathode current when energized with no signal. The next step in the alignment is to switch to Tune mode and advance the Mic/CW gain until relative power is seen. It wasn't. Time to start tracing the signal through the various stages and learn where something has gone awry. 

After some work this afternoon, I had a few conclusions. First, the carrier crystals are quite close but not exactly on their 'ideal' frequency. Second, the output of the isolation amplifier is essentially zero. 

I looked at the output of the carrier oscillator, which varies based on the mode of operation. That shifts the carrier to the left or right for SSB operation, or directly on the tuned frequency for CW or Tune modes. 

For CW and Tune, the oscillator should be at 3395.4 KHz but was about 200 Hz high, an error of ,006%.  USB should be at 3396.4 KHz but was off by 20 Hz or less than .0006%. LSB should run at 3393.6 KHz but was off  by 30 Hz, also vanishingly small. I could cut a trace on the PCB and put in a trimmer capacitor for the CW crystal, but the error is too low to care about.

The output of the isolation amplifier was varying randomly from about .85 MHz to about 1.6Mhz for receive modes, then in the Tune mode it sometimes ran at about 2.3MHz and other times was off. This is not good. 

I think I should have the carrier oscillator frequency amplified here for Tune or CW, and the carrier balance control should eliminate the carrier entirely for USB or LSB. That means that output for LSB or USB will only be on the sideband away from the carrier when there is voice on the microphone during transmission. CW only passes the carrier if the code key is pressed, biasing the IF amplifier on. 

I need to back up earlier in the chain, before the isolation amplifier, where I should see the carrier oscillator signal. Next spot to look was the output of the cathode follower that drives speech signals to modulate the carrier.  

I have to admit that I performed a local modification, swapping the coupling capacitor between the speech amplifier and cathode follower/mixer to a larger capacitance to allow through a bit more bass from voice input. It shouldn't affect the carrier oscillator at all, but if I messed up the change it could be an explanation.

If the carrier is not getting through then the isolation amplifier will be boosting random coupled signals. I need to probe further back but it is slightly challenging to find scoping points. With a PCB and tightly fitted components, it takes time to find a place to hook a scope probe anywhere near the desired signal. 

Also, the high voltages require extra caution. With 800V, 350V and -130V present, I could get quite a shock. Only the final section uses 800V  but across the rest of the unit it is possible to bridge 350 and -130, almost 500V of relative potential. 
Section of schematic I am observing

If I have the signal on the cathode of the cathode follower but didn't see it on the other side of the capacitor that couples it to the next circuit, That could be because one of the diodes is shorted, since the voltage drop across working diodes would leave the AC signal observable. 

I had to redo the tests on the grid of the cathode follower, because I should see carrier on all four settings, even if the follower isn't letting that signal through. The signal that is missing comes up into the diode bridge, so that any coupling of it back to the cathode follower is coincidental. The cathode follower exists to drive the audio signal into the bridge.

The routing of the carrier oscillator signal to the diode bridge is somewhat indirect. While I know that the carrier is produced correctly for all four mode settings, I am less sure about how it gets up to the bridge and the schematic is a bit circuitous ( pardon the pun ).

Thursday, February 23, 2017

C-3 and HW-100 work, construction of DDS box


While removing the old capacitors that will be replaced, I found a funky modification made to the underside of the chassis. Someone had soldered 20 ohms of high wattage resistors across a new tie strip that was added. It is a poor quality change, just as I found quite a few solder joints which were amateurish at best.

Resistors inserted in series with magic eye tube filament
Upon investigation, I found these resistors were added in series with the filament on the magic eye tube. Looking at the tube, I discovered it was a 6E5 magic eye, not the 1629 tube supplied with the kit. Aha - the filament on the 1629 is 12V but the 6E5 requires a lower 6V supply. The series resistors were intended to drop the voltage.

Until I replace the capacitors and power this up, I really don't know if the 6E5 tube is working. If it works well, I will dress up and leave this little mod in place but if I need a new tube, I will just go back to a 1629 and drop the extraneous resistors. The wiring of the socket for the 1629 is different from that of the 6E5, thus I have to rewire that also if I switch back.


I am working on the layout for the external box that will hold my direct digital synthesizer that replaces the VFO inside the HW-100. I need to lock in the orientation I want and the placement of the LCD, rotary encoder and three pushbuttons. Further, I need to decide on the power supply and RF output connector types and placement.

Aluminum box for DDS VFO

DDS and LCD to mount inside
Rotary encoder and RF output jack

Buttons and knob to complete the front panel
Perhaps I should create a small board to deliver 12V from inside the HW-100 for use in the DDS VFO. I would likely take the filament power, rectify it and regulate it, putting some small board inside the HW-100 to accomplish all of this.

Meanwhile, the microphone arrived - a historically accurate choice, an Astatic D-1604, and I hooked it up to the transceiver. Now, I could perform the transmitter alignment steps from the Heathkit manual. That is, I could if I hadn't disassembled the DDS VFO. First, I have to complete the construction of the DDS in its box.

I punched the holes for the three pushbuttons and mounted them. The hole for the LCD is cut out, although a bit uneven because of the way I had to make the rectangular slot. My hole for the rotary encoder is too small, thus I have to step up one size and punch again.

I will need to carefully measure, mark and drill the small holes for the mounting hardware that will hold the LCD in place behind the panel. At the same time, I should create the holes on the bottom to mount the main DDS PCB.

I still have not finalized the RF output and power input jacks, thus can't punch the holes in the rear yet. I have a number of heathkit style rubber feet and will place four of them on the bottom of this enclosure to allow it to stand a bit above the HW-100 cabinet top.

I chose to run over to Anchor Electronics to pick up the RF and power jacks, which would let me complete the project tonight. Sadly when I mounted the LCD panel, the backlight broke from flexing of the board. It works but need external light.

After dinner, I had both boards mounted and was wiring up the controls. I finished at night but wasn't ready to test it out until the morning.

This and that

Worked on one of the 1401 systems at CHM today, replacing a front panel switch we thought was bad but the problem persists. The Start Reset button does not reset the machine properly. This means that once an error condition arises, the system must be power cycled to run anything else. 

Participated in an oral history session with Bob Feretich who, along with Grant Saviers and others, created the Tape Emulator box that helped restore the first 1401's Tape Adapter Unit (TAU) about a year faster than would otherwise have been possible, and provided virtual tape drives and educational program support. 


I picked up a nice aluminum box to mount my DDS VFO into, beginning to cut the openings and mount up parts. I removed the rotary encoder and jacks from the circuit board and will place them on the face of the metal box. The LCD is also to be installed on the face of the box. 


I picked up this unit, which uses a 'magic eye' tube to indicate when a resistance or capacitance bridge is balanced, in order to determine the component value. It also tests for leakage at a variety of voltages up to about 450V, which is a great way to test and reform old electrolytic capacitors. 

I inventoried the parts that likely need replacement and ordered them from Digikey. Essentially it is all but one of the capacitors inside, since they were types that are notorious for becoming leaky with age. When the parts arrive, I will replace the old components and power this puppy up. 

Tuesday, February 21, 2017

Aligned receiver section, preparing to do transmitter alignment of HW-100


I wrapped up the conversion to using the DDS VFO, powered up and confirmed that the unit is tuning properly. I haven't confirmed that the output is free of coupled frequencies through the cable that joins the DDS to the VFO, but at some point before operating on the air I will hook this to a spectrum analyzer and look at the output of the VFO. 

I proceeded through the receiver alignment, using the 100KHz built in calibration oscillator. All this assumes that the oscillator itself is right on frequency. There is a procedure to follow to validate this, which will be my next move.

I had a plastic project box that I initially though I would use to mount the DDS, but it is too small and not shielded, both are a problem. I am looking for a larger metal project box which I can install my DDS VFO and its LCD screen. It will sit atop or alongside the HW-100 rig.

I notice that there are minor differences in the frequency that the 100KHz signal peaks as I switch through the bands - this indicates that the crystal oscillators for the final frequencies that are mixed with the IF are a bit off their nominal values, each differently. There are trimmer capacitors on those crystals, I believe, which would allow me to get this spot on to the desired frequency. All that REQUIRES that I can trust the 100KHz calibration frequency, making that tuneup a priority.

I have about ten feet of wire stuck into the antenna jack and sprawling across the floor. Not much of an antenna but I was able to find some CW (Morse Code) going on in the 15 meter band. I need a really strong, i.e. local, signal to hear anything without a proper antenna, but at least I have reception, tuning and so forth. 

My dummy load arrived by early evening, however I still couldn't move on to aligning the transmitter section of the transceiver. I didn't have a morse key or a microphone with PTT yet, thus no way to trigger the transmitter. 

Checkout of HW-100 transceiver and preparation to use DDS VFO in place of the heathkit supplied VFO


I did some initial cleanup on the apparent rust on the HP-23 Power Supply and discovered that it is mostly a brown gunk adhered on top of the paint on the main transformer and on top of the varnish on the choke. I got about half of it removed with a very mild cleaner (Formula 409) and the areas that were exposed show the paint or varnish, not bare metal. 

I went through all the initial resistance checks using my V7 VTVM, with its 11M Ohm input resistance and everything checked out spot on. I flipped the unit on its side to give me access to the underside of the circuit boards where I can make all the voltage measurements specified in the checkout section of the manual.

I flipped the power switch on, but nothing happened. HP-23 plugged into mains, cable connected between HP-23 and HW-100, but zero signs of life. This may be yet another problem in the cable, due to the flimsy small gauge wire and total lack of strain relief. I will pop the cap off the connector at the HP-23 where the earlier defects occurred and take a look. 

Nothing was visibly wrong, so I plugged it all back in and voila, the transceiver turned on. I began the alignment as outlined in the manual - using the built-in 100KHz oscillator selected by moving the slide switch to CAL. I haven't verified the correctness of the frequency of this oscillator, but it is something I can do using my frequency meter before I go too far through this process.

I decided to pull out the VFO (main tuning knob and its associated electronics) to see how I would modify it to work with my direct digital synthesizer (DDS) VFO. The existing box sits behind the main dial, contains a variable capacitor for the tuning and an oscillator, plus a 6AU6 tube which amplifies the oscillator before its signal exits the box and travels down to the rest of the transceiver circuitry. 

My DDS needs the tube amplifier,  thus I just need to disconnect the oscillator half of the circuit in the box, run in the DDS signal and connect it to the grid of the amplifier tube. I have found two components, a 10 uf capacitor and a 47 ohm resistor, which are the only coupling between the oscillator and the tube amplifier.

Plan to disable oscillator on left, add input jack for input from DDS VFO
I disconnected those components, mounted a signal input connector for the generated frequency from the DDS, and made sure that the oscillator was disabled from operation. I had to choose a good spot for the new connector, permitting the existing oscillator to be reconnected at some future point if desired. 

Upper half is the oscillator, bottom half is tube amplifer
Top of VFO box - need to find entry point for DDS signal
I decided to remove the trimmer capacitor labeled "SHIFT ADJUST" and use that hole to pass the DDS signal inside. I completed this and remounted the oscillator box.

Trimmer oscillator to be removed to route input cable to tube grid
While soldering on the B+, bias and filament wires to the feedthroughs, I had the small wire connecting B+ to the PCB below snap off. I had to remove the wire fragment and install a new wire to power the VFO box. 

I have to hope that the shielded cable that runs out of the VFO box and through the transceiver cabinet out to my digital signal synthesizer will not pick up spurious frequencies from other circuitry such as BFO, IF or final outputs. If it does, I shall need to install the DDS in a metal box just above the VFO unit. That complicates things because I will need to remote the rotary encoder, pushbuttons and LCD panel then.

Monday, February 20, 2017

powered up HW-100 and fixed first problem


Marc will rerun the boot attempts with the pack I wrote and with a known good booting pack, adjusting the capture and formatting rules to give me traces I can compare more successfully.


The dummy load and cable are at the local post office but with the postal holiday tomorrow, they won't arrive until Tuesday. I can't switch on the transmitter until the load is applied, but I can check out the unit under power, make initial adjustments and see how it receives signals with an inadequate short wire antenna.

I set up the power supply, joined the cable between it and the HW-100, and switched it on. Nothing. The power supply light didn't turn on at all. I pulled off the cable and tested the power switch pins on the connector that will complete the power supply primary circuit. They worked fine.

At this point, I decided to check continuity in the cable. Several pins had no connectivity at all! Bad cable. I opened the caps on the two plugs and the problem was immediately obvious. The wires had failed due to being yanked, separating several leads from the remnant down in the pins.

Wires snapped off on the power supply end of the cable
I took the opportunity to remove the phone jack cable that had been piggybacked on the connector to fit some local modification by the original owner. I removed all the wire remnants and prepared the cable ends to be soldered back in the connector. The design of the cable is very odd - no strain relief at all for the wiring on either end.

Receiver end, with spurious phone jack connected to unused pins

I soldered up the connector, checked connectivity with the meter and then attempted power up of the transceiver once again. It now powers up and I can hear hiss from the audio section when I turn the audio gain up. No antenna at all and not surprisingly, no reception.

I did the initial tests from the instruction manual, which showed that I have correct voltages delivered to the board and that all the tube filaments glow. I set the function to Calibrate, which will produce signals every 100 KHz to allow calibration of the receiver. I can hear the tones at various points on the tuning dial, but the S meter does not deflect from its zero setting.

Hooked up a short bit of wire and tuned in some CW signals on several bands. I clipped a voltmeter across the meter connections and saw that current is flowing through the meter. I then noticed that as the current increases with stronger signals, the meter pins below zero. Hmmm. Something to debug.

I will be using a scope and some reference signal to do alignment, but still lack a microphone to drive the rig. It will arrive in about a week, after I have the dummy load to let me also align and tune the transmitter.

I found some tests to do for the meter circuits, but none of them moved the S-meter upwards. This is not good, seems like I have a continuity break in the meter wires or the meter itself is defective. I did check the two resistors hooked to the meter terminals, and found they both were high. That would subject the meter to higher voltages on each post.

With the meter removed from the panel, I hooked up a resistor in series with my ohmmeter to see if the needle will deflect. Worked perfectly, thus the problem is not the meter itself. The wiring seems good but all I can check is ALC, the first of three meter settings, because the other two require a dummy load since they switch on the transmitter (Relative Power and Plate Current).

The meter is connected to an IF amplifier tube circuit in two places and will measure the relative voltage of the two lines. When I have the RF Gain all the down, the meter should be at full deflection. As I advance the RF Gain, the meter deflects downwards, which is the correct direction.

Earlier I had picked up some CW stations but that wasn't working tonight. I did swap two 6AU6 tubes, perhaps one of them is marginal. I will swap them back later. I picked up carrier that appears to be SSB transmissions on 21 meters, but the receiver is too far out of alignment to capture audio.

Lots of noise and erratic contacts, which is to be expected. More cleaning will be required. 

Saturday, February 18, 2017

HW-100 power supply restored, disk tool archiving packs correctly


My copy of Contralto was corrupted, but when I reinstalled it, I was able to take the cartridge image I had uploaded using my disk tool and boot it up just fine. Whatever problem is keeping cartridges from booting, it does not impair the reading capability.

While we took traces from the logic analyzer yesterday, they are formatted oddly compared to the way we traced prior reading attempts. As a result, I am having difficulty following them to see what is different between booting one of the cartridges I wrote and others that work fine.


My update kit for the HP-23 power supply arrived today, I assembled it and installed it into the unit. I then tested the unit to check that it produced the +800, +300 and 130V DC as well as filament power. While it also generates an adjustable bias voltage, that is not used with the HW-100 receiver.

HP23 Power Supply as received
Original parts inside (plus big filter capacitor cans)
Parts stripped out and ready for new circuit board

Soldering the board itself was easy, but getting all the loose wires from the chassis to the proper points on the board was more devilish. One of the wires was just too short to reach the hole on the board, yet it was the 800V transformer output which I really didn't want to splice. 

I ended up with no choice in the matter, thus I had to put in a join on a spare terminal strip. There were issues with ground connections that required some improvisation. Finally, the original RF filters on the input were gone but the wiring had to be restored to stock connections between plug, power lamp, power switch, transformer and the AC switching pins on the connector that runs out to the transceiver. 

After this was all done, I double checked the wiring against the schematics to be certain it was right. Some careful comparisons, continuity, resistance checks and other tests were performed before I would allow power to flow in the supply. 

New PCB visible under holes from removed filter capacitors
Plug in, switch on and I measured everything on spec. A bit over 800V for HV, over 300V for LV, under -130V for fixed bias, about -80V for adjustable bias, 13+ for the 12V filament supply and 6,5V for the 6V supply line. The unit is ready to go.

Restored and updated HP-23 power supply

To do the testing, I used my recently restored V-7 Vacuum Tube Voltmeter. One reason I got it was to measure the relatively high voltages in old tube equipment.

V7 VTVM ready for use
When I received the power supply, it was mounted inside the Heathkit SB-600 speaker enclosure. It was cheaper to buy the power supply WITH the speaker enclosure than to buy one along - and I have a speaker to use with the transceiver as well. Needs a bit of cleaning on the paint job and some dings handled, but otherwise is fine.

Dusty, dirty and dinged speaker cabinet

Friday, February 17, 2017

Did more Alto testing and fixed up the VTVM


I was able to get to the test setup in the afternoon for a quick test of a corrected cartridge image and to try out my new writing logic. I write an image and can read it back word for word. Those written cartridges will NOT boot, however. Not sure why at this point.

I collected data during the testing and looked it over at home. Unfortunately, it was not definitive. I did archive another cartridge, with zero checksum errors encountered. What I don't have is a known image of that cartridge to compare against what I read. I will try to boot it under Contralto, the Alto simulator, to see if it seems good. However, for some reason my Contralto application went bad.


We also worked on the ethernet tool that Ken Shirriff is building, making some progress as he debugs it. Writes and reads packets but has problems when the packet gets above about 18 words. He believes he has a timing issue within the Beaglebone and his code.


I kept at the VTVM, flexing things and examining everything. Finally, I spotted a nearly invisible less than hairline crack in a trace joining two pins of a tube socket. I used a bit of spare wire lead clipped off another component to create a flexible connection across the break. Things are working now, with the exception of the 6V flashlight bulb that illuminates the meter scale, since it is burned out. I will order the bulb but have put the tool on the bench now since the light is merely a cosmetic touch.

Thursday, February 16, 2017

Random work on restorations


Today I prepped the HP-23 power supply for the update kit that is coming. All the capacitors, resistors and diodes inside are removed, as are the tie strips. A single PCB has the replacement parts installed and it will sit inside the power supply enclosure. 


I took the VTVM apart, checked the circuit board and it seemed fine. I powered up with the parts loose and it worked perfectly. Put it all back together, and the bad behavior is back! I did this a few times and can't figure out what intermittent problem is causing this. Frustrating. I will have to start tracing circuits and monitoring voltages until I can narrow in on a suspect area.

Wednesday, February 15, 2017

Got power supply for HW-100, replaced capacitors on VTVM


I finally completed the setup of the direct digital synthesizer (DDS) so that it implements the eight bands of the transceiver, generating the subtractive intermediate frequency of the VFO it will replace. When tuning any band, the low end of the band generates a 5.500 MHz signal, then as the dial is rotated upward through the band our output frequency declines down to 5.000 MHz. 

Each band is 500 KHz wide. The frequency counter verified the proper output frequencies for all eight bands. It was painful to set this up, since I had to reverse the earlier settings - e.g. take the band 8 start frequency from 29.5 MHz to -29.5 Mhz. 

That step alone took almost 120 rotations of the dial. The IF offset had to move even further, about 140 rotations. Multiply that by seven other bands, albeit with decreasing numbers of rotations particularly on the lower bands. 

The remaining tube arrived - a 6AU6 - which I verified on the tube tester since it was tossed into an envelope and mailed. It checked out fine so I installed it into the socket. 

Today I received the Heath SB-600 which is a speaker cabinet with the HP-23A power supply installed inside. This power supply will deliver the 800V, 300V, -110V and filament voltage needed by the HW-100. I have to check out all the filter capacitors and replace any that are still original.

The unit was packed foolishly, thus the outer carton was partially torn open, one foot of the speaker cabinet was lost and I had some minor shipping damage. Still, it is in good shape for the restoration I need.

I removed the power supply from the speaker enclosure and moved it over to my bench. All of the capacitors are original electrolytics. They all must go. I built up a list of the values I needed in order to go shopping. 

However, I came across a nice kit that replaces everything inside, ordered it and will wait for it to arrive. I found that my HP-23 has been modified - need to figure out what was changed and either transfer it over to the new kit or dump it.

The major modifications are filter inductors on the AC input line, plus bypass capacitors. These are intended to block any RF from leaking back into the AC mains, I believe. This was probably a change made by the prior owner as a concession to neighbors. 

The other modification I found was a phono jack attached to the power connector where it would plug into the prior ham rig it powered This jack makes use of a unused pin for ground and borrows the variable bias power pin for the center lead. On the power supply side, the cable is not soldered into the variable bias pin, thus that voltage is not carried on the cable at all. Fortunately, the HW-100 does not make use of the variable bias line. 

The Heathkit manual has a resistance chart, which I used to see that the resistances appeared close to the published values at all the points they listed. When I came to measuring tube V5 on the bandpass board, however, two points had open conditions but were listed to display relatively low resistance.

More mysteriously, when I looked at the PCB I could see that those two tube pins, 8 and 9, were on traces that had absolutely nothing else connected to them. The open circuit was what I should have, yet the PCB has holes missing components and the chart gave target resistances.

Hmmm. I took to the schematics where I found one half of the tube (V5A) but nowhere could I find the other half of this dual tube. It dawned on me that in fact the documentation was wrong about those resistances, because the V5B half of the tube is not implemented on the HW-100. 

The same circuit boards are used for the lower cost HW-100 and the higher priced SB100/SB101 transceivers. The HW series was created by dropping certain features from the SB version to lower costs. For example, the SB allowed use of an external VFO, but it was dropped in the HW machine. I believe that the second half of tube V5 hosted one of those dropped functions.

On the SB-100 schematic, V5B is used as a "crystal oscillator" for a fixed frequency, which the operator could use for transmit, while still receiving using the tuning dial, or could force both transmit and receive to the one fixed frequency. As an example, this was used with the MARS (Military Auxiliary Radio System) in which ham operators work with the military as a means of supporting and augmenting communications for the armed forces. 

Thus, I solved the mystery of the missing tube section and know for certain that the resistance chart included with the HW-100 manual is partially in error as it contains values that would be found on an SB-100 instead. 


My new capacitors arrived from Digikey, thus I put the VTVM on the bench and began to replace the originals. With everything in place, I fired up the meter. Unfortunately, the behavior was not good. I suspect I got further cracking in the circuit board from mounting it back together. 

I will put this aside and open it up again tomorrow, when I am ready to deal with the funky condition of the meter and repair as much as possible on the fragile PCB. It had been working nicely before the replacement of the capacitors and resulting reinstallation of the PCB on the meter and brackets. 

Working on HW-100 and V-7, build of Digital Direct Synthesizer oscillator


I bought a kit to build replacement probes for the VTVM, which arrived yesterday and is assembled and ready for use. Heath used a phone jack for the main DC probe, plus banana jacks for ground and AC/Ohms probes. 

New probes for the VTVM
 I reinstalled the potentiometers that are now turning freely. With them soldered in place, I just need for my replacement capacitors to arrive from Digikey before I can wrap up the restoration and put this meter into service. They are projected to arrive on the 16th.


My remaining tube and the power supply are still drifting through the USPS system. The tube was due yesterday but is now showing "delayed delivery" without an updated date. The power supply is supposed to arrive tomorrow. 

I bought a direct digital synthesizer (DDS) to replace the VFO in the transceiver, giving me higher accuracy and stability for both transmit and receive. It was a kit that I quickly assembled and checked out today. It appears to work well, passing self test and displaying appropriate messages on the LCD panel.

The unit will mount atop the HW-100 cabinet to avoid having to hack the front panel apart to mount controls. I will need to manually synchronize the band settings on the transceiver with the appropriate DDS bands in order to have the LCD display the true frequency. 

Otherwise, the HW-100 would still work properly, tuning an intermediate frequency over a 500KHz range, which is added to the crystal selected frequency to yield the desired band.  That is, the DDS always produces signals over a 500KHz span, but it might display this relative to the wrong band's starting frequency. 

If on 80 meters, the unit is tuning to 3.500 to 4.00 MHz, while on band 2 (40 meters) it is tuning from 7.000 to 7.500 MHz. The VFO produces the same IF value in either case, but displays based on the DDS logical band I have selected. If I put the HW-100 in band 1, tune to 3.720 Mhz, but tell the DDS I am on band 2, the DDS LCD display will have 7.220 Mhz showing while I am truly tuned to 3.720M.

The HW-100 first produces a 3.395MHz IF, modified slightly depending on whether SSB or CW is chosen. The VFO then produces a signal that ranges from 5.000 to 5.500 MHz as the second IF. This is mixed with the first IF to create a third IF at 8.395 to 8.895 Mhz. The crystal oscillators and bandpass filters adjust this IF to the final RF frequency of 3.5-4, 7-7.5, etc all the way up to 10 meters at 29.5-30 MHz. 

You may wonder how it produces a final frequency of 3.5 Mhz with an IF that ranges from 8.395 to 8.895. The crystal frequencies are mixed and produce but the sum and difference. The bandpass chooses only the difference signal. Thus, the crystal at 12,395 MHz is mixed with 8.395 to give 21.79 sum and 4.0 difference. As the VFO increases from 5.000 to 5.500, the second IF goes up to 8.895 giving 21.29 sum and 3.5 difference. 

From this, you can see that the HW-100 VFO actually starts at 5.500 MHz when the dial is at .000 and decreases to 5.000 while the dial states .500, added to the band start. By this, I mean that to tune 3.510 the VFO has to be at 5.490 to give 2nd IF 8.405 and difference 3.510. 

Fortunately, the DDS can be set up with a negative or positive IF offset, which means I can have it move downward in frequency between 5.500 and 5.000 as I tune upwards through each band. It just requires some setup parameters to define my eight HW-100 bands and the appropriate offset. The goal is to have the DDS always generate between 5 and 5.5 MHz. 

By the end of the evening I had set up the first 5 bands and the following morning I wrapped it up. Putting the DDS on the frequency counter, I verified that the output frequency is always 5 to 5.5 Mhz on all bands, while the LCD display is showing the band. However, I still have a problem.

The dial direction is still inverted. That is, on the 3.5MHz band, with the dial set to 3.500 MHz output, I should be seeing 5.500MHz and as the dial gets up to 4.000 MHz the true output should have dropped to 5.000 MHz. Instead, this begins with 5.000 MHz at a dial indication of 3.5 and goes upward as the indication moves to 4.

I have figured out that there is a way to reverse dial indication, but it means a new round of configuration of the DDS. I will take care of this later today.


We may get a test shot late Friday, but several people have doctors appointments in the morning that restrict our time together. I would check the ability of the tool to write a bootable image and to read an existing cartridge correctly, assuming I get the chance. 

Monday, February 13, 2017

More restoration work on HW-100 and V-7 units


I wrapped up all the modifications moved on to replacing the broken rubber belts which couple front panel controls to variable capacitors inside the chassis. Later, when I get a mini jack, I will replace the large phone jack on the front for the headphones so that I can directly plug in modern headphones.

I am still waiting for one vacuum tubes and for the power supply, plus will need to restore the PS before power up. Still, I am working through all the resistance checks listed in the assembly manual.

 I did receive a direct digital synthesizer kit which I can use as a replacement for the VFO inside the transceiver, to produce a more accurate, stable and reliable frequency than the Heathkit oscillator. I will be wiring this up tomorrow.


Two potentiometers on the unit were frozen - they trim the zero and infinite ohms positions of the meter. While I have replacement units on order, I decided to disassemble one to figure out what caused them to bind up.

It turned out to be frozen grease at the point where the shaft enters the front of the pot, under a circlip that holds the shaft in place. When I removed the circlips and forced out the shaft, it dislodged the bad grease and allowed me to reassemble the pot which now had free movement. Electrically they work fine as well.

Thus, I will reinstall these original pots into the meter, rather than the new pots on order, but still need the replacement capacitors to complete the restoration. Those parts should arrive by the end of the week. I have received the probe kit to build the proper probes for use with the meter.

Sunday, February 12, 2017

HW-100 Transceiver modifications continue


This morning, I worked on the modifications to the transmit/receive relay and to the IF board. Slow but steady, removing old components and installing changed value components to fit the mods. Very hard to get access to some of these. 

I visited with my daughter in the east bay midday and was just able to get started on the modulator board changes when I returned. One of the resistors was almost completely beneath a chunky capacitor and wedged against other parts. I finally got it out, but decided to insert it on the underneath rather than on the topside since it was nearly impossible to thread the new resistor's lead into a hole underneath the capacitor. 

I did complete the modulator and bandpass board changes, then started into the headphone/speaker mods. Tickets to a show tonight, so won't get everything done, particularly the modification to allow low impedance headphones and have them mute the speaker when inserted.

These changes improve the performance, remove sidetones, remove various kinds of noise, add a bit more bass to voice, support low impedance headphones and a number of other improvements that have been developed by HW-100 and HW-101 owners over the years. Since the HW-101 was one of the most widely used Heath ham radios, it garnered a lot of attention from owners and the result are these improvements. 

Saturday, February 11, 2017

Restoration work V-7 VTVM, HW-100 SSB XCVR


I tracked down the line cord defect to loose screws inside the wall plug - but the unit still isn't working. I don't even see filaments glowing in the two tubes. There is power coming from the power transformer secondary on the filament windings - but the two tubes are not glowing.

The tubes work fine on my tube checker, so the problem lies somewhere on the circuit board between the pilot light where I measured the voltage and the tubes themselves. Sigh. I had expected this to be a mostly working unit to restore, but the reality is different, as so often true with ebay purchases.

At this point, I will partially disassemble the meter to get access to the board upon which the tubes are mounted and do some continuity and solder joint quality testing. I quickly found the problem. The PCB had cracked, breaking three traces including the filament voltage to the two tubes. 

Board cracked, three traces lost continuity
I bridged the three cracks in the traces with short jumpers, but took the time to inspect and test the components on the board in case any of them warrant replacement. I also saw plenty of oxidation on the rotary switch contacts, which has to be cleaned to return this to useful condition. 

After bridging those traces, I discovered a secondary teeny crack that might be visible in the left side, about middle height in the picture. It interrupts the filament voltate to the tube socket that is off screen, as well as another trace. Will repair those and continue checking. 

Jumpers in place, now time to check out resistor and capacitor values on the board. Just in case, I will replace all six capacitors on the unit. The resistors all seem close enough to spec to be good. Firing up the unit, I found that tubes now lit, the meter did its swing up and down as the 12AU7 dual triode warmed up in the bridge circuit.

I tried the ohms with a dead short and used the VOM resistance measuring device to inject a DC voltage. Behavior seemed reasonable. I have two frozen potentiometers - the zero adjust and ohms adjust dials on the front. I can't get them to move even with shots of Deoxit inside and pliers on the shaft. I suspect these will need to be replaced as well.

I have ordered replacement pots, plus the capacitors, for delivery in a few days. Digikey again, when I can tolerate a few days shipping delay and don't seek near instant gratification.

Meanwhile I cleaned up the contacts on the rotary switches and got this into as good as shape as I can prior to receiving the replacement components. I will set it aside for now.


I received my Digikey shipment of the remaining parts necessary to apply all the recommended modifications to the transceiver. I can't do live testing until the power supply I just bought arrives, but I certainly can install the mods while I wait.

I had been shopping for the power supply itself - HP-23A/B - but realized that it also comes mounted inside the SB-600 station speaker, thus it was worth looking at those auctions. I realized that I could buy an SB-600 for about the same as the power supplies were closing at, but also get the speaker and enclosure. 

By early evening, I had made all the modifications to the audio board and was moving on to other boards. There are many wires hooked to the boards, making it quite difficult to work on them. They have to stay in place on the chassis, which leaves limited access for soldering irons and other tools. 

Each part has to be laboriously unsoldered and removed before a replacement can be put onto the board. In a few places, additional parts are added which at least means no desoldering first. For example, a .005 uf filter capacitor was added on a voltage rectifier tube to sop up high frequency noise. 


We have set a plan for the testing of the fixes I created - this coming Friday during our next Alto session. Hopefully this will get everything working well enough to begin archiving all the cartridges. 

Friday, February 10, 2017

Preparing for modifications to HW-100 and testing key components


I accompanied a friend who was inspecting some tape drive equipment for possible purchase. These were two 3420 model 6 drives (9 track reel to reel, 75 inch per second, at 1600 and 6250 BPI) plus the 3803 model 2 controller. They were in surprisingly good condition and he will be buying them. 

He has to arrange for a trucking company to go to the site, pick them up and bring them to his home about 100 miles away. These will join the single 3420 mod 6 he already owns. 


I located some recommended modifications for the unit - then checked to be sure that none were applied on the unit already. I will collect the parts and make the changes over the next couple of days, while I am waiting for the two missing tubes and for a power supply.

I also picked up a Direct Digital Synthesizer VFO unit that will replace the Heath supplied VFO, to provide for much higher frequency stability as well as offering split mode and RIT features.


I received a V-7 (late 1950 era) Vacuum Tube Voltmeter which is able to handle nice high voltages (up to about 20KV) and has 11 Megohm input resistance to avoid dragging down the measured circuits. I inspected it, plugged it in and switched it on. Nothing. No sign of life.

Some quick diagnostic work showed that the defective component was the line cord from the wall socket. One of the prongs on the plug had no connectivity through to the wire as it mated with the circuit board inside. No, this is not a fused plug, it is just a complete and total break somewhere in the cable, with no visible signs of damage to be spotted.

The solution is a new cable, where I will implement a three prong plug and ground the chassis for the protection that was missing in the 50s era design. 

Wednesday, February 8, 2017

Waiting for disk tool test time, meanwhile restoring a Heathkit HW-100 ham transceiver


All is ready for testing but I have to get to Marc's house to do this, which means a couple of days of waiting.


I pulled out a HW-100 that I had bought a few years ago with the quixotic intent of converting it into a kit (desoldering everything, buying new components for all resistors, capacitors etc) and then building it 'as if' I had bought the kit way back when.

Heath HW-100 SSB Transceiver

I actually have many of the replacement components but have changed track. At this point, given how nicely it appears to have been built, I will simply restore the unit to operation and align it carefully. In order to do this, I need to buy the power supply, which was an external unit called an HP-23A/B/C that furnished the 800V, 300V, filament and other power needed for the transceiver itself.

Inside the HW 100

I opened it up and began checking out the vacuum tubes inside. I know that several of them are quite rare and very pricey to buy, such as the 6GW8 audio amplifier tube. Of the 20 tubes that are used in the unit, 18 were in place and astoundingly, every one of them tested good. The only two that are missing are pretty common and reasonably priced tubes - a 6AU6 and a 12AT7.

Testing the vacuum tubes

I found that the electrolytic capacitors inside the unit had been replaced by a previous owner, thus were modern and safe substitutes for the originals that would have been dangerously aged. Of course, I still need to deal with the big electrolytics coming in the HP23 power supply, but this unit is in good shape.

There are three rubber belts which turn variable capacitors to tune and load the RF sections, which had turned to brittle material and already snapped in one case. I have bought a replacement set on Ebay. I am also ordering the two missing tubes so that I can be ready to move forward when the rest of the testing and restoration is complete.

Cracked and broken rubber belts needing replacement

I will go through the unit and check out components, to the maximum possible extent without unsoldering them. The high impedances of vacuum tube components makes this more feasible than with low ohmage solid state components.

Until I have the power supply in hand, restored and checked out, I can't actually check out the operation of any part of this. I will need a dummy load before I can test the transmit functions and should have some sort of an antenna to test reception.