Heathkit DX-100 Restoration

History of this DX-100

About 30 years ago I was given a Heath DX-100. It belonged to a fellow ham, now SK, and it has resided in my mini-museum for all those years. I had it on my list of restorations but because the front panel was in pretty poor condition I was not that kean on restoring it. I put out the word for a replacement front panel from a junker as the gentleman from Electric Radio classified no longer makes replacement front panels. A local ham had a junker and I was able to get a not perfect but much better panel from that DX-100. So this winter, 2022, I started the restoration.

This turned out to be a big change inside the transmitter but the front panel remains stock and except for performance, which is considerably better, one would not know anything was done to it. I am very pleased with the end result but before I talk about that I will show and describe what I have done in text, photos, and schematics.

One important warning! There is HIGH VOLTAGE present in this transmitter. Always make sure circuits are discharged before servicing. I make it a habit of pulling the plug whenever I am working on the transmitter otherwise there is always the possibility you could come in contact with energized parts.

Quick description of changes

Here is a list of most of the changes made to my DX-100. A more detailed description of changes will follow below.

The transmitter was fairly clean inside. The copper although slighly pitted in places was in remarkably good shape. The kit building was another story. Some sections like the RF areas around the 6146's and output coils were very well done but others were not anywhere near my standards. I found many poorly soldered connections and a couple of mistakes. I really wonder if this rig every saw much on the air time. Fortunatley I made so many changes that it was fairly easy to correct the problems as I went along. Whomever built it did not follow the correct wiring harness positioning but that was mostly corrected. The underside of the chassis in some areas now hardly resembles a stock DX-100. A dozen or more terminal strips were added to accommodate the change in circuity. Quite a few holes were drilled in the chassis for these added components.

Heath used the classic dual fuse two conductor power plug with gray lamp cord. Not only is this unsafe but it does not meet today's electrical code. In 1955 grounded outlets or polarized plugs were a thing of the future. It was a 50/50 channce when you plugged it in that a fuse would be on the hot side of the line, so they fused both sides of the line. The problem is if the neutral side fuse opened you would still have AC to the device. Not being grounded and no concept of GFI's was yet another issue. Certainly there were many more possibilities of electrical shock with this early equipment. I do my restoration in the basement and code is that basement outlets must be on GFI's. This is also an important troubleshooting tool because if the GFI trips you know you have a primary circuit problem in the device you are working on. GFI's trip if ground and neutral on the feed end are shorted which of course they should not be. Also be very careful when working on tube equipment that involves high voltages like the DX-100. The high voltage can reach close to 900 volts. I always turn things off and pull the plug when I am working on a restoration and not running tests needing the unit to be powered.

Photos before Restoration

The following photos show the rather sorry state of the circuitry in the DX-100 I before the start of resotration. Fortunately it wasn't beyond repair.

Click on any image below for a larger view, click again for a closer look where your mouse is positioned

This photo shows the back areas of the LV power supply and the below chassis 6146 and tank components. Restoration had already started as the LV choke has been removed and a rear panel fuse and new power cord are in place.
Broader view of the LV supply and modulator section. A portion of the RF section at the bottom.
Closeup of the line input and bias circuitry before restoration. Only new line cord and fuse installed. LV choke and air coil inductors on the line removed.
View showing entire LV and modulator section.
Unfortunately a little out of focus but you can see the messay HV section. Starting to remove the rectifier sockets.
The low level modulator section with the copper shield removed. Lots of changes were made here as you will see in a later photo.

Now the serious restoration!

For the restoration I suggest removing the front panel. It is much easier to maneuver with it off. It is also necessary to take it off to service the VFO which is recommended. One caution is that when you have the transmitter upside down without the front panel the VFO tube and adjustments would be damaged unless you had a way of supporting the front side. I used a piece of angle aluminum bolted to the side of the transmitter. Make sure it is long enough to ensure that the VFO tube or anything else vulnerable are protected. Here is a photo showing the support,

The support is bolted to the modulator side of the transmitter. It should be long enough so the VFO tube has clearance when the transmitter is upside down. In this photo the front panel is on but it would be off when using the support.

The DX-100 has several under chassis sections. I will start with the LV, Bias, Modulator section. As mentioned before the LV choke which Heath placed under the chassis was removed to allow easier access to the terminal strips used by the AC input and the LV transformer. That corner of the chassis is very crowded. When this kit was built the chassis came in part, the top panel and the bottom frame. The kit was partially built with the top and frame not connected making it easier to get to things. Unfortunately once the kit is finally assembled it becomes very hard to access some areas. So I pondered once removing the choke how I would get it back in as access to the bottom holes was very limited. Fortunately this problem was solved when i realized that I could mount the choke on the top of the chassis after removing the unneeded rectifier socket and can electrolytic.

View showing the LV rectifiers and copper board covering holes from removed LV rectifier socket and electrolytic can. Upper right shows modulator screen MOSFET 250V regulator. Rear panel shows added modulator bias pot and RCA jack for modulated B+ scope sample.
Another view of this area showing LV rectifiers, Modulator screen regulator, and rear panel bias pot and modulated B+ RCA jack.
Another view of this area showing power input, added rear panel fuse and updated bias supply. Heatshrink placed on transformer and choke leads for protection.
Wide view of LV and bias supplies and modulator driver transformer and 1625 modulator sockets on right. Large 1 Meg blue resistor is for modulated B+ sampling.
Top chassis view showing LV choke remounted where the LV rectifier tube and electrolytic can formerly were. A double sided copper clad board was cut to cover the holes and mounted under the chassis.

The low voltage power supply was tested with no loads connected except a 20K bleeder resistor. Without load it was about 285V. Perhaps I should explain why it is a good idea to change the LV supply from capacitor to choke input. The original capacitor input supply that Heath design gave, by their 1950's voltage chart, about +370 volts. The was at a nominal 115-117 VAC line. Today most line voltage is in the low 120V range so in most cases just plugging the rig in today would give even higher voltages. The +370 voltage exceeds even the CCS ratings of the tubes it supplies in the DX-100. There is no reason especially with the scarcity of tubes to abuse them like that. I my mod I also used the 5V winding formerly supplying the HV rectifiers to buck the primary. My line voltage is very steady at about 121V +- about a volt, subtracting 5V put me right in the ball park of the original design. This gave almost exactly 6.3 volts or 12.6V to the filaments.

Modulator Section

Now onto the low level modulator section. This is located in the front corner of the LV section of the chassis and has a copper plate shield separating it. Part of the circuitry is mounted to this copper shield so you have to remove the screws holding the terminal strips attached to it and then remove the shield. I completely re-arranged the circuitry so nothing would be connected to this shield adding a terminal strip on the chassis side and on the front panel. The front panel terminal strip uses a flat-head recessed screw that does not interfere with the front panel. Since I was also adding PTT I removed the single conductor mic plug and replaced it with a two terminal Amphenol jack. The new jack is larger in diameter than the old one and the way these jacks are designed they need to go through the front panel and chassis BEFORE they are soldered in. I didn't like that so I permanently mounted the connector in the chassis after enlarging the old hole and punched the front panel (15/16" round punch) to accommodate the outside diameter of the jack. This allows you to solder the connector in and test it without having to remove it again to install the front panel.

All coupling and bypass capacitors were replaced in the speech amplifier and driver sections. Values are shown in the schematic link below. The mic input circuit grid leak was increased to accommodate a high impedance microphone like the D-104. The PTT line to the PTT relay in the HV section used an unused wire in the harness. The copper shield was reinstalled turned 180 degrees and in a slightly different position. With the 1625 modulators tubes removed a scope was placed across the output of the driver transformer and a microphone connected to test the circuit changes.

Because of the compactness of this area it is a little hard to see all the changes, see the schematic link for details. The terminal strip on the side chassis was formerly on the copper plated shield and a terminal strip was added to the front side between the level pot and the mic connector using a recessed flat head screw. This allows a good and close ground to the pot and mic and PTT bypass caps.
Although capacitor values were greatly increased the actual size of today's components is considerably smaller making it easier to update.

VFO Restoration and Modification

By far the biggest complaint you will hear about the DX-100 is VFO drift and I was determined to attack that problem. First of all, as mentioned, Heath "cooked" the miniature tubes in the DX-100 with higher than required voltages and certainly higher than tube data maximums, the VFO was no exception. They had +370 volts on the 6AU6 plate and +150 on the screen. It is no wonder the VFO was drifting. So I used a MOSFET to regulate the plate to +160 volts and the OA2 gas regulator was replaced with an OB2 bringing the screen down to +105 volts. The 6AU6 was replaced with a 6AH6 as recommended by Heath although for some reason they never actually made the part change even in the VF-1 which is almost exactly the same circuit. I also re-soldered many of the connections in the VFO and using a large gun soldered grounds to the copper plated chassis in several places. I also rewired all the connections to the VFO. The 6.3 filament line had been fried probably by a short at one time on the pilot light circuit which feeds through the VFO. I ran a separate line outside the VFO to the dial and meter pilot lights and also changed them to LED's using replacement LED bayonet style cool white bulbs. They are available on the Internet.

Disassembled VFO. It is not hard to remove once the front panel is off. First remove the U shaped cover held with sheet metal screws and lugs to the bottom of the chassis, then the remaining assembly is held by the switch shaft which you need to disconnect under the chassis. You can see the burnt black filament wire
Another view showing the burnt wire fused to the other wires.
Another example of the poor wiring. This is why for any good restoration you need to take the VFO apart.
Another view showing original VFO compartment wiring
All wires out of the VFO were replaced with teflon and encased in heat shrink. VFO Ground connections soldered to copper clad chassis in several places.
Another view showing re-wiring and chassis soldering. All connections in the VFO were gone over.
Completed VFO restoration under test before reinstalling.

The VFO in the DX-100 operates with 160 meter output for the 160 and 80 meter bands and 40 meter output for the 40 through 10 meter bands. It also has an 11 meter band which I ignored in the restoration. I found a problem maybe caused by the reduction in voltages where the 160 meter output of the VFO rolled off considerably at 80 meters causing the grid drive to be low on that band. It turns out they used a 22uh choke in the plate circuit of the VFO which rolls of the 160 meter output considerably. This was not a problem on 160 meters but manifested itself in creation of the second harmonic when tuned to 80 meters. I changed the choke to a 750uh that I had in the junk box and it solved the problem. Interestingly the VF-1 actually switches this choke to different values with the band switch.

Replacing and testing the new plate choke in the VFO.
This shows the preparation of the MOSFET's for installation. The wiring looking at top of the device is gate (blue), drain (red), source (orange). The device I used is totally insulated and can be bolted directly to the chassis with a dab of heatsink compound. Note that a MOSFET is a high gain device and it can oscillate. The leads from it should be kept as short as possible. If you notice oscillation adding a 1K resistor to the grid lead close to the device should clean it up.

The final results of the VFO were outstanding. Very little VFO drift, on the order of a hundred cycles or less from startup and after a half hour. After hearing so many bad things about the DX-100 VFO I was very pleased with the results. I also have two VF-1's that I intend to look at stabilizing with the same mods. The results might even be more significant on the VF-1 as its B+ input voltage is from whatever the rig it is attached to feeds it. A MOSFET regulator in the VF-1 with a change of the VR tube and VFO tube might make it a much more stable VFO. The gas regulator tube could also easily be replaced by a MOSFET regulator in both the DX-100 and the VF-1 but I decided to keep the tube as I have pelnty of OB2's.

HV Supply, Antenna Relay, and Switching Circuits

The original HV supply in the DX-100 used dual 5R4's in a choke input circuit with two series connect 125uf 450V electrolytics with a 30K 20W center tapped bleeder. This equates to almost 25 watts of power through the bleeder. The 5R4 sockets were removed as well as the electrolytic cans and bleeder. The bleeder was under rated for the voltage and they used the center tap to feed the 1625 modulator screens. This put as much as 400 volts on the modulator screens where there should have been around 250 volts. Yet another Heath design flaw or maybe just cost savings.

So as you will see in the photos and schematic I completely changed this. I used eight 1000V 3A diodes, four per leg replacing the 5R4's, two 250uf 500V can electrolytics with two 30K 10W equalizing bleeders. This greatly reduced the dissipation in the bleeders. The 1625 screens were fed from a MOSFET 250V regulator that is in turn fed from the LV supply and switched by an available contact on the antenna relay. I also incorporated a concept Collins often used in their designs of resonating the HV choke. This essentially makes the choke a higher impedance (parallel resonant) at bleeder load reducing the unloaded HV. This greatly improves the regulation and also keeps the unloaded HV below 900 volts. The HV is switched by a solid-state relay in parallel with the antenna relay. Voltage for the relays is supplied by an added 12V transformer, bridge rectifier, and filter capacitor.

View of the rear of the HV section showing the antenna relay (top) and solid-state relay (bottom). The BNC jack on the rear panel is the receiver connection. The 8 pin octal accessory connector has been rewired as per the schematic.
View showing the high voltage diode strings, filter capacitors where the HV 5R4 rectifiers were, and bleeders. At top right is the choke resonating capacitor and lower right the 12V relay transformer.
Close view of the antenna relay and two MOSFET switches, one keying the cathodes, the other supplying a closed circuit on PTT for the accessory socket.
Closeup view of the accessory socket. MOSFET to the left of the socket gives an open circuit on PTT for the accessory socket. The resistor back in the corner is the .1 ohm Plate meter shunt. The resistors just out of view at the far left are additional resistors in parallel to calibrate the plate current reading.
The HV choke resonating capacitor is shown at top center. It is a glass capacitor with screw studs on the ends so it is wrapped in heatshrink. New 1 Meg HV meter resistor is to its right. This was moved from the LV section in the original Heath design.
12V PTT transformer, bridge rectifier and filter capacitor at right. DPDT HV switch is only using one pole which controls the PTT circuit,

The plate current shunt was missing in my DX-100. Another reason why I question if it ever worked. Without this resistor you would have no high voltage. Heath schematics describe this resistor as both a .91 and a .1 ohm resistor. I had a nice .1 ohm 1% resistor but in checking with a digital voltmeter the reading was a little high. Something like 220Ma on the DX-100 meter for 194Ma actual as read on the digital meter. Two 1 ohm resistors in parallel got it close so I left them in the circuit. While I am mentioning the meter it is a good idea to parallel the meter with back to back diodes for protection. You also cannot compare readings on the digital meter and the front panel meter at the same time as the digital meter has a higher series resistance and the front panel meter would peg. Take a reading on the front panel meter and record it. Switch the meter to another position, then insert the digital meter and take a reading on it. You then need to parallel resistors to get the same plate current reading on the front panel meter with the digital meter out of the circuit.

Top chassis view showing high voltage filter caps mounted where the rectifier tubes formerly were. A copper clad plate is installed at the front where they were formerly mounted.

It would have been interesting to measure the line power on an unmodified DX-100 and my final restoration. I would guess it would be significantly less. All the the rectifiers filament load is gone, LV is significantly reduced and the HV bleeder value is doubled. Unfortunately my DX-100 was not in operational condition when I started and for fear of damaging something I would never have turned it on in that state. The 900V series rating of the original HV electrolytics was very close to what the unmodified voltage would have been without load such as when HV is on in CW key up. The 500V, 1000V total of the new capacitors gives a good margin. I would not put two 450V capacitors, 900 total, in a DX-100 especially if your line voltage is high. Heath measured voltages closely follows an input line voltage of 115-117VAC. So for example if your line voltage was 125VAC the HV could be elevated by 80 volts or more.

Rear panel addition of new power cord, fuse, and modulator bias pot.
Additionof rear panel BNC receiver antenna shown next to accessory connector.

Exciter RF section

Closeup of the drive pot MOSFET replacement.
Overall view of the RF section.

Final RF section

Few changes were made to the final RF section. I did not modify the loading circuit to a single loading capacitor instead of the switched caps and fine loading control. The original circuit worked fine and as long as you don't ever switch the coarse loading when keyed it should be OK. A hole was drilled to accommodate two coax runs, TX output and Antenna, over to the high voltage section where the antenna relay was mounted.

RF section showing antenna output and RF pickup for scope to an RCA connector.
View shows hole to high voltage section for connections to the antenna relay.

The 6146 screen dropping circuit was changed to have both modulated and unmodulated voltage. The calculation can be done using the following -

6146 Sceen Resistors Calculator
B+ Voltage: V Created by Don, AC2RS
Screen Voltage: V
Total Screen Current: mA
C1 Value: pF

R1 Value: _____Ω
R2 Value: _____Ω
C2 Value: _____pF

Screen Voltage MUST be 200V. Screen Current for 2 tubes should total 22mA. Control Grid current should be set to 6mA for 2 tubes. These values are appropriate for a plate B+ voltage of about 650V. They come direct from the RCA Transmitting Tube Manual.

NOTE: Under no circumstances should the Control Grid Current exceed 3mA per tube or 6mA total. The RCA Tube Manual specifies absolute maximum Control Grid Current of 4mA per tube or 8mA total. Tube Life will be dramatically shortened if you exceed these parameters. It is not necessary for good linearity to challenge these grid current ratings. Ignore misguided information on the web which suggests you do so.

This technique should work on any transmitter which employs 6146 finals. It improves AM positive modulation peaks and has no adverse effect on CW operation. For example, it works on the DX-100 (which Dan McGorill originally developed it for), the B&W 5100, and the Johnson Valiant (don't forget to compensate for use of three tubes). The Heathkit Apache TX-1 will require more adaptation, due to its unique tune up circuit in the screen of the finals.

In my case I use a 30K/10W and 40K/10W in series for R2, a 45K/10W for R1 and a .001/2000V for C2 based on B+ being 800 volts, screen current 22ma and C1 of .01 (2-.005)) This is an easy mod and greatly improves the modulation linearity.

6146 sub-chassis showing new screen resistors and C2.
Top view of RF section
Whenever I do a restoration/modification I like to picture all the parts I removed. This is most of them although a few might be missing.

Project Epilogue

This was a very rewarding project. Like most of my restoration projects I had no idea where it would end up and how much I would need to do. The modifications progressed as I went along, taking it one section at a time. I probably spent more time on this project than the many other I have done but I learned a lot doing it and I am not sorry I picked the DX100. It is a great old boat anchor. The final results are in the great on the air audio reports I have received. I would be glad to answer any questions you may have. My email is good on QRZ. I hope to work you on the air sometime.

73, Doug, WA3DSP

Heath DX100 Links

This page last updated 4/18/2022

© 2022 - WA3DSP