Scout CPO

Share Button
Share Button

3CPO Code Practice Oscillator / Keyer / Trainer

Share Button


Share Button

Mini-Yack Iambic Keyer/Trainer

Share Button

NEW Advanced Iambic Training Mode – Is it a training or is it a challenging game?  The new advanced Iambic training mode will send a call sign which must be properly re-entered with Iambic paddles.   On each successful entry the keyer will advanced in speed by 1WPM and on each wrong entry will decrease by 1WPM.  When training is complete (by pressing the COMMAND button) the keyer will return to the default speed and send what the last speed during the session, the number of calls entered correctly and the total number of calls sent.


Mini-Yack-BoardMini-Yack Mini-Yack-ButtonsMini-Yack Buttons
 Mini-Yack-Connected Mini-Yack-HW-8SMini-Yack Inside an HW-8
Follow THIS LINK for a neat enclosure you can build by VE3FWF.

Comparison of the Mini-Yack to 3CPO

Mini-Yack 3CPO
Side Tone Square Wave Twin-T Shaped Sine
Side Tone Adjustable? Yes Yes
Iambic Trainer Yes Yes
Straight Key Bypass Yes Yes
Memory 2 @ 100 Characters each 2 @ 100 Characters each
Beacon Mode Yes Yes
Keying Positive Only Positive AND Negative Grid
Input Voltage Protection Yes Yes
Key input protection No Yes
Power 2*-15V 8-15V
Buttons Command Only on boardPLAY1, PLAY2, QRS, RESET, and Command External Buttons Available Command, QRS, Play1, Play2, Reset, Power
Factory Hardware Reset Yes Yes
Enclosure Available No Yes

Download the Mini-Yack-Manual

Want to add more or different features?  Hack the Yack!  The YACK code is open source to view and modify. Mini-Yack-Source.

Share Button

Chinese Super RM Rockmite AKA Octopus

Share Button

These kits usually come with a 7.023MHz crystal(s) which are in the EXTRA class bands.  Check out our Crystals page to move these little rigs to the General Tech/Novice frequencies.

Looking for a good English step-by-step guide to assembling your Super RM Rockmite?  Need Troubleshooting information?  Try this:

Super RM Assembly Manual


Follow THIS LINK for a neat enclosure you can build by VE3FWF.



Here is an English version of the software: (30Mb)

Note: I do not personally use the software and have bypassed the Keyer IC but the READMEFIRST.txt does have setup instructions.  After that your on your own…….Best of luck!

Kit Cost: $30

Kit Contents:

Bag of Parts, Printer Circuit Board, Schematic (blurry!), Parts list, Layout page.

No other instructions, however, instructions available on-line.

A zip file of instructions & schematics is available here.

BEWARE OF BAD TOROID CORES and BAD/Low Voltage Capacitors!

Several incidences of Rockmites with low power output (1-3W) were the result of bad T37-2 (red) toroid cores.  In each case good T37-2’s were replaced resulting in 5W output.  Be suspect of cores with a red texture to them.  The good toroids should be smooth in appearance.

There is also reports of some disc capacitors being bad.  Good or bad consider upgrading to 100V or better capacitors in the final amplifer, low pass filter area, and input to the receiver section.  More than 50V can appear in these areas at times which is marginal since these appear to be 50V caps.  Check and replace C4 (33pF),  C8(100pF), C9(470pF), C11(470pF), C12(1000pF), and C25(AKA C26 .1uF).



 ***** 2017-04-04 UPDATE *****


It’s been a while since playing with the Rockmite.  Thanks to Bernie, VE3FWF, I now have this cool LASER cut enclosure to put it in.  As long as we’re adding an enclosure I tricked it out with Pixie Switches, Super VXO, and replaced the keyer chip with a YACK engine.

VE3FWF Enclosure Plans (.SVG, .PDF & .JSON files) along with SVXO instructions are available here:


Rx Filter Pixie switch

& YACK Keyer Engine

Pixie Switch

& Super VXO

New clear enclosure Ready to play!

There are a few caveats with it all together.  The VXO drops out at the extreme end of the band but not enough to worry about.  Along with the VXO the Tx/Rx offset varies from -1KHz at the low end to almost zero at the high end.  This was predicted when playing with Super VXO’s earlier this year.   Retro-fitting the YACK engine was the hardest part as the button logic was entirely different.  The YACK sidetone is also a little on the loud side so maybe a little resistor divider or trimmer might be the next mod to adjust the sidetone volume.

The SuperVXO gives a nice range with the following:

Switch Low High
A 7036.6 7046.2
B 7045.2 7056.2
C 7100.0 7115.2
D 7110.6 7123.2
Tx Offset -1KHz 0Hz

Super VXO Parts:   150pF PolyVaricon, 15pF cShunt, 18uH Toroid w/ 4.7uH Axial.

STM to YACK conversion:

STM YACK Function
1 1 Reset
6 7 SW4 (Now QRS/CMD/PLAY1)
10 4 Ground
12 5 Tx
14 6 Sidetone
15 2 DAH key
16 3 DIT Key
20 8 +5V

The following mods are also needed for YACK:

Add a 4.7K resistor between pins 6 and 20

Pull R27 (reset resistor) – move between pins 1 & 20

Remove R17, R18, & R21 next to the switches

Add a 1K between R18 and R21 (or a 510+560 Ohm) on the switch side (Use the RIGHT pads when looking at the top of the PCB)

Add a 1.8K between R21 and R17 on the switch side (Use the RIGHT pads when looking at the top of the PCB).

Switch SW4 is now the QRS button

Switch SW3 is now the Command button

Switch SW5 is now the Playback-1 button

And now back to our regular scheduled program……….


  • The same issues as the Pixie above sans the color coded inductors as there are none.
  • Schematic diagram was blurry and hard to read, impossible in spots.
  • Kit was missing a 2.2K resistor but ended up with an extra 22K resistor as well as extra .1uF (104) and .01uF (103) disc capacitors.  Kit was also missing the 10uF capacitors but received extra 1uF capacitors.
  • There is a location next to the audio jack for a 100pF capacitor.  This is not marked on the schematic but installing it had no effect.
  • There were NO instructions on what the inductor values are or even how to wind them!  Fortunately KB2HSH’s web site had instructions.  Can also be found on the SUPER-RM Yahoo group.  There are several versions so make sure you download the correct one!
  • The hole for the BNC center connector is slightly out of  alignment and too small.  I ended up breaking the flimsy wire and fixed by placing a small piece of wire in the PCB then splicing into the broken end.
  • In a few areas non-joining pads are spaced too close together.  Watch your soldering and excess lead lengths for shorts.
  • Winding the toroids are not too hard to do.
  • Pick up a strip of break away machine pin headers.  Break into 3 pin sections and remove the middle pin.  Then insert these into the 3 frequency selective crystal areas.  You can then easily change frequencies by swapping crystals.  Note: you may end up with a little more noise by not grounding the 2 input filter crystals.
  • RIT is a trimmer and should outright be a potentiometer w/ knob.  A good upgrade is putting this in an enclosure.
  • NO Mounting holes!  Assumes you are using an enclosure designed to exactly fit the PCB.

Operation and comments:

I’m getting a nice 5W signal from the Tx output.  This is with a cheap 1A 12.0V power supply.  The Rx also works but there are some issues overall:

  • The Program in the micro is not great. it could use a major upgrade!
    • The DIT and DAH inputs do not auto repeat.  Makes the keyer useless!
    • The Sidetone only has a few different tones and sounds funny.
    • Have not even attempted to try the serial port/software.
    • Could not get it into straight key mode!
  • Since I have my own keyer,  and don’t like the tone even in straight key mode, the following modifications were performed:
    • Remove the CPU and MAX232 IC’s – they are not needed.
    • On the CPU Socket Jumper pins 12 (Tx line) and 15 (keyer tip jack) with a piece of wire.
    • Jumper a 4.7K pullup resistor between pins 15 (Tx Line) and 20 (+5V).
    • Snip the lead of resistor R10.  This 470 Pullup is too much for the 3CPO keyer (or jumper D501 on 3CPO instead of the 4.7K pullup and R10 snip).
  • There is a horrid tone on the audio around 7.5-11KHz.
    • The problem is somewhere in the audio circuitry.
    • The problem is there with the SA612 mixer, micro, MAX232, and Osc xtal removed from the circuit.
    • By lifting pin 1 off the op amp and using a simple 470K resistor makes the tone less harsh (still present).  Of course you can’t drive your headphones this way.
    • Looking at the circuit between the 2 op amps I knew something just didn’t feel right.  My guess was no resistor between the 2 sections, effectively giving the second stage a very high gain (Rf/0=infinity) and causing oscillation.  Here is a simple solution:
      • On the bottom of the board, cut the trace between U3 pin 7 and where is leads to capacitors C14&20.
      • Place a 50-200 ohm resistor between U3-7 and C14.
      • I used a small trimmer and found anything below about 25 ohms causes oscillation.  Going above 25 seems to have little effect except to slightly alter the overall noise tone.
  • With the keyer and tone problems solved this is looking like a neat little rock bound QRP rig!
  • FLDigi and HRD receiving software does not work well with this board as it sits.  I believe this has to do with the way the 2N7000 driver is configured.  However, tapping directly off pin 1 of U3 using a DC blocking capacitor (10uF) with no attenuator between radio and computer seemed to work the best.
Super-RM-bypass-Mod Super-RM-Audio-Mod

Additional information & pictures for the 131019 build (Click on each to enlarge)





Y1 and Y3 uses a socket for a Pixie Switch.  Y2 is bypassed with a jumper and C3 is removed.






U4 Bypass

Share Button

QRP Radios

Share Button


Heathkit HW-8 CALF-3CPO-HW-8
Chinese QRP Kits Pixie2
DC Receiver Tuning HW-8-Dial
Share Button

Station Accessories

Share Button
Audio Attenuator Audio-Attenuator-V2-Board-small
Reverse Polarity Protection RPP-in-HW8
Code Practice CALF-3CPO-HW-8
CW Reception CALF-H-Front
Share Button

CW Reception

Share Button
 New – CALF Vs NESCAF Smackdown Comparison.  We compare the NESCAF switched capacitor filter to the CALF active op amp filter.

Read the article:  NESCAFvsCALF.pdf

Audio samples:


I have a hard time concentrating on CW when multiple signals are heard at the same time.  Maybe it’s due to my hearing impairment (Yes, even one eared I do play music and copy CW!).  For me I try to copy everything I hear, high and low, which results in gibberish and missed characters.  Just after creating my HW-8-LM386 audio amplifier, I met up on the HW-8 Yahoo reflector with Glen, KK4LPG, which re-introduced me to op-amp audio filters.  Glen happens to be a wealth of knowledge in op-amp filtering and pulling in weak signals.

The CALF (CW Audio Limiter Filter) project began as a simple way to improve the not-so-great filtering of the Heathkit HW-8.  My first thoughts were to remove the LM3900 IC and replace it using a piggy-back PCB with better filters on it.  The project began growing after looking at some of Glen’s ideas of adding a limiter (and LED display) to the circuit.  So why should a limiter be in the circuit?  There are many good reasons!

  • Several great articles by N7VE discuss filtering and the need to add a limiter circuit to protect your hearing.  This makes sense in saving your ears when you turn your volume way up to hear a weak signal only to go deaf from a large static crash.  Although in this circuit the limiter is at the front of the circuit instead of just before headphones, it still helps.
  • In relation to static crashes above – Take for example trying to listen to an S5 signal with a summer lightening storm a hundred miles away generating S9+10 static crashes.  The storm takes the foreground while your signal takes a back seat.  Now add a limiter in the circuit setting the limit threshold to just above your S5 signal.  Anything over that, such as lightening crashes, is limited (think of it as cutting the positive part of a sine wave in half and squaring off the top).  You still hear the crashes but they are now more like S5 or S6 crashes against your S5 signal… a level playing field!
  • In a weak signal environment you can use a limiter to it’s limits so to speak.  For example take a nice quiet day on 15M (no lightening, just background noise) and you hear a faint signal.  Placing the limiter up until the background noise itself is limited may sometimes make that signal sound stronger.  Maybe it’s a trick of the ears but it really does work!

So now at this point there is a limiter as well as bandpass filters on the protoboard.  The gears started rolling with questions…….

  • How do I add the limit control and LED’s to my HW-8 without drilling into the case?  (no good answers here!)
  • Will it work with my other rigs?  The HW-7?  The HW-16?  The IC-725?  The TS-520?  (yes it does!)
  • Can I even fit this larger circuit in my HW-8 where the LM3900 goes? (if I used SMD devices it could)

With those questions a decision was made to make it external to the radio, polish it up a bit with some input buffering and an output amplifier, and create a PCB for it to keep the project clean looking.  Because I like to tinker with circuits I thought ahead and added a few other goodies to the PCB such as the ability to socket the R/C values to make different types of filters (High, Low, and Band pass!) and to add jumpers to be able to re-arrange the limiter and filters as needed.  The power switch is also an audio bypass allowing the audio signal from the radio to go straight to the speaker/headphones when the power is turned off.


CALF kits and assembled units are available in the store.

How does it perform?  Well first I have a disclaimer…..  Compared with one of my IC-725’s with a built in 250Hz IF filter, there is no comparison.  An IF filter will easily outperform strong signal interference compared to any AF filter, even a DSP one.  That being said, there are still plenty of reasons to use an AF filter.  For many commercial radios cost is a big issue with IF filter prices going from $100-$250+!   Other radio’s may contain an IF but a filter is not available (buy custom Xtals and roll your own!).   Yet for many QRP type radio’s, such as the HW-8, there is no IF to filter.   In these cases an AF filter is better than no filter at all!  Personally I prefer the sound quality of CALF over my IC-725 narrow IF fiter.

That being said, it does a pretty good job.   Along with the bandpass filtering, the limiter really does help the signal.  Both during static crashes as well as trying to grab that signal in the noise floor.    The filter was designed with a 700Hz center frequency.  Although there are 6 stages of filtering, only 5 are actively selected with the switch.  3dB Bandwidths were measured as follows:

 700Hz   Measured as 689Hz with a 5db/octave roll off - Great for AM/SWL listening
350Hz    Measured as 360Hz with a 13db/octave roll off  - Great for SSB listening
175Hz    Measured as 159Hz with a 34db/octave roll off
125Hz    Measured as 117Hz with a 41db/octave roll off
<100Hz   Measured as  79Hz with a 41db/octave roll off - A little too narrow to tune with the HW-8 unless tuning with a lower position first.  However once tuned worked great!
calf-proto1-100mv-manualMeasured response of CALF


Here are a few audio samples.  In each sample is a few seconds without filtering then the same sample played again with the limiter/filter:

40M on a bad day

40M with Buzzing sound



Noise, QRM, and QSB


Example of working with High and Low pass circuits.  Section 1 is High pass, Section 2 is Low Pass.  The resulting plot shows each filter along with combining them in series.  Note in the picture we used a jumper to bypass stage 1 for a stage 2 only measurement.  In this test the center frequency was 1200Hz.






 Want to design filters for the CALF?  Check out the TI FilterPro cheat sheet page.

Sometimes I get a request for a specific bandwidth or center frequency, so at the bottom of this page I will post values for requested settings.



CALF-V2-MANUAL – including parts list and schematics

Interested in experimenting or need a nice filter?  Kits are available and can be found in the store.



CALF now has an enclosure but here are a few pictures getting from point A to point B…….

CALF-H-FrontThe final CALF coral



CALF & 3CPO on Batteries

CALF  & 3CPO will run on batteries (for example 9V or a bank of AA/AAA’s).  It’s recommended to be closer to 12V or to use at least 6-AA’s.  Connect the battery positive (red) to the positive power terminal on the CALF/3CPO.  Connect the battery negative (black) to the middle unused power jack terminal.  This connection is normally shorted to ground when nothing is plugged in and will open (disconnecting the battery) when external power is plugged in.

For the battery I usually find the 9V & AA/AAA battery holder wires to be real thin & fragile.  If you use the terminal blocks then you may need to solder a small piece of wire (or a clipping from a part lead) to the battery leads so the terminal block has enough to clamp down on it.  You may also want to use the mounting hole next to the power block along with a small cable tie to secure the battery wires down and keep them from breaking.


Values for specific CALF filters

Refer to the schematic for node numbers – Values in red indiciates parts NOT in the original CALF filters.

Filter Type Fc BW Q N1 N2 N3 N4 N5 CALF-600Hz

600Hz bode plot

using (sub) values

1 BP 600 600 1 15K .01uF .01uF 56K 47K
2 BP 600 300 2 47K .01uF .01uF 100K 6.8K
3 BP 600 200 3 82K .01uF .01uF 150K (sub 200K)
4 BP 600 120 5 56K .022uF .022uF 120K 1.2K
5 BP 600 100 6 56K .022uF .022uF 150K (sub 470K || 200K)
6 BP 600 75 8 82K .022uF .022uF 180K (sub 200K)
Same as above except optimized using values existing in the CALF filters (red)
1 BP 550 550 1 27K .01uF .01uF 56K 27K
2 BP 550 275 2 62K .01uF .01uF 120K 8.2K
3 BP 550 185 3 91K .01uF .01uF 180K 5.6K
4 BP 550 110 5 68K .022uF .022uF 130K 1.3K
5 BP 550 90 6 82K .022uF .022uF 160K 1.1K
6 BP 550 70 7.8 100K .022uF .022uF 200K 820

500Hz Center Values – optimized using values existing in the CALF filters (red)

Filter Type Fc Bw Q N1 N2 N3 N4 N5
1 BP 500 500 1 33K .01uF .01uF 68K 33K
2 BP 500 250 2 56K .01uF .01uF 120K 8.2K
3 BP 500 167 3 82K .01uF .01uF 180K 5.6K
4 BP 500 100 5 82K .022uF .022uF 150K 1.5K
5 BP 500 83 6 82K .022uF .022uF 180K 1.2K
6 BP 500 63 8 120K .022uF .022uF 220K 820

500Hz Center Values – Using 0.1 uF capacitors

Filter Type Fc Bw Q N1 N2 N3 N4 N5
 1 BP 500  500 1  3.3K  .1uF  .1uF  6.8K  3.3K
 2  BP 500  250 2  5.6K .1uF .1uF 12K 820
 3  BP 500  167 3  8.2K .1uF .1uF 18K 560
 4  BP 500  100 5  18K .1uF .1uF 33K 330
 5  BP 500  83 6  18K .1uF .1uF 39K 270
 6  BP 500 63 8  22K .1uF .1uF  47K  180

450Hz Center Values – optimized using values existing in the CALF filters (red)

Filter Type Fc Bw Q N1 N2 N3 N4 N5
1 BP 450 450 1 33K .01uF .01uF 68K 33K
2 BP 450 225 2 82K .01uF .01uF 150K 10K
3 BP 450 150 3 120K .01uF .01uF 220K 6.8K
4 BP 450 90 5 82K .022uF .022uF 150K 1.5K
5 BP 450 75 6 82K .022uF .022uF 180K 1.2K
6 BP 450 56 8 150K .022uF .022uF 270K 1K


Calculated 450Hz Center Values – Using .1uF 5% capacitors and 5% resistors (KC9ON tested Preference)

1% resistors are preferred.  Note the last 3 stages are all Q=5 causing steeper skirts.

Filter Type Fc BW Q N1 N2 N3 N4 N5    CALF-450Hz-Preferred
1 BP 450 450 1 3.3K .1uF .1uF 6.8K 3.3K
2 BP 450 225 2 8.2K .1uF .1uF 16K 820
3 BP 450 150 3 10K .1uF .1uF 22K 560
4 BP 450 90 5 18K .1uF .1uF 33K 330
5 BP 450 90 5 18K .1uF .1uF 33K 330
6 BP 450 90 5 18K .1uF .1uF 33K 330
Share Button

Code Practice

Share Button

A few short years ago I decided to start playing with QRP and purchased an HW-8. Since this is a CW only radio I brushed up on my code using software called RufzXP until I was confident to have that first CW QSO in over 20 years. Although reading code at 15WPM went well I sure had QLF (Sending with the left foot) missing or hitting extra dah’s and dit’s with an IAMBIC keyer. It was even worse on a straight key! So in searching for a training tool I ran across the DK3LJ YACK keyer project and discovered it’s ability to be used to help train my IAMBIC fist. The only down side was the typical piezo buzzer / square wave oscillator…. that buzz buzz sound I can’t stand.

Last fall (2013) someone on the QRP-L mailing list mentioned wanting a nice sounding code practice oscillator.   That is when I decided to combine a twin-T oscillator into the YACK keyer. The result is a nice little keyer IC with well shaped CW audio.  A few extra goodies were added like reverse power protection, protection of the keyer inputs (every accidentally plug your -65V grid key rig into the key jack and watch the magic smoke?), and of course the ability to use either positive keying (for modern day equipment) or negative grid keying (for my TS-520 & 530). To add icing on the cake the controls are mounted to a second PC board to make chassis mounting easier and cleaner.

3CPO kits are available in the store!

Don’t need the horsepower of 3CPO?  Check out our Mini-Yack.

Manual and schematics (v2): 3CPO-Manual.

CALF Coral Enclosures with overlays are available in the store!

Modified YACK Source Code: 3CPOv2-Source

3CPO View-small

CALF & 3CPO on Batteries

CALF  & 3CPO will run on batteries (for example 9V or a bank of AA/AAA’s).  It’s recommended to be closer to 12V or to use at least 6-AA’s.  Connect the battery positive (red) to the positive power terminal on the CALF/3CPO.  Connect the battery negative (black) to the middle unused power jack terminal.  This connection is normally shorted to ground when nothing is plugged in and will open (disconnecting the battery) when external power is plugged in.

For the battery I usually find the 9V & AA/AAA battery holder wires to be real thin & fragile.  If you use the terminal blocks then you may need to solder a small piece of wire (or a clipping from a part lead) to the battery leads so the terminal block has enough to clamp down on it.  You may also want to use the mounting hole next to the power block along with a small cable tie to secure the battery wires down and keep them from breaking.

3CPO Alternate XMIT output test

The resistance checks in the KEYING circuit tests may vary depending on the DMM or VTVM being used.  Sometimes the DMM will only work for this test in the continuity setting which provide slightly more current, other DMM’s may need to be taken out of AUTO ranging mode.  In any case where the meter method does not work an alternate method of testing is available as follows:

Parts Needed:

  • an LED – any size of flavor
  • A resistor – The value is not critical but something in the range of 2.2 to 10K.  4.7K is a good medium value.
  • An isolated power source, such as a 9V battery.

Wire the circuit below.  Connecting P1 and P2 should make the LED light.

  • Connect P1 (Battery +) to the XMIT KEY+ and P2 (resistor free end) to XMIT GND.  The LED should light as you key down.
  • Negative grid block keying test – Swap the XMIT KEY+/GND connections.  The LED should still light during key down.


Share Button


Share Button

Fun with an HW-8

Video Release – Our first video featuring the HW-8 along with the 3CPO code oscillator/keyer and CALF audio filter looking for the CBLA (Color Bust Liberation Army) on 3.579545MHz.


 QSK Mods for the HW-8 and HW-7:

Here are a few mods found in the HW-8 handbook.  They are also highly recommended for the T/R switch below.


HW-8-tr-C92Change C92 to 1uF – Located between the relay and delay trimmer. R27aChange R27 to 470K – Located under the Audio output jack. HW-8-TR-CapAdd a .47uF capacitor between the Base and Collector of Q1

And a mod to fix keying issues for the HW-7 – A great find from Rich KR7W via Frank K0JQZ.

It appears the HW-8 and HW-7 use the same relay driver circuit.  The only difference on the HW-7 is R14 which is 10x too big (should be 4.7K).  Also on the HW-7 C19 is 25uF but on the HW-8 it is a mere 10uF.  The simple resistor fix seems good enough where I have not bothered trying to change C19 as well.    However……. When installing the TR board in the HW-8 it is recommended to change C92 to 1uF per the HW-8 handbook QSK circuit so if it’s still not fast enough QSK then you may want to consider lowering C19 down to 10uF or even 1uF.



T/R Switching:

T/R Switch kits are available in the store.

2015-03-17 – Update – 4 boards now built.  1 stand alone and the other 3 in my HW-7 & HW-8’s.  All seem to be operating great.  Pretty cool hearing between the characters.  Kits are made up and ready.  Just need some time to update the web site now.

2015-03-05 – Update – The PCB’s are in and 2 boards built.  First board was an outboard T/R switch.  Was able to handle 100W when it started smoking!  Not bad for a 10-20W design!!  Switch was working so good board #2 went into the HW-8.  So far it’s working as planned.  Rx isolation in transmit mode is > 40dB and the receive loss is < 3db (less than 1/2 an S-unit).


HW-8-TR Board HW-8-TR Installed in HW-8 HW-8-TR installed in the HW-7
HW-8-TR-Side HW-8-TR-Installed2 HW-7-TR2


2015-01-25 – The annoying click-click-click of the HW-8 relay finally provoked me into replacing it with a more modern circuit.  To make things more interesting I also plan on using this design in a future rig with a 10W output.

The initial breadboard test was from the HW-8 handbook using diode switching.  But there were a few issues.  One was the high amount of current consumption in both transmit and receive.  Another was the isolation was mediocre but probably acceptable.  There was also 3-6dB losses involved in both receive and transmit.  So on to the next design…….

The next version was a MOSFET one based somewhat on the N5ESE design and also used in the project.  The breadboard did not seem to have the isolation as in the diode design.  I have also read that the 2N7000 MOSFETs can be destroyed under high SWR/abuse conditions.   The OpenQRP project used a ZVNL110A which helps that problem.  Unfortunately only 2N7000’s were in the junk box.   So they say the third time’s the charm………

A hybrid of the 2 designs by placing one of the diode sections from the first design in front of the MOSFET’s of the second.  Isolation seemed better (>50dB) with no Tx loss and <3dB (1/2 an S-unit) Rx loss, low current consumption over the first design, and protection of the MOSFETs,  A good compromise.


For the low power of the HW-8 you can probably jumper out the diode T/R section leaving only the MOSFET section.  However, the design is for other uses later on, such as a 10W amplifier.


 Audio Boards:

Audio Replacement kits are available in the store.

Back in 2013 ago I purchased a Heathkit HW-8 e-clunker from Ebad. It took some work to get it running, and I still have an issue with microphonics, but I am having a blast with it. One of my caveats is the original audio board can’t directly drive a speaker or 8 ohm headphone. So I decided to give it some more oomph and replace it with a basic LM386 audio amp and PCB designed to fit in it’s place. It’s nice to hear it off a speaker now.


Download the assembly manual and schematic: HW-8-LM386-V4-Manual.

Download the assembly manual and Schematic for the new V3 version:  HW-8-LM386-V3-assembly.

HW-7 Owners

This board also works on a HW-7 with a few modifications:

  1. Disconnect the “Phones” wire going from the PCB to the phones jack.
  2. Connect a wire from the Phones jack GROUND to the GND hole on the audio board.
  3. Connect a wire from the power switch to the +12V hole on the audio board
  4. Connect a wire from the OUT hole on the audio board to the phone jack tip.
  5. On the audio board place a 1K resistor across the IN hole to one of the GND holes (do not solder the IN hole yet!)
  6. Place one end of a 33K resistor also in the IN hole.  Solder.
  7. Connect the free “Phones” wire to the other side of the 33K resistor.
  8. Set the audio board pot to half way.  Turn on your rig and enjoy!



MC1496 Adapter board

MC1496 Adapter kits are available in the store.

If you smoked your MC1496G IC1 you will find it’s getting harder to find now days and the prices can be anywhere from $16-27 EACH on Ebay!  The 14 pin DIP is available but also getting harder to find and is pretty large for the circuit.  I found the SMD version of the part is small, CHEAP, and fits well using a circular carrier board.  I have a few spares available in my store area.  The assembly manual can be found here.





Improve the HW-8 Audio (quick and easy mod!)

Thank you Georges, F6DFZ, for this great mod!

This modification used to remove some of the raspy sound by reducing the gain of the op amp from infinite to a more reasonable value. This in turn removes the clipping of all audio into square waves.

a) Remove the wire from the center terminal of the Wide/Narrow switch.

b) Clip one end of the 4.7K resistor so only about 1/2″ of lead remains.

c) Solder the clipped end to the center terminal of the Wide/Narrow switch.

d) Clip the other end of the 4.7K resistor so only 1/2″ lead remains.

e) Solder the removed wire to the end of the resistor.

f) Make sure the resistor and soldered wire do not touch the chassis or other components. Optional – Tape or shrink wrap the resistor and connection.



Repairing loose slugs (Bad drifting and micro-phonics)

I noticed my VFO tuning coil, L9, was extremely loose and very touchy when trying to adjust it.  I found a post on the QRP-L group which NN6CW suggested wrapping Teflon plumbers tape around the slug to make it tighter.  I ended up putting about 2 turns of thin tape on.  The coil adjusted very smoothly!  In addition the micro-phonics is now eliminated and my drift is less than 1KHz over the first hour and around 100Hz drift up to 9 hours later!



Shorting pre-selector capacitor lets the smoke out of Q1

I had a problem with one of my HW-8’s where turning the pre-selector capacitor fully clockwise, C301B section would short out and kill the front end MPF102 transistor. I just could not figure out where it was shorting and how to repair it.  So instead of spending big bucks for a new cap, or trying my hardest not to put the pre-selector fully clockwise, there is a simple fix.  Cut the wire going from the pre-selector C301B to point A on the circuit board.  Insert a .1uF capacitor in line.  Don’t forget the shrink tubing or electrical tape.  Problem solved!



Improving Receiver Sensitivity (as suggested by F6DFZ)

A Simple way to improve receiver sensitivity is to replace the front end transistor Q1 with an MPF102.



Add an S-Meter to the HW-8 (WB7OVJ via QST Hints and Kinks)

Only 4 simple parts and a hunk of wire to add a simple S-meter.  I didn’t have a 12K resistor handy so used a 10K, which seems to work.


 Keeping the Hw-7 and HW-8 from loosing the magic smoke!

This is a simple mod to prevent these rigs from going up in smoke when you accidentally swap the power supply lines.  All you need is a rectifier diode and piece of shrink tubing.  I use 1N4007’s but any of the 1N400x’s will work.  Just cut the wire between the power connect + lead and the power switch.  Slide on the shrink tubing.  Solder in the diode, banded end (cathode) going toward the switch.  Then slide the shrink tubing over the diode and exposed leads and shrink.  You will loose .7V (diode drop) going to the board but you probably won’t miss it and prevent an accident from happening.

UPDATE – I no longer use 1N400x series diodes for blocking.  These have been replaced with a reverse polarity protection board (P-MOSFET).

Reverse Polarity protection kits are available in the store.

More HW-8 MODS and tips:

If you know of a good tip or mod for the HW-8, please let me know and I will post it on this page!


HW-9 Mapping

Recently I had to repair and re-align an HW-9 for a friend.  While in there I decided to save O-scope screen shots along the way.  This is a nice companion to the HW-9 schematic when poking around and the folders are laid out in the order of the alignment instructions!

You can find it right here: hw-9-mapping



Share Button


Share Button

Want to know more about Amateur Radio?  The ARRL is a great place to start.

Ready to get your license?  We recommend, and instruct with, the W5YI books.

Read the story about 3CPO – a code practice oscillator/keyer kit based on my QLF (sending with the left foot).


Interested in HF Digital modes?  View the ARRL Michigan HF Digital Modes presentation.


A great and way to start listening to HF Digital is with the WA8LMF audio attenuator project!


Have fun with a Heathkit HW-8.

Share Button