Re: The QRP Mindset: Who is responsible for the QRP signal?


Nice thread. There is something magical about putting out a signal and getting a reply, regardless of the power. But doing so with a minimalist setup is very gratifying. The person on the other end is usually amazed, unless they too are running QRP. 

I operate QRP POTA quite a bit and it's always fun when a DX station does all they can to pull me out of the noise and work the elusive county. I try my best to put out quality CW using a respectable rig and the best antenna I can fabricate (usually my 29' vertical wire with a 9:1 unun). My goal is to rapidly deploy a station using battery power and 5 watts or less. Theirs is to get me in the logs. We both get what we want, which is all that really matters.

Re: The QRP Mindset: Who is responsible for the QRP signal?

Ron Carr

On Tue, Oct 12, 2021 at 01:14 PM, Pat wrote:
signal, the recipient may have to fine tune the signal, filter out some noise, and may have to turn an antenna in a little different direction.  The recipient may have to reduce interference from other competing noise sources or ask others to stop talking for a minute or two.  A ham may have to put on headphones or amplify the signal, switch antennas, or use other signal-enhancing steps to hear the QRP station.  The entire time the receiving station is going through their signal-refinement repertoire, the QRP station just has to keep sending their signal. 
Boy! If he is doing all that work, why does he send back a 599 signal report?

Re: Battery power for the QDX? #power

Curt wb8yyy

Consider how and where one will use this qdx. In the shack the dissipation of an adequate 9 volt linear regulator is just a little more heat, on battery well not so nice. One could install the regulator network in an adapter unit for in shack use, retaining 9v operation for field use. Or an excuse to build some kind of low noise 9v supply. 


Re: U3S w/ 10W HF Linear 8m project, some pics and questions

Al Holt

Oops, of course I forgot my question(s)!

I'd like to try the amp out with a 19V power supply. Do I need to rebias the amp to 125mA per side on its finals when using this higher voltage supply?

Any other problems running the amp at voltages higher than 12?


U3S w/ 10W HF Linear 8m project, some pics and questions

Al Holt

Hello all,
My 8m U3S w/ 10W HF Linear amplifier project is coming along and I'm pleased with the outcome so far. First, the pics:

Some details:
  • It's all boxed up in old Radio Shack "T-Sin" Deluxe Metal Enclosure (270-229) that I got off of eBay a few years ago. A little oversized, but plenty of room to grow. 
  • The low pass filter was built with parts from LPF 10m & 6m kits sitting on a I/O breakout board. I used SimSmith software to determine the best values for the toroids. I bridged the kit caps with 10pF which seemed to help improve return loss at 40.6 MHz. I'm learning the software package, and it's VERY nice!
  • I'm using 9V for the U3S' PA voltage. With a 12.6V external power supply I'm getting about 6W out of the 10W amp during transmit. I haven't measured what the U3S output is with this configuration. I plan to do some testing at 30m to get some kind of benchmark. 
  • The U3S needed some modification getting the signal out to the amp. Capacitive coupling is needed, so C5 in U3S was retained; I'm using the pads marked "RF-I" (RF input the onboard LPF connections), because externally mounting the LPF is necessary with it connected on the 10W amplifier's output. 
  • I was able to cobble together a nice 9V/5V linear PS for the U3S out of a 'Proto board' power supply kit from Nightfire Electronics in Ocala, FL. Their PCB supports both fixed and variable regulators and it was easy to break the connection between the voltage I/O's on each end of the board to have both regulators installed and usable. The heat sink chunk of aluminum bar is from an old computer power supply, I think. 
  • Speaking of heatsinks, the one on the U3S' BS170 PA is one I came up with made from a sheet "hobby" copper. Probably more of a belt-and-suspenders solution.
  • Keying the 10W amp is done with an 'active low' output from the U3S. But, its "Key" output is 'active high.' I found I could use the relay control signal for the U3S's onboard 'Relay 0' to achieve this goal. It's too bad the "PPT Delay" feature of the U3S does not support PTT (a feature I rely on) they are mutually exclusive at the moment. Hint hint, Hans! So the leading edge of my transmissions may be a little flakey.
Now, all I need is my Experimental licence; fingers crossed! I plan to operate WSPR mainly and perhaps QRSS, who knows, U3S supports a wide range of digital modes!

I hope this all might whet your appetite to get involved in 8m experimentation with hopes that FCC gives US hams privileges in this bridge band between 10 and 6 meters. Also, with introduction of QDX there'll be more interest in the 10W HF Linear kit and its use.

Comments and questions alway help!  73!!


Re: The QRP Mindset: Who is responsible for the QRP signal?

John Kirby


Very interesting thread :>)

Thought  I  would share one of my QRPp , full QSK, ransceivers

72 73 74

Re: QCX Mini Constant Tone

Mont Pierce KM6WT

On Thu, Oct 14, 2021 at 12:55 AM, Ken Thomas - N6KEN wrote:
ent to test my key I tuned up another radio and I can hear a tone being generated by the Mini.
This is normal for any Direct Conversion receiver.  The VFO is on the same frequency as being received.

If your other radio is a superheterodyne radio, you will hear it's I.F. oscillator too if you tune the mini or another receiver to the I.F. frequency of the superhet radio.  You just don't normally notice it as it's on a different frequency then being received.

An old trick for hearing CW/SSB on AM only shortwave radios was to build an oscillator on a small perf board on the AM radio's I.F. frequency (usually 455KHz). Then, putting it next to the AM radio it would act as a BFO.


Re: QCX Mini Constant Tone

Ken Thomas - N6KEN

Makes sense. Thanks for the response Alan!

Re: QCX Mini Constant Tone

Alan G4ZFQ

On 14/10/2021 08:55, Ken Thomas - N6KEN wrote:
I tuned up another radio and I can hear a tone being generated by the Mini. I unplugged the headphones and my key and it still produces a tone on the frequency it's set at.
You will always hear this. It may seem loud on a local receiver but all it is is a small leakage from the local oscillator.

You will probably hear it with no antenna on the RX. Then you can make a crude estimate of how strong it is compared with the TX output.

73 Alan G4ZFQ

QCX Mini Constant Tone

Ken Thomas - N6KEN

I bought my QCX Mini when it was first released as a assembled unit by Hans. I'm just now getting ready to use it this weekend and I decided to power it up and play with it. Went to test my key I tuned up another radio and I can hear a tone being generated by the Mini. I unplugged the headphones and my key and it still produces a tone on the frequency it's set at. 

Any suggestions on why this is happening? Anything to worry about?

Re: The QRP Mindset: Who is responsible for the QRP signal?


On Wed, Oct 13, 2021 at 11:55 AM, Jim Manley wrote:
I’ll bet that, with one of the locally-available barbed-wire fences, I’d be able to outperform the majority of people who only throw money at the problem by buying “real” antennas.
I didn't mean to offend you, or anyone else in Montana, Jim.  I don't doubt your capabilities, either.

By "real antenna," I just meant one intended to radiate RF signals, commercial or home brew, and definitely not high-falutin'.  This isn't about money.  My favorite antenna to use portable is a NorCal doublet, and that cost me the cost of some ribbon cable.  I am proud that I built a well-documented, effective radiator that delivers decent signal reports.  Is tuning up the neighbor's barbed-wire fence, or whatever a good scrounger can scrounge up, something to be proud of?  Is that your best work?  

The point of my post was that we should operate quality equipment (not spark-gap transmitters) with a properly-deployed antenna, operating as effectively as we can and without interfering with others, and I think that's especially true for the QRP operator.  


Re: How about a QCX-4? Anyone?

Dave Bauer

The QDX has low pass filters on the TX output side, and band pass filters on the RX input side, per the manual. Bandpass filters are critically needed on the RX side, to ensure detected signals are band specific.


Re: Battery power for the QDX? #power

Tony McUmber

I'm not replying to myself to be seeming to generate traffic -- I just couldn't think of another way to add to my previous remarks.

Anyway, Gregg got me to thinking about this power dilemma, and thinking always seem to confuse me in many ways.

I understand about using a 9V in-line regulator to provide 9V from a 12V source.  It seems, though, that it may still be necessary to find a way to switch between a 3-turn secondary and a 2-turn secondary because the 3- is optimized for 9V feed and the 2- is optimized for 12V feed to get the best performance.  Is this not so?  If so, then is it practical (and practicable) to find a way to shoehorn in an extra couple of turns of wire and a switch?  Am I losing my way here?  It seems to me that one must decide at the outset (because it is the first component to be installed!) how to wire the secondary, thus committing forever to an unchangeable input voltage.  I'm pretty sure that this would not make an untenable difference in one's life, whichever potential is chosen, but it seems like it would be neat to be able to adjust the power supply and then make a counter-adjustment to the rig for optimum performance under a wider range of conditions.

Maybe such a mechanical beast would be incompatible in any practical way with today's sub-atomic circuitry.  A kid can dream, no?

Re: Battery power for the QDX? #power

Tony McUmber

Well, Gregg, I absolutely had not thought of that.  I will look for such an item.  
<$1.00, he says (plus shipping, of course).
Still, for me, the thought of a tiny double-throw retro knife switch on the back panel has its appeal.

...and some troll earlier was complaining about all the stupid questions!  Hah!


Tony N0BPA

Re: New kit: QDX - QRP Labs Digital Transceiver

G8DQX list


there's a good Wikipedia article on MFSK at <>. Most MFSK systems send a single tone at a time, typically through a Gaussian filter to soften the transitions and conserve occupied bandwidth.

There are systems which use multiple tones at once, DTMF signalling, for example.

HTH, 73, Stay Safe,

Robin, G8DQX

On 13/10/2021 23:44, Gregg Myers wrote:
Perhaps I misunderstand MFSK but I thought  this does not necessarily mean these modes are concurrent tone modes (which the QDX cannot do). THROB is a concurrent tone mode. But I didn't think Olivia and DominoEX send more than one tone at a time, in which case it should be possible for the QDX to send/decode right? Or am I mistaken in what MFSK means? From what I read MFSK only means that multiple tones are employed in the modulation, not necessarily multiple tones at the same time. Right or no?


Re: Battery power for the QDX? #power

Daniel Holmes

My thought was to use a USB C PD battery pack with a negotiator to put out 12V. The other port will run the rPi. 

(Now to figure out if my order ever went through—got a confirm from the cart, but no email. Gonna wait a few days to make sure the charge goes through before I worry about it’s)


. Please pardon any mispelings or errors.

On Oct 13, 2021, at 6:06 PM, Fred Spinner <fred.spinner@...> wrote:

I have several different types of switcher boards, and I will try to figure out the best one by trying them. Putting a LDO after a switcher cleans up low frequency noise better than LC filters, but not the higher frequencies. Unfortunately the feedback loop in most LDOs is pretty low frequency. 

Hans' earlier comment of you have to try it with switching regulators is pretty spot on.  You sometimes can move the frequency of whatever interference you get from a module to somewhere in frequency you don't care about.   You can can "dither" the switching clock to spread out noise peaks. You can LC filter and LDO filter together.  You can Faraday shield the supply.

Newer higher frequency devices are getting pretty good, but still this would be a great problem to solve. 

The 9V voltage range is good to protect the AMS1117 LDO regulators, as that is a safer input voltage range for them.  But this also means that the 3.3v and 5v lines already have a LDO on them.  So one on 9V might not make as much sense.

The 9V range is bad for the Lithium chemistries directly but it's a compatible voltage for NiMH or even alkaline. 

I am going to try a 9v buck converter with my 13.6V nominal output LiFePO4 first and if its meh I am thinking a pack of NiMHs next.  Cheap enough, small enough, and in a decent mAh range for the QDX.  It's pretty miserly on receive current for what it is... 

One wonders if tapping off of the laptop battery with a buck converter is doable too... But that could get ugly... 

Fred W0FMS 

On Wed, Oct 13, 2021, 4:46 PM Brent Minchey <brentm@...> wrote:

My home and portable operating is all battery powered, either 3S Li-ion or lead-acid. That means a supply voltage maybe hitting 13V all the way down to 9V and usually hovering in the 11-12.5V range. The nominal 12V, 3:2 winding, option for the QDX wants 13V to put out 5 watts. At 12V, it's down to 4 watts, at 9V, about 2.1! I'll stick with the 3:3 winding.

The quiet option for bringing that down to 9 volts looks to be an LDO regulator, maybe a 2A, 
BA90DD0T. That's throwing away ~2–4 watts on transmit most of the time. A buck, or buck boost DC -DC converter will be a lot more efficient, but the noise could negate a bunch of Hans's excellent work. Shop for a clean one? Build one with shielding and a 100µH inductor or some more sophisticated filter? There's a trick in mobile phone gear of setting a switching regulator at the drop out voltage ahead of an LDO and using it as a filter. A trade-off in efficiency, and I couldn't find any info about how clean that will be at HF. 

Are there preferred switching frequencies for the buck converter to avoid in-band harmonics? Has someone already figured out a way easier solution to this? 

Thanks, –Brent WT4U

Re: Qcx


By the way, what exactly do you mean when you say "disable the transmitter" ? Are you talking about practice mode?
And what exactly do you mean when you say "keys up"? Do you mean that you actually get RF output? 
Julian, N4JO.

Re: How about a QCX-4? Anyone?

Fred Spinner

QDX has low pass filters on TX with tuned harmonic traps for all but 30m. (20m is a little different than the other filters too allowing that bank to be used for both 20m and 30m.) These are switched by the 1N4007s acting as PIN diodes.  Bandpass filters on RX, switched in through the second FST3253. 

In general what you would expect in any correctly designed transceiver, but with less parts. 

Clever design for filtering too, not just the way TX is done.  

Way better than FCC requirements on TX with the very accurate phased push pull TX already suppressing even harmonics quite well. 

Fred W0FMS 

On Wed, Oct 13, 2021, 6:17 PM Ronald Taylor <wa7gil@...> wrote:
Dave, they really are lowpass filters. Read Hans’ discussion about them. They might be a little different from what we are used to seeing. They have harmonic traps as part of their design. But they’re definitely lowpass. 

73… Ron

On Wed, Oct 13, 2021 at 18:00 David Birnbaum <dbirnbau@...> wrote:
I just checked in the manual for the QDX and the filters are NOT low-pass but bandpass.


Re: Battery power for the QDX? #power

Fred Spinner

So one know-it-all comment on syncing the clock in the field.  You can get a GPS USB dongle for about $13 from the evil A to Z empire.  A quite decent one.  Now for that price and physical case, you are missing the PPS output, so I would not run my network's NTP server off of it.  But, even so, running a clock sync over a consumer grade GPS without PPS will still get you a very small fraction of a second accuracy sync.   Way better than the 2 second accuracy needed for FT8.

If you rig up one with PPS you will likely be better than network derived NTP sync.  But that is a whole another project.  

You can then stop the program or daemon in Linux, remove the dongle and you will be accurate enough for the WSJT-X modes for the outing and maybe for a week with a normal PC RTC.   And that is a cheap solution for a Raspberry Pi not having a built in RTC too.

The dongles also emulate a serial port on USB and tend to just work. 

Fred W0FMS 

On Wed, Oct 13, 2021, 7:16 PM Gregg Myers <gregg.w7grm@...> wrote:
Hi Tony,

For me, adding a 9v regulator inline to the power feed is way simpler that trying to switch a tap on the transformer. A 9v linear regulator or an LDO regulator costs less than 1$ after all. This is the simplest way, IMO.


On Wed, Oct 13, 2021 at 8:09 PM Tony McUmber <1tonymc@...> wrote:
I would be interested to hear from anyone about the feasibility of winding a 3-turn secondary and a 2-turn secondary and then switching between them depending on the available input voltage.  Surely this never could work, especially with a plain-old mechanical switch.  Yes?  No? Take a hike?

Tony N0BPA

Re: Qcx


Hi Gary,
Well that should certainly be enough...
So two things have to happen in order to transmit: 
1. DC power has to be supplied by Q6 through L4 to the output stage Q1-Q3, and
2. The VFO frequency has to be gated through the the output stage at the output of IC3A pin 3.

For (1) to happen, the "KEY_OUT" signal (processor pin 16) has to be at logic 1 (close to 5V), which will turn on Q4 and Q6; and
For (2) to happen, the "RX" signal (processor pin 14) has to be at logic 0 (close to 0V), which will create logic 1 on the "TX" signal (IC3 pin 11, also IC3 pin 5).

I would invite you to see if you have a logic 1 at IC3 pin 5, and the gate (center pin) of Q4.

If IC3 pin 5 is logic 1, then IC3 pin 3 will appear, on a DC voltmeter, to be something around 2.5V, and would show a square wave of VCO frequency on a scope.

If "KEY_OUT" is high, you should expect to see a DV voltage at the Q6 side of L4 close to your supply voltage.  In Rx mode, if I remember correctly, the voltage at L4 will be undefined, as both Q6 and all three output transistors will be switched off, leaving that point at very high impedance, thus floating.

Julian, N4JO.

6921 - 6940 of 78554