Saturday, July 29, 2017

Sticker Shock

CW keys cost too much, or do they?

I read various forums and blogs for the same reason you might be reading this one.  I'm always interested in learning more or trying to answer someone's question in a forum, etc. etc.

A recent forum post went something like... 
"CW straight key prices are so obscenely inflated that I lost interest in learning CW..."

I'll admit that I've been shocked at the prices of some CW keys but to say that because some CW straight key prices are high that would make me lose interest in learning seems a bit non-sequitur.  I'm not going to debate the logic/illogic of not learning CW due to the high cost of straight keys but I would like to explore the current cost of straight keys.  So are CW keys really that expensive?  Let's have a look...

Morse Express offers a good selection of keys and I've had good experience purchasing from them.  At the time of this writing they offer an AMECO straight key for $12.50, the Hi-Mound model 705 for $50 and a Nye Viking Speed-X for $85.  Prices can certainly go up from there... A Begali Spark (available from Begali) is approximately $150 and more custom keys and materials can go up and up and up.

There may be less expensive new straight keys than the AMECO $12.50 model but let's start there.

So from a 1958 radio shack advertisement we see:

1958 Radio Shack ad

The 99 cent "Skillman 'jr' Speed Key" appears to be equivalent to the AMECO AM-K1 advertised at MorseX site.  What is 99c in today's money... about $8.50  So yes by that measure the absolute cheapest key you could buy from Radio shack in 1958 would cost only 75% of what you'd pay today.   What about the next model up, the "Deluxe High Speed Brass Key" is probably equivalent to a Nye Viking Speed-X.  So $2.95 versus $85... Hmmm inflation says the Nye Viking should cost $25 based on that comparison.  Maybe the Speed-X is better made than the 1958 Radio Shack "Deluxe High Speed Brass Key" but probably not 3-4 times better.

My goodness, straight key manufacturers are gouging us aren't they?

So it would seem straight key prices are significantly higher now than in the 1950s  Why is that?

Could it be the cost of manufacturing... 
As quantity of production increases from Q to Q2, the average cost of each unit decreases from C to C1. LRAC is the long run average cost (Source Wikipedia)

There was a far higher demand for keys in the past than at present

So according to this 1958 FCC Report there were 180,738 licensed amateur radio operators in the USA in 1958.  ALL of those operators were required to both learn morse code and were originally restricted to only operate CW until a license upgrade occurred.  Even after that I'd argue (and I'd like to hear from operators from that era) that CW continued to be a far more common mode of communication for hams than phone modes at the time.  

So guess what?  All these operators were purchasing morse code keys.  I don't think the electronic keyer had come into common use at the time so they would be buying straight keys or bugs of some type (not discounting cooties but hey let's admit using a cootie takes a special kind of strange).

Fast forward to the PRESENT:  
  • Morse code is not required to obtain a FCC amateur radio license.
  • There are ~744000 licensed hams but ~373000 are Technician class; so although they might be exercising CW in their limited band portions I think we'd all agree that's highly unlikely.  So that leaves us with 371,000 operators that could be operating CW.  Seems like a big number.
  • My personal anecdotal experience from visiting local Amateur radio clubs tells me that less than 1 in 20 licenced operators regularly use CW and of those the majority use electronic keyers with paddles or operate with a keyboard, rather than using a mechanical key.
  • The SKCC is a club/association for straight key and mechanical key operators.  From what I can tell it is the most popular CW related club at presently.  As of this writing it has 17385 members.  Since all these members are using mechanical keys that's a good gauge of active straight key operators at present.

Current market for straight keys

Ok making a semi-sketchy educated guess out of this combination of my assumption filled math moderated by anecdotal experience and one actual real number; I'm guessing there are 18,000 active mechanical key operators at present in 2017 compared to 180,738 in 1958 (not taking into account other potential buyers straight keys outside of amateur radio).

There is likely less than 10% of the 1958 market for straight keys now.  Add to this that's it's likely easier to obtain a used key today from an auction or ham swap sites that further dilutes sales potential of new straight keys.

So with greater than 90% market erosion we look back up at the cost of manufacturing chart and see why straight keys cost so much now compared to when you were an aspiring novice back in the 1950s.

What is a CW key worth?

I've mentioned this in other articles but
Of every item in your shack, what are you going to actually touch as much as your CW key?  
You will physically interact more with you CW key than any piece of equipment you own.  It's kinda like comfortable shoes.  Sure you can buy a pair of $10 flip flops or you could have $3,000 custom made Italian shoes.  Somewhere in between lies something that's both functional, comfortable and durable.  

The same is true for CW keys.  You could buy the $12.50 AMECO plastic key or a $540 Begali sideswiper.  Both work, but what you can be happy with likely lies somewhere in-between.  Folks will happily spend big bucks for a ham radio microphone to achieve that "full fidelity sound" (in ahem 2.8kHz bandwidth) but they balk at spending over $100 for a CW key.  

Seems to me that keys aren't getting the love they deserve.

So before you decide that you won't learn CW until the price of Morse Code Straight Keys comes down from today's obscene prices, consider what it is you're actually buying and why they just might cost more today than when you were a lad and all automobiles had to be hand cranked.

That's all for now...

So lower your power and raise your expectations.

Richard, AA4OO

Tuesday, April 11, 2017


Hard fought QSO with the 1Watter

I haven't used the 1Watter in a while so I thought I'd pull it off the shelf and hook it up for a QSO.  The 40m band was in pretty good shape but the frequencies covered by my XTAL controlled 1Watter are pretty limited and trying to find an open frequency was tough.

I received a response to my call from N1WHT up in CT, but the station was a little off frequency and the 1Watter doesn't have a RIT so I had to change frequency to bring him into my passband. Then another much stronger station came in right beside us which overwhelmed the 1Watter AGC when I sent my final response.  I've tried a number of things to correct the AGC issue on my radio and haven't had success.  I've just learned to deal with having the audio from my sidetone drop out when sending in such circumstances.

This was a rather hard fought QSO.  I used my Elecraft AF-1 audio filter kit to help eliminate the stronger station on receive but the passband was so narrow that I had to turn it off to hear some of my sidetone when sending.

So I made a contact but it wasn't the most elegant or pleasant QSO of the evening.  In any event, it's communication so soldier on.

Here's a video of the QRPP adventure...

That's all for now

So lower your power and raise your expectations

Richard AA4OO

Tuesday, April 4, 2017

Space, the final frontier in CW

Give your CW a "break"

I was listening to CW stations tonight through my SDR having extended QSOs on 40m between 7015 and 7035.  There seemed to be a good number of stations sending between 24 wpm and 28 wpm, maybe a bit faster. The signals were strong and I had good signal to noise on most of the conversations.  24-28wpm is generally faster than I'm able to comfortably copy but I use these listening sessions to improve my head copy skills.

As I listened I realized that I could copy some stations nearly 100% while with others sending at about the same speed I just could not head copy no matter how hard I concentrated.  One particular QSO had two stations operating at the same speed and about the same SNR.  I'd guess they were operating about 25wpm.  One of the stations I copied easily without thinking much about it while the other I just caught a word here or there.  The difference was their space between words.  Not necessarily space between characters but between the words.  I started paying closer attention to the station I could copy clearly and I could count about a one second pause between each of his words.  The other station was running one word into the next.

Then I started switching back and forth between a number of QSOs and I recognized that my comprehension was very dependent on word spacing.  I could even copy bug operators who had their DITS flying over 30wpm with 20wpm spaces clearly as long as they paused between words.

Silence is golden

The artist Sting is a famous bass player.  I am also a bass player, certainly not a famous one, but I follow bass players and enjoy their different styles.  

In an interview in 2000 Sting said:
For me, the sound is only half of music - the space between the notes is also vitally important... 
Is space the final frontier in being able to copy CW?  Why is it that some CW operators, and I'll venture to say most CW operators don't put adequate space between their words.  CW is not a fast mode of communication so why not give each word the importance it deserves?  Why be in a rush?

I'm going to strive to put more space between my words in my next QSOs.  It may just make copy for the other operator a bit easier and rather than them bailing on you after an exchange or two you may just chat away for an hour because they enjoy the solid comprehension of every word you send.  If you work me and don't hear me putting an adequate pause between my words, call me out on it.

That's all for now

So lower your power, and    pause   before   your   next   word,  then raise your expectations

Richard AA4OO

Sunday, March 26, 2017

More power to you with the Elecraft PXA100

When 5 watts is not enough

Gasp!  My blog is called "HamRadioQRP"... so what sort of heresy would lead me to post a topic on using more than 5 watts...
Will I have to turn in my QRP card?
I bit the expensive bullet and bought a used PXA100 amp for my KX3 this past week. There, I've got it off my chest.  I've come out of the closet. Go ahead and unfriend me and unsubscribe from my blog...

Normal Elecraft KX3 -- Notice the RF scale goes from 1w up to 12 watts

Elecraft KX3 with a PXA100 -- note the RF scale now goes up to 110 watts

OK, for those still reading...

This is my first time into the bottom of a solar cycle running QRP

If you've been operating for the past 6 months I'm sure you've noticed the band conditions have been getting pretty poor, and while I continue to try and make contacts and have ragchews using QRP power levels sometimes I feel bad for the other fella who is not QRP struggling to copy me.

I enjoy the challenge of making contacts at QRP levels but I also enjoy exchanging more than my call and an RST.  When the other station gives me a 559 or worse, or the QSB is getting bad I want to give them a few more dB to work with.  

I've written before that QRP operators can be compared to certain persons with dietary habits.  Sometimes vegetarians will eat dairy but if your vegan then you're abstain from anything that didn't grow from a root.  Well I'm more of a QRP vegetarian who will occasionally put a slice of bacon on my sandwich if I think it will improve the taste.  I'm certainly no vegan when it comes to operating QRP alone.   I've used my TenTec Eagle and TenTec C21 to go beyond 5w when the mood struck and in the case of SSB I've rarely operated at QRP levels with success so there I'm usually running well over QRP levels but I'd pretty much stuck to QRP for CW until recently.

Elecraft PXA100

Elecraft makes some very elegant equipment.  The KX3 by itself, running on internal batteries and with a built-in auto-tuner is an amazingly portable and full featured QRP rig.  With the newest version of firmware (MCU 2.70 / DSP 1.49, 2-27-2017) the KX3 will produce up to 15 watts of output on 80m, 40m and 20m when used with an external 13.8v power source, but if you want more than that you need an external amplifier.

I've been reading lots about the Hardrock 50 external amp and had even bid on a couple of Ten Tec 405 QRP amplifiers (which I lost) recently.  The Hardrock 50 was compelling for the price because it could be built with a full QSK option, whereas the old TenTec 405 did not have a full break in option but both would get my output up to 50 watts which is about 3-5 dB more output than the KX3 on it's own. 

But my normal operating mode is to begin at QRP and if that's insufficient I turn up my power.  In the case of most external amplifiers I would be over driving them if I kept the KX3 at 5 watts input to the amp.  It would require dropping my KX3 input power to the amp and fiddle with the output until I was at the power level I wanted.

On the other hand, the Elecraft PXA100 is fully integrated with the KX3 and I can simply turn up the power knob of the KX3 and when the power exceeds what the KX3 is capable of on its own the PXA100 electronics take care of adjusting the KX3's input into the amp to give me a smooth power output up to 100 watts.  Additionally, when using the internal tuner option with PXA100 I can use two antenna outputs from the PXA100 and switch between them using the ANT button on the KX3 while the PXA100 tuner maintains independent tune memories for each antenna. It is a very slick package but an expensive one.  Purchasing a used KX3 and PXA100 together puts you well over the $2k mark which you can certainly find more full featured 100w desk radios in that price.  So for price conscious hams this is not a sensible option.  I won't argue that this makes sense from a financial aspect, but I did it anyway.  Given the vibrant used market for ham equipment I wouldn't pay new prices for these but it still comes out to be a pricey desk rig even when purchased used.

OK, back to the PXA100.  One absolutely brilliant aspect of the PXA100 when using the internal tuner option is that its tuner seems every bit as good as the KX3's internal tuner.  I think it will tune a spoon on 160m (admittedly I haven't tried that).  It has no trouble tuning my ladder line fed attic Doublet which has some serious impedance mismatches on certain bands, so being able to quickly switch between my attic doublet and my outdoor Windom has been a real boon that I don't think other desk rigs can match with their internal tuners.  I could be wrong, but Elecraft's tuners and super fast matching algorithms are pretty much legendary in the industry.

Another nice thing about the KX3/PXA100 combo is that full break-in QSK is maintained and there are no fans. My shack stays very quiet (I hate fans).  The amp has a massive finned heat sink on top that while getting quite warm to the touch has not become overheated even on extended transmissions at 100w into non-resonant loads.  I still haven't delved into digital modes but I'm confident that while the KX3 struggles to maintain its cool when used for full cycle modes at 5w and above, when used with the PXA100 this combo avoids that problem because the KX3 is only running between 2-3w intput into the amp and can dissipate its heat at that level with no problem.

So for the past week I've been using this combo and I'm quite pleased.

Here is a video that I made today in a contact on 30m where I started at 5w and moved up to 25w and switched antennas during the QSO.

So lower your power and raise your expectations, but when expectations fail, use an amp

Richard AA4OO

Sunday, March 5, 2017

Non-resonant versus resonant antenna for receive only use

Well... that's not what I expected

I've been playing around with my SDRPlay and have been thrilled with its sensitivity when connected to my attic antenna.  My attic antenna is a 68 foot doublet that weaves around my attic and is fed with 450 ohm ladder line.  I use a 1:1 balun connected to the ladder line and a short run of coax with the RF-front-end-protection  to feed my SDR.  

My Doublet antenna is not resonant on any ham band and requires matching to bring the impedance in line with something my transceiver wants to transmit into.

My outdoor antenna is an 80m OCF Dipole (aka Windom) in the 1/3 configuration, meaning that the short side of the dipole is 1/3 the length of the long side.  Its been a good performer for me and is resonant on 80m, 40m, 20m, 17m, 12m and 10m .  While using the indoor Doublet antenna as a receive antenna for the SDR (with front-end protection and a relay to switch out the SDR during transmit) I use the Windom connected to my transceiver.

This configuration has worked well and the SDR feeding HDSDR as a panadapter has been a real boon to finding stations quickly as well as being a wideband audio feed to CWSkimmer.

HDSDR fed by my SDRPlay

For grins I wondered how much better the Windom would be as a receive antenna for the SDR on the resonant Ham bands... Well it was between 6dB to 10dB down on reception on most bands when used with my SDR.  I tested this by alternately connecting the Windom and the Doublet to the SDR for repeated tests.

This doesn't make sense to me.  I assumed that a resonant antenna (and one that's much longer) would be a better antenna than the Doublet for the SDR but it's not working that way in my case.  

When I use the Doublet with the transceiver it is used with an autotuner.  In general my A/B tests using my transceiver between the Doublet and the Windom are about 3dB to 6dB weaker for the Doublet on receive but I often get better RBN signal reports using the Doublet by a few dB on transmit.

I'm really perplexed.  I expected the Windom to be a better receive antenna for the SDR.

I have a lot more to learn about antennas.

That's all for now...

So lower your power and raise your expectations

Richard, AA4OO

Friday, February 24, 2017

Spruce up a basic transceiver with an SDR

Using SDRPlay as a panadapter for the Ten-Tec Eagle

SDRPlay produces a nice middle of the line Software Defined Radio (SDR)

SDRs are a lot of fun to play with simply as broad band receivers.  I've used the regenerative receiver I built for some shortwave listening but that isn't even in the same universe of capability as a good software defined receiver.  SDRPlay is significantly better than what you'll find with simple RTL-SDR dongles but costs around $150 as compared to $25 for the dongles.  Since there are other reviews of SDRPlay out there so I won't go into details other than to mention that it offers continuous coverage from 10kHz to 2 GHz, has a 12-bit ADC rather than the 8-bit ADCs in dongles and provides up to 10 MHz bandwidth at a time.

One of the ADCs on the SDRPlay board

Using SDRPlay as a panadapter

In addition to using SDRPlay as an incredible broadband receiver, I wanted to use it as a panadapter for my Ten-Tec Eagle.  My Eagle is an early run model that does not provide an IF output or the circuitry to drive a panadapter.  It is possible to splice into the Low Pass filter line of the Eagle with some minor surgery to provide receive antenna input to the SDR but if I did that the SDR could only be used when the radio is turned on and its receive bandwidth is limited to the Low Pass filter switched in the Eagle at the time.  I didn't want that limitation or to have to keep switching the receiver's antenna inputs when I switched back and forth.  So I decided to keep the receive antenna external from the Eagle. 

But to do that I had to protect the front end of the SDR from the transmitter or it would be overloaded and destroyed.

WARNING: Input to the SDRPlay cannot exceed 0dBm without causing damage so be very careful when using your SDR with a transmitter

Front end protection

There are a number of designs for protecting receivers from overload used in multi-station events.  A common circuit is a bulb with two reversed diodes to shunt off extra energy.

I considered building one but when I priced out the parts and a metal enclosure I decided to just buy one.  I found one made by a ham in Slovenia (Aleš S59MA) that was reasonably priced and available on a famous auction site.

This allows me to use a receive antenna connected to the SDR with this protector in-line while I transmit on my outdoor antenna.

That works but the panadapter display really lights up when I transmit and while the specs say everything should be safe I'll feel better switching out the receive antenna on transmit.  I ordered a kit from OK1RP.  The kit requires some extra parts (2x2N2222 transistors and 4x4k7 resistors) to complete that most hams already have but my junk box is pretty shallow so I ordered the 2N2222 transistors from DigiKey.  The part specs are listed on OK1RP's page.

Simple receiver switching circuit can be built in 30 minutes

Note that this relay defaults to the receiver output being switched off when there's no power supplied to the relay.  The relays are so quiet I have to hold it up to my ear to hear it operate.

A bit of extra armor

The SDRPlay RSP1 comes in an unshielded plastic enclosure but I wanted to keep noise and near-field RF at bay.  So I added layer of protection from near field RF by putting it in an aluminum enclosure.  

Original plastic enclosure to the right

$20 aluminum enclosure

Making it work as a panadapter

Bring in the CAT - Computer Aided Transceiver 
SDRPlay feeds IF/IQ to software defined receivers like HDSDR. That provides a way to view and listen to all that can be received by the SDR, not just ham bands but any signal from 10kHz up to 2GHz.  But to use it as a panadapter for my Eagle I needed a way to control both HDSDR and my transceiver.

OmniRig is CAT control software that simplifies software control of receivers and transceivers.  It communicates through COM ports.  The Eagle has a USB to serial interface that uses a hardware COM port but HDSDR is software with no physical COM port.  HDSDR can be CAT controlled as well using either a Virtual COM port or DDE interface.  

So I created a Virtual serial port using VSPE software to create a virtual serial port "pair".  The first COM port is considered the INPUT from HDSDR.  The second pair of the COM ports is the OUTPUT to be used by the listening application (OmniRig).  OmniRig needs to know the "type" of "radio" it's talking to.  HDSDR uses the Kenwood CAT protocol so I chose Kenwood when configuring OmniRig.

HDSDR also provides the ability to synchronize the CAT traffic between two OmniRig connected radios.  This ability to sync two rigs is the key to making this work.  It syncs with the transceiver so that changes made in HDSR are sent via CAT control to the Eagle and vice-versa.

SDRs have huge bandwidth

SDRPlay can send up 10MHz of bandwidth to the SDR software receiver.  That much bandwidth isn't terribly useful other than at the macro scale of seeing if there are signals.  More typically I set the IF bandwidth to 2MHz which easily covers an entire ham band and use the zoom control in HDSR to narrow it even further.  What's further, I can feed IF audio from what's being received by the SDR to programs such as CWSkimmer.  CWSkimmer has an interface for the SoftRock SDR to receive up to 192kHz.  The issue was how to get the audio output from HDSDR to the CWSkimmer application.

To accomplish that I use VB-Audio virtual audio cable to route 192kHz bandwidth of audio from HDSDR to CWSkimmer's SoftRock IF/IQ interface. CWSkimmer also has an OmniRig interface so that is configured to read the second virtual COM port pair from HDSDR.  Thus I can control the Eagle from either HDSDR or from CWSkimmer and vice-versa, changing the VFO on the Eagle commands HDSDR and thus CWSkimmer to follow it.

It's a thing of beauty.

Video Demonstration

That's all for now

So lower your power and raise your expectations

Richard AA4OO

Tuesday, February 14, 2017

All keyed up

Sometimes I can't decide which CW key to use

As I've been posting videos I've received requests to display the translation of the CW as it's received and sent.  I have tried narrating a bit in some of the videos but that doesn't cut it.  So I did this little video test to see how FLDIGI would translate the code and capture that translation from my computer screen then display it in the video at the bottom.

It worked reasonably well for this short message but I think when I have my next video QSO I'll need to capture the entire FLDIGI translation window and include it as a video insert.  This was just a test for me to see how to use iMovie to overlay a video within a video.

Hopefully you'll enjoy this little test... No QSO, just me rotating sending different words across each type of key I had on the desk.

That's all for now

So lower your power and raise your expectations

Richard AA4OO