The Grand Illusion

Five by Nine... QRP... How can it be?

What does an S-Unit really mean and how can a QRP station sound like a 500w station?

The real gain doesn't come from transmitter watts

I rarely go up to the phone portion of the bands.  Since I started doing QRP and CW last summer I think I've only made two SSB contacts just because my focus has been CW.  But last night I was working on a bench project and tuned the Elecraft KX3 up to the phone portion of 40m for some entertainment.

A station in Rhode Island was testing his amp with a new antenna and was getting reports from various stateside operators.  There was a lull where no one was answering his CQ so I thought I'd give him a call.  He was peaking at S9 on my KX3's meter and he also sent me a signal report of S9 with some 10 over peaks.  I then told him I was a QRP station running 10 watts and heard the usual surprised reaction from a QRO station.

I gave him reports as he switched back and forth from running barefoot ~75w to his amp ~500w.  With his amp on he was an S9, running barefoot he was S8.  So the amp gave him one additional S unit.  In terms of hearing him I would have been hard pressed to tell much difference simply by volume.  He was perfectly copyable without noise with the amp off.

We also did some tests with my station at reduced power (as if...)  At 10w-12w output I received a S9 to S9+10.  Reducing output to 5 watts netted me a S8 report and when I reduced to 1 watt (one watt) he was still able to comfortably copy me and I received an S5.  So with my station at 1 watt and his running 500 watts we could still converse via SSB.  Ladies and Gentlemen you don't need as much output wattage as you think you do.

He was running through a newly constructed homebrewed vertical while I was using my 80m OCF Dipole.  We didn't discuss what sort of radial system he had installed, but the difference in antennas was likely the deciding difference in our stations.  We were about 500 miles apart and his vertical probably had more low-takeoff gain than my dipole, or possibly more ground loss so I probably had better high angle gain on him for the short 500 mile skip.  I think the take-away from this exercise is that the antenna is generally the key rather than transmitter power.  If I can get 6 dB of gain from my antenna I have in essence quadrupled my effective radiated power.  Quadrupled?  Yes, Quadrupled effective output power.

Power and S-Units

You've probably heard it before but it's worth a refresher...  Our radio's meter display a measurement in "S-Unit".  To move the meter up one S-Unit the receiver must hear 6 more dB (decibels).  So one S-Unit is 6 dB.

A decibel is:

• A Power Ratio:  dB = 10 Log P₂/P₁
• A Voltage Ratio:  dB = 20 Log V₂/V₁ 

You increase 3dB each time you DOUBLE your power

That's worth repeating... 

You increase 3dB each time you DOUBLE your power.  So to gain one S-Unit you must quadruple your power.

So if you are running 100 watts, to gain one S-Unit using the same antenna system you must increase your power to 400 watts.  If you are running QRP SSB (10 watts) that is only 1.5 S-Units down from 100 watts.  100w / 2 (-3dB)  = 50w / 2 (another -3dB or 1 S-Unit) = 25w... etc.

Antennas offer the cheapest increase in dB

Note the statement above "using the same antenna system".  That's the key then isn't it?  It's easier to get 6 dB of gain from an antenna than from wattage.  A 40m doublet can offer significant directional gain especially when operated on higher bands.  Now unless it's rotatable you will be at the mercy of the directionality of its lobes but if you have trees or tall structures you can very cheaply string up a few dipoles oriented in different directions and for far less money that a 500 watt amp (6 dB).  If you can only have one wire antenna you may miss out on some DX in the antenna's nulls but you will have some stellar gain in the direction of the lobes.  Of course rotatable yagis and beams are the best but now we are talking about real money again.  I'm talking bang for the penny.  You don't have to buy a wire antenna.  Some Dacron rope or weed-eater line and some surplus insulated wire is all you need.  You can even make your own feedline cheaply from electric fence wire and insulators.

We have a great hobby, but there are so many aspects of it that sound like common sense when they really are not... like increase your power.

Increasing power gains you very little compared to a better antenna systems.

That's all for now.

So lower your power and raise your expectations (or your antennas)

72/73 (Note:  72 is a common substitute for 73 among QRPers... as in "not enough power for 73") 
Richard, N4PBQ

1 Watt and a Wire... in the Attic

You can't always get what you want, but you try sometimes...

Recently I've dialed my normal 5 watts down to 1watt (one watt, singular) for all my contacts.  To throw some water on the fire I've decided to use my attic antenna which weaves all around my metal ductwork and electrical wiring.   Mostly this was to prove a point to myself but it may be enlightening to deed restricted hams that they can use a qrp radio and an attic antenna successfully.

Key lineup... Palm Single (paddle), Vibroplex Bug (circa 1970s), Kent Hand key

It only seems pointless until you try

Calling CQ with 1w QRPp into a poor attic antenna isn't as pointless as it would seem.  I didn't have to wait long when calling CQ before I got an answer most of the time.  

Now am I going to bust a pileup with 1 watt ?  Possibly not but I think that my assumptions about both how much power I need and how big an antenna I need are usually out of proportion with reality.

1 mighty watt

My assumptions are often incorrect

I made QSOs on 30m, 20m, 17m and 10m this morning all at 1 watt.  The solar conditions report was not really fantastic, especially for 10m.  Yet 1 watt through the attic antenna bagged the only DX I heard on 10m.  I had a couple of other multiple exchange contacts on 20m, 17m and one good old fashioned 25 minute long ragchew on 30m where I received a 599 report for my one watt from Bob (NR8M) in Ohio.  Admittedly, Bob was booming in and we had good propagation to each other.

http://www.hamqsl.com/solar.html

Video

The recording below was number 4 or 5 this morning.  I wanted to post this one because I was working another QRP station in Arkansas (K5EDM) and we did NOT have great propagation to each other.  He was running 5w while I was running 1w so it was QRP to QRPp.  In the video you can see that I'm using some of the KX3's tricks to pull the signal up because there was a lot of QSB and noise (note the GEOMAGNETIC FIELD UNSETTLED in the solar report).  

I had the volume maxed and was using the RF gain control mostly.  I eventually had to turn on the preamp which really washed me in noise but I dropped the RF gain more and eventually switched in the APF (audio peaking filter) which performed magic on this contact.  Often I find that APF doesn't help but this time it made a big difference.

That's all for now

So lower your power and raise your expectations

72/73

Richard N4PBQ

The need for speed in CW

CW QSO speed statistics... 

As I continue on my Morse Code Journey I have been using a desktop application called Morse News to work on my code comprehension of natural language sentences for ragchew training. I set the program to send at speeds higher than I can copy to stretch me.  Presently I have it set at 26wpm character speed and 23wpm Code speed.  I can usually figure out what the news story is talking about but I miss a lot of the specifics.

I was a kid in the 1970s when Speed Racer was a popular cartoon

The need for speed

Currently I'm OK working up to 20wpm qsos as long as the other station has good spacing.  Sometimes I lose the thread or the specifics in a ragchew at 20wpm and I just ask them PSE AGN? 

In sprints I can copy a callsign sent at 25wpm to 30wpm if I hear it a few times but that's no good for real contests where you need to catch that call sent at 30-40wpm sent the first time.  My friend AA4XX has asked me to help him in multi-op contests but I know he is sending around 30wpm and I just can't hang there at this stage in my Morse Code journey.

Statistics show 25wpm is the magic number

So it got me to wondering what was the actual "average" qso speed.  I downloaded days of raw data from Reverse Beacon Network.  RBN collects the call sign, the SNR and the SPEED.  The raw files contained between 40 to 50 thousand CW CQ calls logged each day after I filtered out other modes.

The average speed from RBN CW logs worked out to exactly 25wpm.  The standard deviation was 5.34wpm which would mean most of the time you would expect to work stations between 20wpm and 30wpm.  That seems reasonable and matches what I seem to hear on air.  These numbers are from the CQ calls themselves and are only measuring character spacing speed so the code spacing is likely a bit lower in most cases and for ragchews is likely averaging 2 to 5 wpm lower than the actual sent character speed.  Since I am not much interested in serious contesting at this stage in my hobby going over 30wpm isn't on my radar.  

So based on those calculated CW speed statistics from RBN data if I can reach the point where I am comfortable in a 25wpm ragchew I should be content (for a while)

That's all for now.

So lower your power and raise your speed.... err expectations

72/73

Richard, N4PBQ

Race to the bottom

1 watt or less?

I've been considering ordering a 1 watter kit but I was concerned that it wouldn't be useful in the downward sunspot cycle. 

1Watter board with initial 100nf caps installed and starting inventory of other parts.
Not shown are the myriad toroids (fun, fun), transistors and pots.

So I spent some time this afternoon making 1 watt contacts on 20m and 40m with my Elecraft KX3 just to see what staying limited to that power would be like.  I've made casual 1w contacts in the past but staying at 1watt was a good test for me.

It turned out that I had a plenty of answers to my 1 watt CQ calls and received reports that were only one or two points down from my report to the calling stations.  The last station I worked dropped down to 300 milliwatts and I still copied him fine. 

So I ordered a 1 watter kit for 40m from Kits and Parts today.

The 1 watter kit is significantly more involved to build than the Elecraft kits I've assembled to date. I hope to learn a lot from the build and alignment and I just might finally have a reason to buy an oscilloscope (hopefully my wife doesn't read this).

Update 2/20/2016:
I received the kit yesterday.  It includes 21 100nf SMT capacitors that the builder "highly recommends using".  Optionally he supplies through-hole capacitors.  I tried figuring out hold to hold the SMT part in place to solder for about 30 minutes and gave up.  I don't have 3 hands. I'd have to have someone show me how to solder an SMT part.  Even in the linked video to building the kit the builder used the through hole components so I don't know why the SMTs were supplied or were highly recommended.  So I used the through hole capacitors. 

Through hole versus SMT capacitors

I couldn't figure out how to solder the tiny SMT capacitors

1watter board with the 21 through hole 100nf caps installed

Inventory of resistors

Next step was installing resistors.  There are six 1/8w resistors missing in the kit:

• R3,9,21,22 (470)
• R12 (1.2K) 
• R14 (4R7)

I haven't built enough kits to have spares like this so I sent them a note and will see how long it takes to get the resistors.  I haven't inventoried the rest of the parts yet but so far I'm a bit disappointed.  I have built 4 kits from Elecraft and one from 4-State QRP and all of those kits had all the parts.  Although in the case of 4-State QRP one of the plastic knobs was broken in shipment.  They sent me another in 3 days.

That's all for now. 

So Lower your power and raise your expectations

73/72

Richard,  N4PBQ 

Skippy 80m

80m QRP 4600 mile skip?

I am still relatively new to operating QRP and operating on 80m. I made a few SKCC contacts last night and had a nice long ragchew with Mark (W1AAF).  When I was shutting down my station I looked at my browser window displaying the Reverse Beacon Network and was surprised to note that my QRP CQ on 80m had being reported by RBN stations in Europe.  The reporting station in Slovenia was over 4600 miles from me! 

I have not seen 80m skip like this from my station so this was new to me.  I have made some cross continental contacts on 80m but never seen it go 4600 miles to the Northeast before. I'm used to that from the higher bands but not my NVIS 80m antenna.

80m Skipping up to 4600 miles

Antenna patterns

This got me to thinking about my 80m OCF Dipole which has been a good performer for me especially to the North and West but generally poor to the South.  However a couple RBN reports also showed my signal as strong as 22db to Puerto Rico and Curacao last night so some sort of 80m propagation pixie dust was being spread around.

Below is a satellite view of my QTH with the layout of my OCF Dipole shown in blue and the direction of propagation to the Wales and Spanish stations super-imposed.

Antenna and propagation

My OCF Dipole is not straight across.  It is bent about 25 degrees as you can see due to my anchor points.  The long end slopes down to about 10 ft above ground while the apex is only at 25 ft tied off to the peak of my roof. The short leg heads off level to a tree in the front yard.  So 25 feet sloping down to 10 feet is certainly not an ideal height for 80m so the take-off necessary for the skip to the distant European stations is very surprising to me. 

Too good to be true?

This antenna has been a good NVIS antenna for me on 80m and a relatively good DX antenna given the recent poor band conditions on 17m and 15m.  Even though I normally operate QRP or sometimes up to 10 watts I nearly always get a signal report that is either equal to or only one S-unit lower than the QRO stations I qso with. I have really been puzzling over this antenna lately trying to figure out why its SWR is so low on so many bands.  It shouldn't be this good.  And if something seems too good to be true it usually is.  Low SWR doesn't always mean antenna efficiency.  When I measure the individual sections of coax connected to this antenna they have SWR and reactance issues.  I still use an old run of coax that comes back into the house that has problems due to a previous lightning surge that was strong enough to melt the solder connections in my spark arrestor so I know things are amiss.  Yet overall SWR, impedance and reactance through all the components connected together is very low on most bands.  The key seems to be my station ground that is nearly resonant on 20m and runs around my house and up to the 3rd story operating position.  The long electrical ground has certainly become "part" of the antenna system, yet I'd expect it to introduce a lot of loss as it ties into 3 ground rods.

But it's nights like this that keep me from "fixing" it.

These sort of mysteries are part of what keeps the hobby interesting for me...

EDIT: August 2016
I did finally replace the coax run back into the house as it turned out that coax braid was damaged.  The result was that my Windom was no longer 1:1 SWR and showed more expected SWR readings.  It's still 1.7 SWR or lower on all the bands except 10m and 6m where it's SWR is 2:1.  So that's likely a more normal response.  The antenna is also quieter now with the replaced coax so all is well.

That's all for now.

So lower your power and raise your expectations

72/73

Richard, N4PBQ

Feeling Extra Hammy

I Passed my Extra Exam

I finally took the time to study for the Extra ticket.  The fine Central Carolina VE team administered the exam and did a very professional job for the group of 8 folks taking their tests last Sunday.

Study Guides

I purchased a well done study guide from http://www.kb6nu.com and used QRZ resources page to take practice exams to prepare for the test.  The Extra test has 50 questions compared to the 35 in the Tech and General and draws from a question pool of over 700 questions.  If you have a background in electronics you will need less study than if you do not.  I am a new student of electronics and the process of studying for the exam was very enlightening.

Why go to the effort?

If you are already licensed as a General class you may not see the need for it as it only gives you a bit more spectrum use on some of the bands but if you really want to chase DX the extra (pun intended) spectrum will come in handy.  But I'd recommend studying for it just for the exposure to concepts you may not have explored and more importantly it will allow you to give back to the hobby some if you pursue becoming a VE.  You can be a VE as a General but you are limited in which tests you can administer so most VE teams want Extra class license holders.  I plan to pursue my VE certification after my ticket shows up in the FCC database.

That's all for now

So lower your power and raise your expectations

72/73

Richard, N4PBQ/AE

Resistance IS NOT FUTILE

The Beautiful Genius Load

In this age of hyper-political-correctness it seems that I should not call my homebrew contraption an "Ugly Dummy Load".  

Ugly is so negative. It probably has a great personality and an inner beauty all its own.

And calling it a dummy is sooo degrading.  Why that's akin to calling it a Dolt, a Dullard, a Dunce, a Fool, a Blockhead, an Ignoramus, a Numbskull or a Simpleton.

So in the spirit of the age where Everyone Gets An Award

Behold the "Beautiful Genius Load"

Beautiful Genius Load

Construction

Up to now I only had one dummy load and it was built into my MFJ Deluxe Versa Tuner II so I thought it was high time to have something a bit more portable.  I already had the UHF connector in my junk box and lots of spare 12g solid copper wire.  So I ordered some 2w/1000ohm 1% resistors for 40 cents each and built the contraption above for about $10.  The bare copper wire is soldered to the bulkhead connector; one wire to the center conductor and the other to a screw terminal.  I simply wrapped the resistors leads around the bare wire and soldered them in place.  In the near side of the photo it looks as though they are not soldered but that is because it was flipped over when I soldered that side.

Concerning 1% resistors... I could have used 5% resistors rather than 1%.  The number of resistors would have averaged out the differences but I like the blue color and everyone wants to be a 1%'er.

Exhaustive parts list:

• UHF bulkhead connector
• 20 x 1000ohm 2watt resistors
• Two stout gauge copper wires

Not an especially good design

This is not a very good design for a dummy load due to its length.  The two parallel wires will generate some capacitance especially at higher frequencies but for ease of building it can't be beat.  In the following section I show the results on the analyzer and you'll see that there is some capacitive reactance that causes the SWR to be 1.1 starting at 14 Mhz.  This certainly isn't ideal but it is sufficient for most amateur uses.  If you think you need totally pure resistance all the way up to 28 Mhz you'll want a different design.

I may place it in a 2 inch PVC pipe with some holes drilled for ventilation to give it some mechanical stability but the heavy gauge wire seems plenty sturdy for now.  Of course, while it's in use you want to keep it clear of metallic objects and not touch it during transmission.  Also it will get HOT after extended use so don't go and grab after you've transmitted into it without determining if it will burn you.

Performance

Well it works mostly like you'd expect a Dummy... errr Beautiful Genius load should.  The twenty 2watt resistors give me plenty of heat dissipation for QRP levels.  I transmitted at 5w into it for a full minute and it was just starting to warm. It should handle QRO loads up to 100w for a few seconds at a time as long as it's given time to cool down between uses.

The SWR is 1.0 on the lower bands and 1.1 - 1.2 on the higher frequencies...

80m SWR 1.0

40m SWR 1.0 

30m SWR 1.0 

10m SWR 1.2

15m SWR 1.1 

17m SWR 1.1

20m SWR 1.1

Make your own Dummy Load... errr Beautiful Genius Load

It's always handy to have a small dummy load handy. You can use more or fewer resistors depending on your needs.  Just remember the formula for calculating resistance is:

Calculating Parallel Resistance

You want to end up with 50 ohm as your total.  Also be sure to use resistors with decent power handling.  Soldering 1/8 watt resistors in parallel isn't going to give you much power handling capacity.

That's all for now

So lower your power and raise your expectations

72/73

Richard, N4PBQ

Making use of Elecraft Mini-module Kits

Connecting the bits and bobs

I have enjoyed building Elecraft Mini-Module kits.  Now to put them to use...

Elecraft Mini-Module Kits

• AF1 Audio Filter
• BL2 Switchable Balun
• CP1 Directional Coupler
• W1 Powermeter

What to do?

I built the kits as part of my learning adventure and to improve my soldering skills.  It's also helped me learn to follow instructions better (my wife says I need to work on learning to follow instructions).  But ultimately these modules are intended to be useful, and in my case they work nicely to when operating my old Ten-Tec Century/21.

My Ten Tec Century/21 is a 1970s CW-only, low(ish) power rig originally intended for Novice license holders of the time.  It has no RF output meter or SWR meter.  It has poor filtering/selectivity compared to modern radios and its analog tuning dial is a bit vague so you generally only know your frequency within 5 kHz. 

The mini-module kits prove useful. I employ the W1 Wattmeter to determine my power output and SWR; the CP1 directional coupler is used to send a 20db attenuated signal to a frequency counter to determine operating frequency, and the AF1 Audio Filter makes operating near adjacent CW signals more pleasant by providing a narrow audio-band-pass filter.  The result signal can be transmitted through a LDG tuner into the BL2 switchable balun connected to my attic Doublet.

Bring out your cables

All these independent modules need to be connected, so tying the bits and bobs together requires a few coax jumpers to route the RF around:

1. UHF to BNC from the radio to the W1 Power meter
2. BNC to BNC From the W1 Power meter to the CP1 coupler
3. BNC to UHF From the CP1 coupler J1 input to switched T1 output to frequency counter
4. BNC to UHF From the CP1 coupler J2 output to the tuner

And other cables:

1. Serial cable from the W1 Power meter to the computer
2. 12v power cables for the W1 and AF1 (unless I want to use 9V batteries)
3. Audio cable from the TenTec C21 to the AF1

So it's definitely not a neat and tidy setup at present. I plan to arrange things more neatly and possibly place the W1, CP1 and frequency counter into a single box. But for now it's fine and I like the flexibility to switch things around or pull a module out to use somewhere else as the mood strikes.

AF1 Audio Filter making crowded band operations pleasurable

  

CP1 Directional Coupler sending off 20dB attenuated signal to the frequency counter

Frequency Counter fed by the CP1 directional coupler.

W1 Power Meter sending its measurement off to the computer

W1 Power Meter Output to Computer

The W1 has a serial output to a PC for use with the Elecraft W1 software.  The software can both configure the meter and display more detail than can be determined from the LEDs.   Source code is supplied and the command set is documented so it would be easy to write your own software for this.

The W1 power meter LEDs give you relatively discrete output information for the lower two ranges (0.1w to 1.4w) and (1.5w to 14w).  But in the high range (over 14w) the LEDs are only displaying 10 watt intervals.  For instance in the high range, when the second LED is lit you don't know if your operating just 20 watts or 29 watts.  It won't trip the next LED until it crosses the 10 watt boundary in the high range so it can be useful to look at the measurement on the computer if you are operating QRO.   I'm not complaining.  I understand that the meter is primarily intended as a QRP meter and for QRP power (less than 15 watts) it offers plenty of information.

Here I brought the TenTec Century/21 up to nearly full input drive (55-60 watts) to see what it could output. The rig probably had a few more watts left in there but I didn't want to push it because I haven't gotten around to replacing some of the out of spec components in the internal power supply.  I normally use this radio under 10 watts (I look for about 30 watts input on the drive meter) but I was curious to see what the old girl could do since I had the meter hooked up to the computer display.

Measuring maximum RF output from the Ten Tec Century/21

The computer interface is a nice touch and the ability to modify the source code to suit is a plus.

Nits and Quibbles

My antenna's native SWR at 15m (~21.08MHz) is around 2.5 so it requires tuning (impedance matching).  After my LDG auto tuner spends a ridiculous amount of time trying to find a match it settles at 1.7 SWR according to the W1 Wattmeter, while the indication on the Autotuner is that it believes the SWR is 1.5 or better, while the radio on the other side of the W1 meter sees a SWR over 2.5.  I only see this behavior on 15m so I think there is some strange impedance reaction occurring in the W1 wattmeter that is changing the reactance on the jumper to the radio.  I've tried a few different jumpers, swapping jumpers, etc.  But it always presents an abnormally high SWR to the radio at 15m.  Now when I transmit into a dummy load I don't see this behavior, so it is some combination of SWR / reactance present at W1 that causes a impedance mismatch downstream toward the radio.  I have more investigating to do but for now I am choosing to not use the W1 Wattmeter in-line when operating on 15m.

The CP1 directional coupler is not entirely transparent and raises the SWR by a bit as signal passes through it.  You would expect there to be losses according to the -20 db taps (one forward and one reverse).  This should work out to about 0.08% loss but I wouldn't expect it to raise the SWR. It adds about 0.1 to your SWR  and occurs even if the forward and reverse couplers are switched "off" and shunt their respective loads to the on-board 50 ohm resistors.  I'm unsure what accounts for that slight SWR bump but be aware that CP1 contributes some very small losses.

Summary

So the Elecraft Mini-modules are fun to build; and with enough jumper cables, can be combined for experiments and general augmentation of other equipment in your shack.  So go out there, build some kits and experiment.  It's a rewarding experience.

I'm trying to decide what I'm going to build next.

That's all for now... 

So lower your power and raise your expectations

73/72

Richard, N4PBQ