During a break visiting relatives in Western NC I went outside and threw my end-fed antenna up into a tree and sat on the front porch swing and worked stations in Maine, Kansas and Cuba.
Elecraft KX3 -- Palm Single Key -- and Index cards for logging
Portable Ease
The Elecraft KX3 is such a fabulously portable radio. It is very totable with sidekicks panels and cover, the only exposed part during transport is the BNC connector. I just make sure that's pointing "up" in the backpack. I didn't take the time to put the internal batteries back in the KX3 so I carried a small 3ah gel cell battery, good for hours of operation. My antenna is a 9:1 balun fed insulated wire. I get it up in the tree using a throwing weight and heavy line, then untie the weight, attach the wire and pull it up. The coax serves partly as a counterpoise, and I clip two more wires cut to different lengths for additional counterpoise.
The KX3's built-in auto-tuner will tune a spoon so it was no problem tuning the end-fed
Operating
The North America QSO party was in full swing and I was looking for casual ragchews so I headed to 30m and heard a number of QSOs but signals were not strong. Checking RBN from my smartphone I saw that I wasn't getting very good reverse beacon reports.
Reverse beacon reports while operating portable
After sending my call out manually a few times I put the KX3's built in keyer memory to work sending out my call and eventually I received answers to my calls.
I only had about 90 minutes to operate but I had a good time working a few stations and listening to other QSOs. Packing up the radio, key and pulling the wire down didn't take long.
I'm still thrilled to have the Elecraft KX3 for such opportunities.
One of the more popular CW straight keys in the venerable Navy Flame Proof. You grip this key using its Navy style knob. The "navy style" is tallish and rounded on top, usually with a skirt underneath. I demonstrate the "grip" used for the Navy style key in an earlier post.
But what's this old knob made from? Compressed horse hair and tar?
Navy Flame Proof Key contains "fillers" in the bakelite material
As I photographed the key I noticed that up-close the knob of my old service key appears to contain fibrous material and it got me wondering about what material they used to make these knobs...
This knob is made of Bakelite
Bakelite is interesting stuff. It was one of the first commercially successful "plastics" developed by the Belgian-American chemist Leo Baekeland in Yonkers, New York in 1907. Leo incorporated "filler" materials in bakelite to give it strength.
All sorts of ground up stuff went into bakelite including asbestos.
So maybe my "Flame Proof" key really IS flame proof...
Bakelite factory
On the other "hand", here's the Kent Hand Key knob for comparison...
Coming back to the hobby in 2015 I re-entered as a CW / QRP operator and I searched for a good QRP radio. The Elecraft KX3 certainly qualified and it has been a great radio to use, especially when portable.
Elecraft KX3 -- stellar radio for portable QRP operations
But a few months after getting my KX3 an old TenTec Century/21 found it's way onto my desk and through numerous blog entries you may note that for some reason I kept gravitating to use it rather than my KX3 when I operated from the shack, even though the KX3 beats the old C21 technically in absolutely EVERY regard. I just enjoyed operating CW with the old radio more than the KX3. I couldn't explain why.
Fast forward to late March of 2016. I purchased a used TenTec Eagle from my friend AA4XX and began using it as my primary radio when I wasn't portable. Here again the KX3 trumps the Eagle in nearly every technical aspect and offers dozens more features. I just kept gravitating to use the TenTec radios rather than the Elecraft.
I used the my KX3 for Field Day in 2016 and after I packed up and brought it home the KX3 stayed in my backpack and only came out for portable outings. It did not go back on my desk. The KX3 cried little electronic tears while the Eagle gloated...
Ten-Tec Eagle -- compact / simple HF transceiver
Why no love for the KX3?
Time passed, and over the new year break I got to thinking about what I missed about having my KX3 on the desk; like its RX/IQ output for HDSDR and the ease working DX splits using it's dual watch capability and it's integration to logging applications like the ability to trigger CW macros from my logging software. The list of "nice-stuff" goes on and on since the KX3 contains multiple kitchen sinks... So I re-organized my desk to make room for the KX3 again and operated with it exclusively over the past few days...
I was getting ear fatigue and my ears rang in the evenings. This was not the sort of ringing in the New Year that I wanted. I had been previously operating the same amount with the Eagle over the past month without the earaches. Something was amiss.
Had I finally discovered why I keep going back to my TenTec radios?
Audio, Audio, Audio
So over time, even when I switched back and forth between radios there was a subtle "ouch" occurring when I used the KX3. I enjoy CW and digging out weak signals can be fun... or it can be painful. I guess when I sat down to use a radio and my hand hovered between the "Oh-so-feature-rich" KX3 and the "Nice-personality" Eagle my brain was saying "choose the nice personality" you're happier that way.
But there was a underlying reality to the choice I was making.
Just the facts mam
I used an audio frequency analyzer to capture audio from each radio by sandwiching the microphone in my headphones. It hears what I would hear. And the graphs tell a tale.
Below is one graph for each radio. The RED graph line in each chart is the averaged "peaked" frequency output audio during the same QSO. Ignore the green line as it was just the instantaneous audio at the time I froze the display between takes. The CW sidetone on each radio is set to 620Hz.
I re-ran this capture for each radio a few times during a lengthy ragchew between two stations. The signal strength was around S5-S7. It wasn't a strong signal which is typical of what I work, especially as the Solar cycle winds down.
I tried the captures with and without noise reduction on each radio. The RF was rolled off as evenly as I could determine for each and both were set to a DSP filter bandwidth of approximately 400Hz. Both radios were using the same antenna and everything was as similar as I make it. RCVR EQ was set flat for the KX3.
Elecraft KX3 CW audio (ignore green graph line)
Ten-Tec Eagle CW audio (ignore green graph line)
The CW audio output from each of the two radios is distinctive
KX3 audio demonstrates
shoulder noise
Eagle has clean audio
There's clearly a CW signal peak around 620Hz in each radio but the KX3 shows a significant shoulder of audio just 9dB down from the peak below the center frequency
Whereas the Eagle has a clear peak presenting a narrow tone range at the sidetone pitch with narrow shoulders down to the filter width.
Confirmation of my subjective tests
When I saw this I literally said "Aha!"
This confirmed what my ears and my subconcious had been telling me. The KX3 is more fatiguing to listen to than the Eagle because it presents more noise in the audio or at least a wider audio signal given the same DSP filter setting. I've always remarked about my TenTec radios that their CW seemed to float above the noise. I believe it's related to the cleaner audio filtering. The TenTec Eagle just has cleaner audio out of the box. It has no audio adjustments beyond AF and NR, no menus for fine tuning. My old TenTec Century/21 sounds the same when using its 500Hz selectivity setting.
In my opinion Ten-Tec just got CW right.
Yes, I have tried using the KX3 RCVR EQ settings to reduce that lower frequency noise and the problem IMO is that the EQ is more for SSB audio. I think the Q for each setting is too broad and when I try to reduce the low frequency noise IMO it just makes the audio sound mushy. I just can't get as "clean" sounding CW tone out of the Elecraft as I can the Ten-Tec.
OK, "sound" is a subjective thing. No two people will hear the same thing the same way and frequencies that bother me may not bother you, but it seems pretty clear from the graphs that the CW audio from the KX3 doesn't match the Eagle.
Summary
Admittedly, my test involved a very small sample size of one radio from each manufacturer. It's just that I'd put the KX3 back on the desk after a many month absence and my ringing ears got me to investigate the cause a bit more scientifically.
I will continue to use the KX3 for portable ops because it is a great self-contained radio and when I work portable I usually operate for much shorter periods so the audio doesn't become an issue.
Man, I hope Ten-Tec can come back from the grave. They sure made some fine radios for CW operators.
I try to keep a log of my contacts, mostly to know if I've previously worked someone and for keeping up with my SKCC awards. I also keep notes from each QSO in my log so I can reference them during followup contacts. I'm not a contester or a DX chaser although I occasionally will take a look at the cluster spots and see if I can work someone across the pond if they're sending slow enough for me to copy.
Mostly I just send out my call and see who turns up.
But with the new year I was curious about how many States I'd worked in the past year and a half, during casual contacts. Turns out I've worked all US states except Nevada and Nebraska (but I could swear I've had QSOs with Nebraska). Unfortunately my logs are incomplete. I've had computer screw-ups where I've lost logs and for about 2 months last year my logging software stopped working due to a Windows 10 update issue and I kept trying to log using the QRZ online site but I found it difficult to use. Also, when I'm portable I rarely transfer my paper records back to electronic logs. I usually just scribble down some notes during my Portable QSOs. So I'm likely missing a few dozen QSOs here and there. I know I should get my logging act together but then the hobby starts to seem like work.
Anyway, poking through my log led me to wonder if I could generate some statistics. I'd previously written a Python application using python hamtools after my first 100 QSOs to parse my log and determine the average age of operators that I'd worked. I wrote about those results in http://www.hamradioqrp.com/2015/09/morse-code-is-only-mostly-dead.html But I couldn't find the program I wrote back in 2015 (I'm not very organized) and as I started to write a new parser it got me thinking there must be existing applications to analyze my log and provide insights.
So how do you analyze your log file? By exporting it in ADIF format.
ADIF
ADIF stands for Amateur Data Interchange Format. Most Ham logging programs can export and import data using this format. It can have lots of fields here's an example record:
<a_index:1>7 <ant_az:3>334 <ant_path:1>S <state:2>MI <band:3>30m <call:4>K8BZ <cont:2>NA <country:13>United States <cnty:7>GLADWIN <cqz:1>4 <distance:4>1045 <dxcc:3>291 <eqsl_qsl_rcvd:1>R <eqsl_qsl_sent:1>R <freq:8>10.11403 <freq_rx:8>10.11403 <gridsquare:6>EN73SX <ituz:1>8 <lotw_qslsdate:8>20170105 <lotw_qsl_rcvd:1>R <lotw_qsl_sent:1>Y <mode:2>CW <my_country:13>United States <my_cq_zone:1>5 <my_gridsquare:6>FM05OP <my_itu_zone:1>2 <my_name:4>Rich <name:25>Steven C Wuelfing (Steve) <operator:5>AA4OO <station_callsign:5>AA4OO <qso_date:8>20170103 <qso_date_off:8>20170103 <qth:7>Gladwin <rst_rcvd:3>599 <rst_sent:3>569 <rx_pwr:1>0 <time_off:6>191503 <time_on:6>185344 <tx_pwr:2>5 <k_index:1>3 <lat:7>43.3504 <lon:8>-84.5603 <notes:97>rig ic7600 ant g5rv, rainy temp 38f, lil snow, left hand sending with single lever, likes my pics <pfx:1>W <sfi:2>73 <app_log4om_validated_callsign:1>Y <eor></eor></app_log4om_validated_callsign:1></sfi:2></pfx:1></notes:97></lon:8></lat:7></k_index:1></tx_pwr:2></time_on:6></time_off:6></rx_pwr:1></rst_sent:3></rst_rcvd:3></qth:7></qso_date_off:8></qso_date:8></station_callsign:5></operator:5></name:25></my_name:4></my_itu_zone:1></my_gridsquare:6></my_cq_zone:1></my_country:13></mode:2></lotw_qsl_sent:1></lotw_qsl_rcvd:1></lotw_qslsdate:8></ituz:1></gridsquare:6></freq_rx:8></freq:8></eqsl_qsl_sent:1></eqsl_qsl_rcvd:1></dxcc:3></distance:4></cqz:1></cnty:7></country:13></cont:2></call:4></band:3></state:2></ant_path:1></ant_az:3></a_index:1>
A web search for ADIF parsers and generators abound. There was a promising Excel macro but it didn't work for me. A bit more searching turned up a really cool site...
QScope to the rescue
QScope (https://qscope.org) is a website that allows you to upload your ADIF file and provides a bunch of statistics and charts. There is standard stuff like "Stations worked", "Bands" etc. and for contesters there are statistics for QSO rates (i.e. how fast are they logging contacts). I was actually more interested in how many of my QSOs were over an hour but it didn't have a statistic for that :)
So here is a small sample of statistics and charts that QScope was able to tell me from my log...
Basic Statistics
Number of QSOs1304
Number of Dupe QSOs (same Call, Band, Mode)254
Imported on2017-01-06
First QSO at2015-07-29 19:25:17+00
Last QSO at2017-01-05 00:04:46+00
Event duration525 days 04:39:29
Distinct callsigns911
Distinct CQ Zones (if specified in the ADIF log)20
Distinct ITU Zones (if specified in the ADIF log)24
Distinct DXCC (if specified in the ADIF log)56
Distinct Locators/Grid Squares (if specified in the ADIF log)863
QSOs per band
Where is your antenna beaming?
This chart is very interesting to me. It shows that 93% of my QSOs are from antenna bearing between 270-300 deg. An OCFD and a Doublet should not be this directional so I'm very curious.
Cool stuff with Maps
The Google map KML support is impressive. You can download KML files generated by QScope and watch animations of your contacts over time and groups of stations contacted in each DXCC region.
Summary
There's a lot of information in your log. So if you're a new operator like me take some time and determine if there are states you haven't worked or if there's a state you overwork (most my contacts this past year were in Ohio).
For starters I'm going to figure out if there are SKCC operators in Nevada and Nebraska that I can schedule some QSOs with to complete my WAS. I'm not looking for an award or a plaque, I just think it would be cool to say that I've worked every state in the US QRP.
It seems as though one of the biggest misunderstandings in amateur radio is the issue of "grounding". I have probably read more than a dozen articles about grounding and sprinkled a little from each regarding setting up my station. But there always seems to be more to learn. This time I learned more about electrical grounding rather than RF grounding.
Situation
I generally try and send my CQ calls at 5 watts RF output or less. If I receive a really poor signal report from the answering station I'll raise my power if I'm able to 20-40w to give them 6-12 more dB to make it more pleasant for them to work me. Propagation has been really poor for me over the past few weeks and I've found myself getting lots of 339, 459 reports etc. and I've been using the Eagle to bump the power into QRO range of up to 80 watts sometimes (gasp).
However whenever I raised my power in such a situation my TenTec Eagle would behave oddly...
After I'd finish keying there would be a brief audio drop and sometimes a pop as the audio came back. I thought an under voltage situation was occurring so I measured the voltage from the power supply to the rig while transmitting into a dummy load. The voltage only dropped from 13.8v down to 13.6v so that didn't seem to be the problem. Even transmitting QRO levels into the dummy load sometimes resulted in the audio pop but it was always after I un-keyed, never during the actual transmission.
I was perplexed. So I put my question to the TenTec Eagle Yahoo Group. Over the past year I have learned more from the technical discussion in that group than pretty much any other single forum... A frequent contributor to the TenTec Eagle Yahoo Group, Bob, K4TAX, answered my question:
Yes, the Eagle chassis is "floating up" that is to say the DC negative is not really at zero volts as current increases with higher power. I suggest a dedicated ground from the Eagle chassis ground terminal to the power supply ground terminal. For all of my ground jumpers I use #10 AGW auto primary wire to which I crimp and solder a proper lug to each end...
I did already have the station ground connected to the AC service ground via a fused connection but it was a small guage wire and apparently insufficient to tie the radio back to the power supply ground. So what did this mean? Bob, K4TAX followed up with an email a couple days later to see if I'd resolved my issue...
Many, I find, believe the DC Negative is adequate between the radio and power supply. In as much as the radio ground system internally has in effect two ground systems, one being for digital ground and one being for analog ground, we commonly find that the resistance between the power supply and the analog ground does allow the radio to float up above zero volts when in transmit. To test for this condition, measure the voltage between the power supply ground and the ground terminal on the radio. Put the radio in CW and key down. The voltage should be less than 0.25 VDC. If it is greater than this, the voltage drop is caused by the resistance in the DC power cable or the ground method used for the station. For this reason it is strongly suggest one have a dedicated ground connection between the radio and the station power supply.
When I performed this measurement I saw that the DC negative between the radio and the station ground was in fact "floating up" by more than 0.5 VDC. When I shortened the DC cable to only 18" between the power supply and the rig that resolved the float but the proper answer was addressing my DC electrical ground.
Station and A.C. Electrical Service Grounding
If not already in place, add Chassis Ground terminal to your Power Supply, and internally bond chassis ground to D.C. negative terminal.
Connect all station equipment chassis grounds direclty to P/S chassis ground. See Figure 1 below.
P/S MUST have three wire A.C. connector, plugged into mating receptacle with ground tied back to A.C. Main Service Entrance ground.
The P/S chassis ground is your Station ground. The A.C Power connector ground bonds to your A.C. Main Service ground.
Figure 1 below depicts ground wire from each piece of equipment is tied back to the Power Supply Chassis Ground.
The Power Supply MUST have Service Ground on A.C.Connector to complete an effective Station AND Service Ground.
The fix
My station grounding was somewhat haphazard. As I'd added equipment I'd added ground straps between new components and whatever was nearby. I had different gauge wires, and sometimes used RF straps for DC grounding and... well it was a mess. Armed with new knowledge from Bob I went to Home Depot and purchased 50ft of #10 AWG with #10 AWG terminal eyes and #10 AWG butt connectors.
I cut a 6ft piece of the #10 AWG wire and every 18” along the length solder/spliced in a 10# AWG wire adding a butt connector resulting in 18” jumpers that allowed each piece of equipment to be attached via a jumper with the butt connector. Each jumper had a male butt connector on one end and a terminal connector to connect it to station grounds at the other. That allows me to easily move equipment around the station ground without unscrewing the ground lugs.
I now have all the radios, tuner, switches and power supply bonded together properly. I also bonded to the chassis ground of my filtered power strip that feeds AC to all the components.
Also at Bob's recommendation I disconnected my station from the separate ground wire that ran to my outdoor ground rods. My outdoor OCF Dipole comes back through an Alpha Delta spark arrestor that still ties into that ground rod system which is also tied into the house service entrance ground.
The result
I'm happier than a slinky on an escalator... Ok I've used that one before but it's applicable again.
First, there is no longer a rats nest of ground wires running between everything. Ok, points for neatness.
I can individually disconnect a component without removing the ground wire through the removable butt connector. Points for not fighting with the ground lugs when I need to pull a component or re-configure the station.
My Eagle can run QRO up to 100 watts out if required with no DC floating issues.
Here's the phenomenal bit... the band noise at my station is now One S-units lower on 40m and 20m and Three S-units lower on 30m. Honestly I have a hard time believing that re-working the DC ground and disconnecting from the long outside ground wire made that big a difference in my noise levels but it has.
Summary
There is so much to learn in this hobby. My "aha" moments just keep coming.
So if you're like me and didn't know that your radio may have separate Negative VDC voltages between its power and its chassis and what the ramifications could be if they were unmatched, well now you know. Just a few Ohms difference between the power supply negative ground and radio chassis ground can have side effects inside your radio. You may not even know there are problems brewing inside there, so ground that station properly for VDC!
Also, if you're running a long ground wire to your station separate from your AC mains safety ground then you're creating a ground loop of at least a few Ohms that gonna cause problems.
I am not saying that you shouldn't (if your run to ground is short enough) run a ground for an antenna grounding system. But that is RF grounding and its a very different topic.
That's all for now...
So get rid of that long ground wire, properly ground your station to your power supply and make proper use of your AC mains safety ground.
In my journey with amateur radio using CW and mostly low powered operations (QRP) I've slowly been accumulating and learning to use different kinds of Morse Code keys.
Navy Flame Proof (front adjustment screw is from my Nye Viking key)
But first...
Before I talk about the Navy Flame Proof let me go backwards a bit and discuss CW keys in general and why there are so many types of keys.
Broadly defined, I think most Hams lump CW keys into two categories... paddles and mechanical keys. Paddles are used with an electronic keyer that produces the actual DIT-DAH Morse Code elements where the operator is only responsible for initiating a Morse element, not controlling the duration. Conversely, mechanical keys rely upon the operator to correctly time the length of the elements. In the case of a "Bug" style key, the mechanical key automatically times the duration of DITS but the operator is responsible for the timing of the DAHs.
I would venture to guess that nearly all contesters use paddles because paddles (either dual or single) require less effort and are less fatiguing to use than a mechanical key. It also seems that the majority of QRQ operators use paddles, likely for the same reason.
But for whatever reason I'm a bit anachronistic and I usually prefer to use mechanical keys.
I started with a cheap, Philco plastic straight key hooked up to a MFJ practice oscillator. It worked, and I made CW contacts using that key (not the practice oscillator). I'm sure a lot of code has been sent over the years by amateurs with such a key but no one would call a Philco a precision instrument. I knew I wanted another straight key besides the Philco. But unless you can visit a ham with a large CW key collection you're at a bit of disadvantage deciding what kind of key you'd like.
CW keys are devices you have to interact with before you know if you'll enjoy them
You can read descriptions and watch videos, but using a CW key is a totally hands-on sorta thing. You might like the looks of a key but if it doesn't feel good eventually you'll set it on the shelf. I find it fascinating how different simple mechanical switches can feel under my hand. We are created in all shapes and sizes and with different preferences. No one is going to determine the "best key" for someone else when it comes to a device that is physically touched and manipulated tens of thousands of times over the course of its use.
OK... let me get back on topic
Alrighty then, so based on that segue it seems that reading a review and watching a video about a CW key is nearly useless. But I'm all about fruitless activity so here goes...
The Navy Flame Proof key was originally manufactured by several suppliers, the Navy 26003A Flame Proof key was manufactured to meet the Table of Equipment needs for ships and planes. During the last three decades of production (ending in 1988), J.H. Bunnell & Co. was the sole source of the Flame Proof key. Bunnell Flame Proofs have "CJB-26003A" stamped on them. Other letters indicate different manufacturers. The "Flame Proof" designation is because the contacts are sealed inside the key where any potential sparking is confined.
There have been different types of Flame Proof keys manufactured through the years but the 26003A specification was the last design of the series. My key was manufactured by J.H. Bunnell & Co. but I don't know the vintage. It seems to have been in service with the Navy because it has a very worn ship stamp on the bottom of the key and the knob itself has seen quite a bit of handling. There are a lot of NOS (new old stock) keys out there that were never in service are are basically "new". I haven't had the chance to use one of those so my review is based on this used service key.
Navy Knob grip
The first thing you may note is the "grip" on the key (see the above photo). This key has what is referred to as a "Navy knob". That means it has a tall-rounded knob where you grip it. This makes it distinct from J-38 style keys that are low to the desk. You can mount this key directly to your desk to reduce the height but it is still going to present a taller grip than a typical American style key. My preference is for tall keys because I operate "European Style" where my arm is not resting on the table. I find it less fatiguing. This article at Morse Express has a good description of the differences in European and American style keys and operation.
Kent Hand (left) Navy Flame Proof 26003A (right)
You can also see my Kent Hand Key in photo above next to the 26003A. It has a different shape to the knob and the disc underneath is different as well. The difference in the two knob shapes results in a markedly different feel to gripping the key. My first interaction with the 26003A was uncomfortable. I was gripping it as I would the Kent and it hurt my thumb, but within an hour of use I'd adjusted my grip and the key became quite comfortable.
You'll note that my key has a non-stock gap adjustment screw in the front and no plastic cover for the terminals (partly why I got it so cheaply). When it arrived it had a standard slotted 8-32 screw in that position but I wanted to be able to easily adjust it so I borrowed a knurled 8-32 screw and nut from my Nye Viking key (which I've never grown fond of) and put it in service. J.H. Bunnell & Co. still has parts and I plan to order the matching screw and nut if I keep this key on the desk.
The Kent is a rather traditional straight key (albeit Euro style) with the contact in front of the bearings, whereas the Navy Flame Proof has the contacts behind the bearings. Both the difference in leverage and travel to the contact gives it a different feel than the Kent.
Interior
So, the difference in grip, knob height and gap from my Kent combines to make the 26003A feel like a very different key. Without trying I naturally send about 2 wpm faster on this key than my Kent. When I try to go fast with my Kent things go awry but I can comfortably operate this key at 19 wpm. A 3rd class radioman in the Navy passed a 20 wpm test so I'm confident that this key can be used faster than I'm operating it.
Ready - Set - Action !
In use, the Navy Flame Proof is quiet. The contacts are inside the metal body and the only real noise is the arm striking the gap setting front screw. Some folks seem to find that noisy and offer various recommendations for quieting that gap screw. I've mounted my key to an inexpensive award plaque (under $3) and it's resting on thin shelf lining. Compared to my Kent Hand Key which sounds like a woodpecker in action this key is downright silent.
The following video was made during last night's ARRL straight key night event and you can see how I use the key during a QSO.
Summary
So should you get one? Absolutely! There's a strong market on e$ay for these keys so if you don't like it you can easily find it a new home. These keys tend to auction from as little as $36 up to $150. The average auction price at present seems to be in the $90 range. I kept bidding low on different auctions and eventually picked this one up one for $41 including shipping, so if you're patient you can give it a try for about the cost of a J-38.
If you've not gripped a "Navy knob" previously give your hand some time to find a comfortable hold, you may also want to experiment with different base heights. I've had mine on a couple different scrap wood bases to try different heights and presently I like this 1/2" base but I might find a piece of steel to attach it to get it down low.
So get a bit of salt air in your shack with a Navy Flame Proof key.
With a bit of time over the Christmas break I'm finally getting around to work on things that have bugged me for a while... like my noisy power supply.
My TenTec Century 21 doesn't have a fan and it has a built-in power supply, so no fan noise. My Elecraft KX3 has no fan, so no fan noise and it can run all day from a small 12v battery. My TenTec Eagle only runs fans when the finals get hot and even then the fans are in the interior of the case and are speed controlled, low-speed, high-volume dudes that I have never heard. I can feel air moving through the Eagles vents so I know the fans work, their just silent. But the Eagle needs an external power supply.
The calming sound of CW wafting forth from the Eagle is soon destroyed by a sound not unlike an air-raid siren emitted from my Astron power supply as it kicks its fan into high gear
Problem
My power supply has a single, large fan that makes me jump every time it whirs into action. Admittedly, my power supply was near the bottom of the range in the Astron lineup. I assume they have speed controlled fans, or at least quieter fans further up the product line. I purchased this one used, about 9 years ago, so who knows, maybe their new power supply's have speed controlled fans.
Anyway, on my model, the muffin fan produces a hurricane force blow switched by a thermocouple located on the transistor heat sink (right hand side). It is a very simple design... when the thermocouple trips it closes a circuit to the fan. When heat sink temperature falls below the threshold it cuts off. So the fan is either ON or OFF. Normally the fan doesn't come on when the rig is idling, and it will only come on once in a while at QRP levels. I just never know when, but I'm always anticipating... like the jerk kid that would flick your ear when he walked by back in school (hmm, I feel as though I may have some un-dealt with issues from childhood).
The fan seems to have a knack for coming on when I'm working a station that is fading into the noise and remaining on until I've missed some key piece of the QSO.
Solution
What I wanted to do was keep the air moving like a gentle breeze rather than a pulsing Category 5, but still allow the fan to come on full speed if needed. So the trick was to get the fan to run continuously at a slow speed whenever the power supply was on but still allow the thermocouple to give it the gas.
The label on the fan says it's 1.4w 12v... I rummaged through my junk box looking for 2w and higher resistors. I didn't find a lot to choose from, since I just haven't been at this electronics building stuff all that long. I had a number of very low resistance ones from 2 ohm to 20 ohm and everything else was 650 ohm or higher.
I clipped test leads to both sides of the thermocouple on the heat sink that controls the fan. I wanted to try different resistor values to see what happened. Predictably, the low value resistors acted pretty much like a closed circuit and the fan ran near full speed. I could have strung a number of the low value resistors in series but that seemed messy and the values were mostly very low. I tried my next lowest value 2-watt resistor, 650 ohm. The fan didn't even come on with 650 ohm resistance. I didn't have anything in between so I put two of the 650's in parallel and it turned verrrry slooowwwly.
You can find TOTAL RESISTANCE in a Parallel circuit with the following formula: 1/Rt = 1/R1 + 1/R2 + 1/R3 + ... "
Ok I was on to something. Turned out that 5 of the 650 ohm 2w resistors in parallel did the trick... That turned out to be near 130 ohm. I left them hooked up for an hour or so to make sure more current wasn't being pulled than I expected, and they only became slightly warm to the touch so I figured my redneck engineering was good to go. I soldered the bundle of resistors leads in parallel and used some heat shrink to keep it neat.
I've used the power supply for a good part of a day and it stays nice and quiet now. Not silent, but good enough to where I can't hear it at all with my headphones on. A key down test into a dummy load at QRO power repeated times confirmed that the fan will go to full speed after a while but in normal operation, the thermocouple never triggers.
By my low standards, I deem this a success.
redneck ham engineering
Perform this mod at your own risk
This is certainly not a manufacturer approved modification and your power supply may be quite different than mine so tinker at your own risk. Oh, and be careful when you have the cover off a power supply there are multiple, high voltage components in there. Make sure it's unplugged and has plenty of time to bleed off power from the big capacitors before you start messing with anything in there.
That's all for now
So lower your fan speed and raise your quality of life...
