SolderSmoke Daily News — Ham Radio Blog
Serving the worldwide community of radio-electronic homebrewers. Providing blog support to the SolderSmoke podcast: http://soldersmoke.com
Mike Caughran, KL7R, SK: Well-known low-power (QRP) and homebrewing enthusiast Michael S. "Mike" Caughran, KL7R, of Juneau, Alaska, died January 22 of injuries suffered in an automobile accident in Hawaii. He was 51. Caughran may be best known as one-half of the team -- with Bill Meara, N2CQR/M0HBR -- that created and produced the weekly SolderSmoke podcast <http://www.soldersmoke.com/>. "I think people were drawn in by Mike's friendly voice and manner," Meara commented on a memorial page for KL7R <https://kiwi.state.ak.us/display/mc/Home>. A member of ARRL and the Juneau Amateur Radio Club, Caughran also wrote articles for the Michigan QRP Club's T5W newsletter and he was an active ham radio contester. "Mike was one of those people who you instantly like because of his honest, straightforward and humble way of talking and expressing ideas," said Mike Hall, WB8ICN, who edits T5W. "His co-hosting of SolderSmoke provided me hours and hours of enjoyment." Caughran was an IT professional with the State of Alaska. Survivors include his wife and son.
We’ve frequently said that is pays to check the W7ZOI web site. Tony G4WIF did just that and pointed us in that direction, noting that the LADPAC “Ladder Package” software is now available for download from that site.
Homebrewers will really want to have that package on their computers. There are all kinds of useful programs in that package: software for designing crystal filters and feedback amplifiers, a program that allows you to think systematically about receiver gain distribution and dynamic range, and many other useful things.
You can get the program here:
http://w7zoi.net/emerrata.html
Also on Wes’s site is this May 22, 2022 picture of Farhan VU2ESE with EMRFD co-author Bob Larkin, W7PUA.
Thanks Wes!
To re-cap: The problem became evident when trying to “net” or “spot” my transmitter onto my receive frequency. Around 18.116 MHz, I could hear at least two tones in the receiver as I moved the transmitter frequency. I needed to get rid of the extra tone.
First, thanks to all who sent in suggestions. They came in literally from around the world, and this is a demonstration of the IBEW in action. I used or at least tried all of them. They were all good ideas.
Following Vasily Ivananeko’s pseudonymous suggestion I rebuilt the carrier oscillator (apologies to G3YCC). I used the carrier oscillator/buffer circuit from Farhan’s BITX20.
Henk PA0EME said I should look at the signal level at the input ports of the NE602 mixer. Henk was right — the VXO input was far too high. I lowered it, but the problem persisted.
At first, I thought that the spur in question was so small that it would not show up on the air. I could not see it in the TX output using my TinySA spectrum analyzer. That was good news and bad news: Good that it was not showing up on the air, bad that I could not see it in the TinySA and use that image in the exorcism.
At first I thought that the spur was being caused by the 10th harmonic of the carrier oscillator and the third harmonic of the VXO. This seemed to fit. So, following VK3YE’s sage advice, I built a little 69 MHz series LC trap (using a coil sent by AA1TJ, on a board CNC’d by Pete N6QW). That trap succeeded spectacularly in crushing the 10 harmonic. Look at these before and after shots on the TinySA:
Spectacular right? But guess what? The problem was still there.
I scrutinized the situation once more. I realized that the spur would be more visible if I put the TinySA on the input of the transmitter’s PA (a JBOT amp designed by Farhan) as opposed to putting it on the output. Watching the spur and the needed signal move in the TinySA as I tuned the VXO, I realized that they were moving in opposite directions. This indicated that the spur was the result of a carrier oscillator harmonic MINUS a VXO-generated frequency (as the VXO frequency increased, the spur frequency decreased). Looking at my EXCEL spread sheet, I could see it: 8th harmonic of the carrier oscillator MINUS the main output of the VXO.
To confirm this, I plugged the values into W7ZOI’s Spurtune program. Yes, the spur popped up and moved as predicted.
For further confirmation I shut down the carrier oscillator by pulling the crystal from the socket, and then just clipped in a 5.176 MHz signal from my HP-8640B signal generator (thanks KB3SII and W2DAB). Boom! On the TinySA, the spur disappeared. Now I at least knew what the problem was: a harmonic from the carrier oscillator.
Following good troubleshooting practice, I turned off the gear and went to bed. When I woke up, an idea came to me: Before launching into a lot of filtering and shielding, just try running the carrier oscillator at a lower voltage, seeing if doing so might reduce the harmonic output. I disconnected the carrier oscillator board from the main supply and clipped in a variable voltage bench supply. Watching the signal on my TinySA, I watched as the spur completely disappeared as I reduced the voltage from around 13V to 10V (see video above). The main signal frequency level did not change much. I tested this by listening for the hated extra tones. They were gone. Exorcised.
Key lessons:
— Spur problems are difficult to troubleshoot. Armstrong’s superhet architecture is, of course, great, but this is definitely one of the pitfalls. Single conversion makes life easier. IF selection is very important. Choose wisely!
— When looking at the TinySA as you tune the rig, pay attention to which way the spur is moving. This provides an important clue regarding the combination of harmonic you are dealing with.
— The TinySA is a very useful tool. It seems like it is easier to use than the NanoVNA (which is also a fantastic tool).
— It can be fun and rewarding to re-visit old projects. In the years between original construction and the re-look, new test gear has become available, and the skill and experience of the builder has improved. So problems that once seemed insurmountable become fix-able.
— Thinking through a problem and thinking about possible solutions is very important. It pays to step away from the bench to think and rest. Rome wasn’t built in a day. Here’s a rough block diagram that I drew up (noodled!) while trying to figure out this problem:
I built the transmitter almost 20 years ago. It is in the larger box, which originally housed a Heathkit DX-40. There is a lot of soul in that old machine. Details on this construction project are here: https://soldersmoke.blogspot.com/2021/12/junk-box-sideband-from-azores-2004-qst.html (The smaller box is a Barebones Superhet receiver set up for 17 meters.)
In the 2004 QST article I discuss a problem I had with “spotting” or “netting.” This is something of a lost art, something that you had to do back in the pre-transceiver days, when running a separate transmitter and receiver. This was how you got the transmitter on the receiver’s frequency. Essentially you would turn on the carrier oscillator and the VFO and let a little signal get out, enough to allow you to tune the VFO until you heard zero beat on the receiver. My problem was that around one particular frequency, I would hear several zero-beats. This made netting the receiver and the transmitter hard to do.
Important note: This is really just a problem with the “netting” or “spotting” procedure — the problematic spur does not show up in any significant way in the output of the transmitter. I can’t see it on my TinySA. But it is strong enough to be heard in the unmuted receiver sitting right next to the transmitter. And that creates the netting problem.
In the QST article, I said that I noticed that the problem seemed to be centered around 18.116 MHz. As I approached this frequency, the tones — desired and unwanted — seemed to converge. That was an important clue. In the article I said I thought that I could eliminate the problem with just one trimmer cap to ground in the carrier oscillator, but looking back I don’t think that this really fixed the problem.
I recently took a fresh look at it. Exactly which frequencies were causing the unwanted signals that appeared in my receiver?
I used an Excel Spread sheet to find the culprits.
— Farhan talks about his practice of taking the pictures of new rigs with the new rig sitting atop the book that was most important in its design and construction. FB.
— I was really blown away by Farhan’s presentation of how the uBITX advertisement was inspired by and in many ways based on the Heathkit ad for an HW-101. Amazing.
— I learned a lot from Farhan’s discussion of SDR theory. I pledge to spend more time with this. I really like Farhan’s hybrid HDR/SDR approach.
— But I have a question: Farhan seems to say that we’d need a big expensive GOOGL computer to do the direct sampling HF SDR. But doesn’t the little RTL-SDR do just that? Without a GOOGL?
— Great to see Wes’s AFTIA being used in the sBITX.
— Really cool that Farhan has his mind on VHF transverters when designing the sBITX. I liked use of the TCXO module to free up one of the Si5351 clock outputs. FB. And great to include an idea from Hans in this rig.
Thanks very much to Farhan (who stayed up until 3 am to do this!) and to the RSGB for hosting.
There he is. Wes Hayward, W7ZOI in 1957. I had never seen this picture before. I found it on Wes’s recently updated “shackviews” web page: http://w7zoi.net/shackviews.html .
There are so many treasures on that page, and on all the other portions of Wes’s site.
Some highlights for me:
— Wes’s description of the station in the above picture.
— On his page about Doug DeMaw, Wes mentions that after Doug edited Wes’s 1968 article about direct conversion receivers, Doug built some himself, experimenting with different product detector circuits. Having used Doug’s mixer circuit in many of my rigs, and having recently experimented with different product detectors for my HA-600A, I kind of felt like Doug was watching over my shoulder, guiding me along as I experimented.
— Wes’s use of a digital Rigol oscilloscope. Makes me feel better about giving up on my Tek 465.
— The page about Farhan’s visit to Wes, and the awesome gathering of homebrew Titans that ensued…
— Wes’s meeting with Chuck Adams.
Thanks Wes. Happy New Year and best of luck in 2021!
This has been a lot of fun and very educational. The problem I discovered in the Lafayette HA-600A product detector caused me to take a new look at how diode detectors really work. It also spurred me to make more use of LTSpice.
In the end, I went with a diode ring mixer. Part of this decision was just my amazement at how four diodes and a couple of transformers can manage to multiply an incoming signal by 1 and -1, and how this multiplication allows us to pull audio out of the mess.
But another part of the decision was port isolation: the diode ring mixer with four diodes and two transformers does keep the BFO signal from making its way back to into the IF chain. This helps prevent the BFO signal from activating the AGC circuitry, and from messing up the S-meter readings. LTSpice helped me confirm that this improvement was happening: in LTSpice I could look at how much BFO energy was making its way back to the IF input port on the diode ring mixer. LTSpice predicted very little, and this was confirmed in the real world circuit. (I will do another post on port isolation in simpler, singly balanced diode mixers.)
At first I did have to overcome some problems with the diode ring circuit. Mine seemed to perform poorly with strong signals: I’d hear some of the “simultaneous envelope and product detection” that started me down this path. I also noticed that with the diode ring, in the AM mode the receiver seemed to be less sensitive — it was as if the product detector circuit was loading down the AM detector.
One of the commenters — Christian — suggested putting some resistance into the input of the diode ring circuit. I put a 150 ohm pot across the input, after the blocking capacitor. The top of the pot goes to the capacitor, the bottom to ground and the wiper to the input of L1 in the diode ring circuit (you can see the circuit in the diagram above). With this pot I could set the input level such that even the strongest input signals did not cause the envelope detection that I’d heard earlier. Watching these input signals on the ‘scope, I think these problems arose when the IF signals rose above .7 volts and started turning on the diodes. Only the BFO signal should have been doing that. The pot eliminated this problem. The pot also seemed to solve the problem of the loading down of the AM detector.
With the pot, signals sounded much better, but I thought there was still room for improvement. I thought I could hear a bit of RF in the audio output. Perhaps some of the 455 kHz signal was making it into the AF amplifiers. I looked at the circuit that Wes Hayward had used after the SBL-1 that he used as product detector in his Progressive Receiver. It was very simple: a .01 uF cap and 50 ohm resistor to ground followed by an RF choke. I can’t be sure, but this seemed to help, and the SSB now sounds great.
A BETTER NAME?
One suggestion: We should stop calling the diode ring a diode ring. I think “crossed diode mixer” or something like that is more descriptive. This circuit works not because the diodes are in a ring, but because two of them are “crossed.” From now on I intend to BUILD this circuit with this crossed parts placement — this makes it easier to see how the circuit works, how it manages to multiply by -1, and to avoid putting any of the diodes in backwards.
Google led me to VK2FC’s amazing site. I was digging up info on product detectors and I landed on Glen’s description of his version of the W7ZOI Progressive Receiver. Glen’s website provides a very detailed, board-by-board description of how to build this great receiver. I now want to build one.
http://www.vk2fc.com/progressive_receiver.php
Glen’s site has many other projects. Check them out:
http://www.vk2fc.com/index.php
And here he is, the Wizard of Horseshoe Bend:
Bill,
Why do you guys make your Soldersmoke podcasts so darn intriguing such that I can’t listen to them in the background while I’m doing something else? Good grief! I start listening and before long you make me stop and chase down a rabbit hole to find something new that you mentioned that I had no clue was out there. Before long I’m doodling out a new sketch or playing with at a new design for something I really need to experiment with or build “next” or something I need to try. It is taking too much of my time!! J
I’ve been listening to your podcasts for years. Way back, before I knew you and before I knew you were doing these Soldersmoke blogs with Mike, KL7R, and just before he was so tragically killed, I was collaborating with him on a simple frequency counter project using a PIC microcontroller. We were making good progress on a neat design. I later completed the project but always kept his contributions noted as part of the source code.
I’ve been making PIC-based VFOs for years – dating back to about 2000 – aiming them at builders who were looking for something to go along with Rick Campbell’s (KK7B) receivers. Rick is a good friend now, after we met in the Kanga booth at Dayton where we both were demonstrating our stuff. (Bill Kelsey (N8ET) of Kanga, was the “marketer” for my kits as well as Rick’s for many years.) My original VFO kits used a DDS (high-end AD9854) that simultaneously produced I and Q signals which made it perfect for Rick’s phasing gear. Rick is a big supporter of my work but he still kids me about polluting his beautiful analog world with my “digital crap” (copyright KK7B term). When I came out with a newer version VFO using a Silicon Labs Si570 PPLL (I can hear already Pete Juliano groaning) it was a big improvement over the AD9854 in noise/spur reduction. I documented this all in a QEX article in about 2011 and Rick (and Wes Hayward) were very supportive/appreciative of my work.
I have used the Si5351 also and I understand Pete’s point of view. It’s “plenty good” for most amateur projects. However, it remains a fact that the Si570 is a better part and produces a cleaner signal. That’s the reason why the Elecraft KX3 uses a Si570. Granted, the newer Elecraft KX2 uses a Si5351 but it’s most likely because they wanted to preserve battery life (the Si570 uses more power but not nearly as much as the AD9854) and also to reduce the cost. I do understand! I also fully understand the ability of the Si5351 to produce I and Q signals via different channels. I’ve had extensive conversations about this with Hans Summers, at Dayton and online. I use a pair of Flip-Flops on the output of the Si570 instead. My PIC code driving the Si570 is ALL written in ASSEMBLER code. Yep! I’m an EE but have had a career mainly in software development and much of it was writing assembler code. I dare say there aren’t too many gluttons for punishment that do it this way. I do it because I want to understand every line of code don’t want to be dependent on anyone else’s libraries. Every line of code in my VFO’s and Signal Generators is MINE so I know I can debug it and it can’t get changed out from under me. (This problem bit Ashar Farhan hard on the Raduino of his BitX. Tuning clicks appeared because the Si5351 libraries he used changed between the time he tested it and released it. I was really appalled when I dug into this and resolved to NEVER use libraries that I didn’t write myself. Similarly, this also makes me have some distaste for Arduino sketches. I would rather see ALL of the code including the initialization code, the serial routines, etc, rather than having them hidden and get pulled in from Arduino libraries. That’s similar to the reason why Hans Summers didn’t use an Arduino in his QCX. He used the same Atmel microprocessor but developed/debugged it as “C” code with the full Atmel IDE/debugger.
By the way, Pete mentioned the Phaser FT8 transceiver by Dave, K1SWL, in a recent podcast. Dave is a very close friend, even though I haven’t met him in person since about 2000. We Email at least daily and some of it is even about radio. J I did the PIC code for the tiny PIC that controls the Si5351 in the Phaser. Yes, it’s written entirely in Assembler again! I do know how to do it for a Si5351. That Si5351 code is not nearly as much “fun”, though. I know, this will make very little difference to guys who write Arduino “C” code to control it but under the covers it’s a world of difference. It takes me about 15 serial, sequential, math operations to generate the parameters for the Si5351. None of them can be table driven and they all have to be performed sequentially. (This is all hidden in about 5 lines of complex, Arduino “C” code but the operations are all there in the compiled assembler code.) In contrast, my Si570 code is almost all table driven. I just have to do one large (48-bit) division operation at the end to generate the parameters. Yes, that’s a bit of trickery to do in ASM. There are no libraries do this.
I will point out one more advantage of the Si570 in comparison to the Si5351. It has the ability to self-calibrate via software instead of relying on an external frequency standard. In my Si570 app I can read up the exact parameters for the crystal embedded inside the Si570, run my frequency-generating algorithm “backwards” and determine the exact crystal frequency (within tolerances, of course) for that particular Si570. Then I update all the internal tables using that crystal frequency and from then on all generated frequencies are “exact”. I love this! Frequency often moves by about 6 kHz on 40M.
Oh yes, I must mention the difference of home solderability of the si570 vs the Si5351. Those little Si5351 buggers are terribly difficult to solder at home while the Si570 is a breeze. I know, many folks will just buy the AdaFruit Si5351 board and it’s already soldered on but, again, I like to do it all myself. No “magic Fruit” for me.
Now that I retired a couple of years ago and am getting out of the VFO kitting business I can finally build complete rigs instead of just making the next-generation VFO’s for everyone else to use. I recently build a tiny, Direct Conversion rig with a Si570 signal generator (of course) and a diode ring mixer (ADE-1). Look at my web page, www.aa0zz.com to see it, along with my VFO projects that I’ve been building in the past. As you well know, Direct Conversion is fun to build and the sound is astounding; however, they are rather a pain to use! Yes, I did make it qualify as a real rig by making several contacts all over the country. (Wes Hayward gave me the criteria: he told me that I must put any new rig on the air and make at least one contact before it qualifies as a real rig.)
The new rig that’s on my workbench is my own version of a phasing rig, experimenting with a Quadrature Sampling Detector (QSD, sometimes called a “Tayloe” mixer), using some ideas from Rick’s R2 and R2Pro receivers and many innovations of my own. At present my new higher-end Signal Generator works great, the QSD receiver works great (extremely quiet and MDS of -130 dB on 40 meters) and the transmitter is putting out about 16 watts with two RD16HHF1’s in push-pull. You can take away my “QRP-Only-Forever” badge too, not that I’ve ever subscribed to that concept! Still more tweaking to do with the TX but now I’m also working on the “glue” circuitry and the T/R switch. The SigGen, RX and TX are all on separate boards that plug into a base board which has the interconnections between boards and the jacks on the back. I’ve built DOZENS of variations of each of these boards. Fortunately they all fall within the size limit criteria to get them from China at the incredible price of $5 for 10 boards (plus $18 shipping) with about 1 week turnaround. Cost isn’t really an object at this point but it’s more of getting a hardware education that I sadly missed while I concentrated on software for so many years. it’s certainly nice to have willing mentors such as Rick, Wes, Dave (K1SWL), Don (W6JL) and many others to bounce my crazy ideas off. Yes, I’m having a ball!
I was licensed in 1964 but out of radio completely from 1975 to 1995. Do you like the picture of my DX-100 on my web page? My buddy in the 60’s had a Drake 2B and I drooled over it but couldn’t afford one.
Now I must finish this rig before you guys send me down another rabbit hole. Too many fascinating things to think about! I literally have a “priority list” on the my computer’s desktop screen. Every time I come up with a new project idea – something I really want to play with such as a Raspberry Pi, SDR, etc, I pull out the priority list and decide where it fits and what I want to slide down to accommodate it. That’s my reality check!
Take care, Bill. Thanks for taking the time to give us many inspiring thoughts and ideas.
73,
-Craig, AA0ZZ
http://www.arrl.org/shop/Experimental-Methods-in-RF-Design-Classic-Reprint-Edition/
And it includes a .pdf copy of Solid State Design for the Radio Amateur.
 |
| Farhan’s DC40 |
Dean KK4DAS asked me to speak to our local radio club, the Vienna Wireless Society. It was a lot of fun. I talked about my evolution as a homebrewer, some of the rigs I made, the moments of joy, and the tales of woe. You can watch the presentation in the video above.
I was really glad to be able to explain in the presentation the importance of people like Pete, Dex, Farhan, Wes, Shep and even Dilbert.
I was also pleased to get into the presentation the N2CQR sign that Peter VK2EMU made for me. Thanks Peter!
Here is the URL to the YouTube video (also above):
https://www.youtube.com/watch?time_continue=3414&v=VHSr-v4QO7Q&feature=emb_logo
And here are the PowerPoint slides I used:
https://viennawireless.net/wp/wp-content/uploads/2020/06/VWS-presentation-Rig-here-is-homebrew.pdf
