LF converter, tuner and antenna

Completed loop antenna, yellow converter box and tuner box

Completed loop antenna, yellow converter box and tuner box

A few years ago at Dayton I was visiting the FAR Circuits tent and ran across a little circuit by Larry Coyle, K1QW, for converting the LF and VLF band to 4 MHz. This was done to allow users to experience the perceived improvement in sensitivity of many receivers in the 4 MHz band as compared to the often poor performance down in the LF band. A very well-written two part article accompanied this circuit and can be found in the November 2008 QST, “A Modular Receiver for Exploring the LF/VLF Bands”. Part one covered the circuit I put together, the converter. Part 2, which I will probably not explore (at least not anytime soon) was a software defined receiver. It is described as a good way to get introduced to software defined radio and might be worthy of consideration.

Converter Circuit

The converter is another SA602-based creation, incorporating a switchable 12 dB attenuator to knock down out-of-band signals which may overload the mixer chip, and a “5-pole elliptic low-pass filter with a cutoff around 550 kHz” (to reject MW broadcast band signals). The parts to build most of this came out of my junkbox and are commonly available. I did have to order the transistor he used for the local oscillator input into the mixer, the 4 MHz crystal, a TL317CLP voltage regulator and the Schottky diode used in the power supply.

I was going to post the circuit but I will refrain as I do not see it in the public domain.

The basic functionality of the circuit is straightforward. The signal from the antenna is low-pass filtered and up-converted to 4 MHz. A transformer off the 602 chip converts impedance to around 50 ohms for export to an external receiver.

The description of this circuit implied that the full spectrum from DC up to about 500 kHz should be received fairly well with a loop antenna and no tuning should be required. However, this was not the experience I had with this circuit.

Antenna for LF

Originally I wound a ferrite rod style antenna, with an inductance of about 1 mH. This was deaf as a fencepost and I could not receive any signals whatsoever either using the converter box or via direct input (mismatch and all) into my Kenwood. I injected signals into the converter box from my signal generator just to verify the circuit was working, and sure enough I was getting output in the 4 MHz band. But no beacons and not even any AM broadcasts were coming through at the low-end of the MW band.

Reluctantly, I wound a larger loop antenna on a wood frame, the wire forming a square of about 2 feet on a side. I had one of these as a teenager and had kept it for a few years but I found traveling with it to be cumbersome and tore it down. This time around I built it to be easily taken apart and re-constructed at some later time. Or at least that was the plan.

Normally these antennas are constructed such that you make some grooves at the ends of the support beams and wind the coil around the frame to sit within those grooves. But I wanted the wires to be contained better than that, and thought I’d instead drill holes at the ends of each beam and just thread the wire through the holes.  Bad move! It took HOURS to wind the loop this way! And there really is no simple method to thread these holes with large lengths of wire. I ended up doing 20 painstaking turns in this manner, yielding about 700 uH.

The pain of winding yielded results though: signals! I could receive some NDB beacons in my local area (two total). And at least the local broadcasters could be heard.

But signals were VERY weak. I found that the beacon was almost impossible to hear, and I could actually hear it stronger off of my end fed longwire tuned to 40m on my TS-2000. Something seemed way off to me.

It could be that because my loop was a little smaller than the author’s the impedance was too mismatched (his was 30 inches on a side to my 24 inches). But the entire band was really dead. I did not hear any lightning “whistlers” even as the Virginia summer thunderstorm season was raging on around me. That was kind of a disappointment.

Tuning the loop

I decided to do a little engineering on the circuit and investigate the impedance. I put the antenna on my TAPR VNA and soon learned that the impedance was very high and inductive. Not surprising I suppose for a big coil. The input impedance of the SA602 chip is also rather high, around 1.5 kohms. So it might not be that badly matched, but for such high impedances my VNA tool is a little hard to interpret (the values seem to swim all over the Smith chart). I’m not really sure if I’m seeing a kohm or 100 kohms.

I wanted to see if I could bring the impedance down closer to 50 ohms at around 200 kHz and just feed my Kenwood with it directly, without the converter, and see if I could at least make out the beacon. In order to do this in a practical manner I needed to use a balun, then tune the hot side with a series capacitance (using the balun means I don’t need to try to simultaneously tune two lines). This actually worked very well! I have a really nice air variable in my junkbox that will tune from a few pF up to about 1200 pF, and I pressed that into service. I found someone else on the internet (see his interesting little writeup at lf loops) who had posted a design for their loop antenna and said that they found a good match by using 2200 pF capacitors in series just before the balun. Having difficulty initially finding a match, I decided to copy his layout and amazingly I got very good matches from about 200 kHz up to 600 kHz! Wow. I felt like I struck gold.

Plugging this into the Kenwood, signals began coming in much better. The level of the beacon (at around 270 kHz) was stronger now than off of un-tuned wire antennas and the local broadcasters were booming in. Progress.

But I still wanted to make use of my converter box! After all, I had built the circuit into a nice metal box with connectors, switches and even a nice little green LED! I even painted the box yellow to signify VLF lightning signals.

Since I knew how to tune this to 50 ohms I decided the easiest thing to do would be to just bump the impedance back up to some arbitrary high value to match what the converter box would be expecting. Admittedly by this time I was losing interest in the project and kind of wanted to wrap it up, so making this match perfect was less important to me than just getting it to work reasonably well.

I ended up rolling two more toroidal transformers, one a 1:4 balun (to bump up impedance and convert back to a balanced transmission line) and one a simple 1:2 transformer. With the two transformations I should get an output impedance of about 800 ohms. I also found that if I switched in another capacitor in parallel with my big variable I could get it to tune down to about 170 kHz. The details and a picture are shown below.


Notes on the tuner circuit for the loop and loop construction


Innards of tuner box (manhattan construction on some copper clad)


Large air variable, yellow converter box and small black and silver tuner box sitting on a wood plank at the base of the loop antenna


The completed system worked very well. Signals and even just the basic noise floor are elevated up to reasonable levels throughout the LF band and a big improvement from the tuned loop by itself. It seems that my Kenwood’s sensitivity in the 4 MHz band is indeed superior to its LF performance. Below 170 kHz though performance rapidly deteriorates and the system is pretty deaf. I could not find values of capacitance or inductance to match down below 170 kHz, although I’m sure such a system could be constructed. Increasing the difficulty is the fact that my VNA does not tune below 200 kHz so I would need to use other methods to design a match.

On the first night after finishing this system I actually heard a broadcast station from Cuba on 570 kHz! It was “Radio Reloj” (https://en.wikipedia.org/wiki/Radio_Reloj) I was surprised to say the least. Apparently when some other stations on the frequency are off-air it is common to hear this Cuban clock station up and down the east coast of the US.

The rest of the LF band was pretty disappointing. I did pick up one more NDB beacon. But there were no pirates or experimental stations heard. And no lightning whistlers! I’ve since learned that there are other design strategies for listening to lightning, employing audio amplifiers. I will either try one of those or contact the author of this article for his thoughts.

I believe I will eventually play with some of the audio-amp style lightning circuits. And I’m also a little curious if anyone else is hearing the megawatt station NAA in Maine that is communicating with submarines in the VLF region (below 100 kHz somewhere). I couldn’t hear NAA, WWVB or the LORAN system at all. So, while I always enjoy tuning things and making baluns and seeing them work, it would also be nice to hear interesting signals!

I’m running out of time before my schedule kills most of my ham radio play time once again. But I’m hoping to look at some Arduino stuff and a Raspberry Pi I picked up at Dayton. Stay tuned and happy soldering.


Posted in antenna construction, LF, lowfer, VLF | Leave a comment

SW-40 completed and on the air!

My completed rig on lower right, inside my pickup truck at the Grindstone Campground in SW Virginia

My completed rig on lower right, inside my pickup truck at the Grindstone Campground in SW Virginia

It’s a great feeling.

My thus far relatively modest journey into homebrewing had another high water point as I finally completed this QRP transceiver, and even better, made my first QSO with it out camping in the woods! The antenna was also homebrew, my multi-band jumper dipole for 10-20 m. The antenna was also used heavily with my commercial Yaesu 857D rig (barely seen on the wooden mount, upper-left). No, I’m not strictly homebrew, I do sometimes like to get on the air with all the bells and whistles of the pro-rigs. But the most joy I had, by far, was the one contact I had on 40 with the SW-40 (I wasn’t actually on the band very long, and had to use a hamstick whip antenna, or else I’m sure I would have made more QSOs).

Balun of multiband homebrew jumper dipole suspended in the trees

Balun of multiband homebrew jumper dipole suspended in the trees

The antenna was created following a much less successful attempt at camping with dipoles last year. The wires were ripping from SO-239 connectors and were very rag tag and falling apart. I knew I could do better.

The finishing touches on the rig came together nicely. I managed to find one of the old, large 1/4″ mono headset jacks from a junkbox given to me by my elmer many years back, and I used that for the key connection (my straight key uses one of these and it’s just kind of a tradition with my rigs, if space allows). You can see it at right (with a key plugged in). It put a nice touch to the creation to have something in there from fellow hams from the past.

Backside of rig

Backside of rig

I also like to put SO-239’s on my rigs. I know lots of people are using BNCs or other smaller connectors, but I like going “old school” and it mates with everything in my shack. I also like having two choices for power hookup on my rigs: the banana plugs and a power socket from Radio Shack (I believe it’s called a “type M”). I suppose I should make up my mind and stick with one choice! But I’ve found it convenient for the power hookups I typically use.

Front side with Bencher and an old "Curtis K5" keyer!

Front side with Bencher and an old “Curtis K5” keyer!

The box, in case anyone was wondering, was purchased from “QRP kits” in 2015. It was designed to be used with their version of Farhan’s BITX rigs, but I really liked it and wanted to use it for different kits too. I am not particularly skilled with metal working, nor do I have access to any specialized equipment. So having a nice box pre-made is a real treat.

The lettering is just your basic transfer lettering I’ve used since I was a teenager. It can be found at art supply stores or online. It isn’t the best way to go but I haven’t yet worked out how to do the laserjet printer transfers. I then spray some clear coat on top to keep the letters on.

This cabinet has pre-drilled holes for a speaker. I don’t currently have a speaker installed and I’d need another audio amplification stage to power one. I may add this down the road or I may not. It would drain a little more current and most of the time I prefer headphones on CW anyway.

The display was also purchased from QRP kits and itself was another kit, the KD1JV digital display. One note about this kit: it radiates some noise on 4 MHz, which is the IF used in this kit. I spent a day just tracking down the source of this “birdie” and it was driving me mad. I finally realized it was the display. Shielding helps to knock it down, but given the constraints of a pre-made PC board from the kit, it is hard to find real estate to solder a shielded box around anything. So I did a workaround of installing a power switch to the display. Once I’m tuned to a known frequency, the display goes off. It also conserves battery power which is nice when running on AAs out at a campsite.

I dug up my OLD Curtis K5 “Lil Bugger” keyer from my childhood and used it with the SW-40. It worked great! I always liked the thumbwheel to dial up what speed you want on CW, and the push button on the back to key up for tuning. I am sure more sophisticated and smaller keyers can now be installed inside the box, but this brought back fond memories of late nights on the air with an old Galaxy III and a dipole on my roof.

My one QSO was with W8IX who has the appropriate nickname “Dit”. He lives in Indiana and is a more serious homebrewer. So it seemed like it was homebrew-to-homebrew operating! Couldn’t have asked for a more fun 1st QSO with this rig, which has taken about 14 years to finally complete!


The unboxed board and display

I’m now moving on to other things. I’ve got a VLF converter board that I picked up at FAR circuits at Dayton a few years ago. I’d actually already ordered most of the parts for this, and the rest I had in my junkbox. I should wrap this up later in the coming week, if all goes well, and I hope to be hearing some interesting noises from lightning down around 5 kHz! The LF and VLF bands are pretty unexplored regions for me, so this is breaking some new ground. And I’ve never built a converter type project before (translating VLF up to 4 MHz). I guess unfortunately it’s another NE602-type project (these are truly everywhere), so I won’t be breaking new learning ground really. But it should be fun and I’ll be posting about that once it’s done.

Posted in antenna construction, Camping, first light, first QSO, KD1JV digital display, SW-40 | Leave a comment

I’m Baaack! Dusting off the station. And, the SW-40 is on the air!

Completed SW-40 kit board from Dave Benson K1SWL

Completed SW-40 kit board from Dave Benson K1SWL

I am back to the blog after an almost 1 year hiatus from all forms of ham radio! It’s good to be melting solder once more. Really, really good!

During my down time, I have seen hits to my blog every so often. Someone seems to find my blog and read several entries, maybe looking for some clues in whatever project they’re working on currently, or just to satisfy their curiosity, boredom, or whatever. I’m glad it seems to still be useful or interesting to some out there. Because I don’t post anymore, I’ve lost any kind of regular following, which is understandable. Still, I’ll pass this blog entry along to Bill Meara of SolderSmoke in case some of his friends might enjoy the entry.

I had stuffed the kit board for the SW-40 last August but I was missing the PA transistor, a 2SC2078. I think I had stolen one from this kit to replace one I’d managed to blow up in the SW-20 some years ago. No worries. Some web browsing had found a supplier and the part (and some spares) arrived about a week later in the mail (I had none in my junkbox and due to the low cost I wanted to just use the intended part rather than make a substitution).

Unfortunately, my time was extraordinarily limited in both the fall and spring. I had ZERO time to play on radio. All that time, the kit, and my radio hobby life, sat idly by on the bench, untouched. I would occasionally look across the room at it, longingly, remembering happier days when I could play radio! It was sad to be honest. But I knew eventually I would find an opening again to play. And, come May 2016, I finally did.

When the dust settled in my schedule, the first order of business for me was to do some late spring antenna work (an SW-40 without an antenna is not too useful). I think many of us play antennas during the warm months of the year (at least those of us who live in colder latitudes).

I found that some severe windstorms had spun my 17 m dipole around on its mast, and it was beginning to torque the coax feed. The U-bolts normally hold quite well but I’m using a cheap plywood mast-to-dipole plate, and after a year some slack had developed, allowing the dipole to rotate a bit. This was solved by running a pair of bolts through the plate and mast. If any rotation takes place now, it will happen at the base of the mast, which can be easily corrected.

The longwire/tuner arrangement I use for most of the HF bands had held up perfectly. The homebrew tuner (see earlier blog entries for that!) is protected from the elements by a Rubbermaid “Roughneck” plastic tub, and it has done its job. The longwire had sagged slightly, which was easily corrected. I set the controls for 40 m CW and had under 1.5:1 VSWR, just like a year ago.

Feeling the ham radio mojo starting to return after the antenna work, I finally came back to my workbench.

This is no simple thing. Having been away for so long, it took me some time to remember what was what. I felt like I’d been in suspended animation and was just coming to.

Even approaching the work area after such a long time away seemed like some type of religious experience to me. As I rediscovered where things were again, the excitement began to build. I really wanted to finish this little rig and get it on the air.

The long-dry yellow sponge of the soldering station once again found water. The iron was switched on. The Kenwood TS-2000 behind me was turned on to listen to the background banter of hams on 40m. I was back in business.

I soldered in the PA transistor, found some pots for volume control and tuning in my junkbox, and soldered up some leads for a key. I did the “smoke test”, applying the 12V DC. And…


No smoke, but no signal anywhere on the 40m band. I double checked my cabling and my monitor rig and tried again. Still nothing.

The current draw was very low and barely changed from RX to TX, which is a very bad sign! It was at this point that I felt some frustration creep in about the manner in which the kit was built and NOT tested, stage by stage, which is what I normally try to do. Because of limited time last summer, I had simply thrown all the parts into the board and crossed my fingers. Let this be a lesson to those novice builders out there! This is NOT the way to go!

Fortunately, I guess, this (and probably most of my near-future radio projects) are simple enough and I’m familiar enough with the circuits that troubleshooting is not that hard. In fact, having it NOT work right away provided a mystery. And forced me to go back, stage by stage, to investigate my creation. Which was what I should have done before anyway.

The scope, which I’d not touched in a year, was activated. Dave’s kit has excellent documentation which actually has almost all nodes of the circuit annotated with both the DC and AC expected signal levels! This almost seems like cheating to me! But hey, it’s there and I used it.

I went back to the LO. No troubles there, a very solid signal around 3 MHz from the Colpitts. Then I checked out the 4 MHz oscillator of the NE602 chip of the TX mixer. Again, solid 4 MHz signal and the voltages checked out. EXCEPT — it died before it reached the input of the following buffer stage and “Q4”.

There are two IF cans, T2 and T3, between the TX mixer and the following buffer stage. The signal was fine on the primary side of T2, but something happened to it by the time it hit the secondary of T3. It was around 500 mV p-p when a 1.5 V p-p was expected. I found out that the capacitor C32 in parallel with the secondary was not the right value. Somehow I’d put in a 150 pF instead of a 47 pF. The change must have resulted in more RF getting shunted to ground, and very little was left to apply to the base of Q4. Once swapped out, I was able to tune T2 and T3 and…voila! The SW-40 came to life!

I was able to get 2.5 W out of this rig, which is the upper range Dave cites as the expected output (1.5 – 2.5 W). That was very satisfying. And the signal is clean as a whistle on the scope.

I was thrilled that the RX side worked fine, first go. I recall that I HAD actually tested some of those stages as I was building the kit, so that probably helped me there. Signals poured in to the little rig immediately and at loud amplification. How thrilling! There is always something special about “first light” for a creation of your own, even a kit.

I feel that the troubleshooting aspect of getting the rig on the air helped to make this kit feel more personally connected to me. Had everything just worked after stuffing the parts, I think part of me would still have wanted to go back through with the scope and measure the various signals coming out of the stages. It seems way too appliance operator-like not to. And all of Dave’s annotations of expected voltage just beckon the hobbyist to play.

What comes next is the boxing of the rig. I know some builders don’t really put their creations in much of a box. But I do like to sometimes take the rigs out camping or picnicking somewhere, and having it in a contained enclosure really helps! I purchased a box and a nice digital tuner kit from QRPkits.com last year, it’s the same one being sold for their Bitx kits (and the one I used for my Bitx17 I built two years ago, the KD1JV digital dial). Another option for a tuning “display” which I had used with my SW20 about 16 years ago was a kind of ingenious kit sold by Dave that would announce your frequency by sending CW through your headphones. It took up very little space. But I kind of like glowing numerals.

So, this one should be a sharp-looking little rig once completed. And I fully expect to make many CW contacts with it. Heading towards another solar minimum, it’s nice to have this old kit completed for a lower HF band. I’ll post a blog once this is boxed.

That’s it! Nice to be back in the “brotherhood”, at least in some small way.




Posted in Uncategorized | 2 Comments

Crystal filter analysis, SW40

filter on board

Three crystal filter from the Small Wonder Labs 40 CW transceiver, connected to the TAPR VNA for analysis

I’ve been away from the blog for quite awhile, mainly because I ran into several roadblocks which stymied my progress with the SW40 build and analysis.

I really wanted to re-visit the type of analysis I did when I built the Bitx20A kit, but the laptop I was using for the TAPR VNA up and died on me!  I really had no time or interest in tearing apart that unit and trying to “fix” it, and it was a sluggish and rather cheap netbook style older machine anyway, so I ended up buying a new cheap laptop to replace it.

Well, unfortunately, that did not resolve matters.  It’s a long and boring story.  But the upshot is, the TAPR software will only run on computers running on a 32 bit OS, and the new laptop was running 64 bit Windows 8.1!  How frustrating!  I contacted the software developer for this tool but didn’t obtain a solution.  I ended up having to partition the hard drive of my desktop and installing a 32 bit version of Windows 7 on it!  That cost some dollars too, unfortunately.  But I’m happy to report that I can at least use this fun and educational VNA tool again, albeit from my desktop via a lengthy USB cable.

If anyone reading this is having similar trouble with the VNA, I am working with a tech rep from Cypress (the maker of the USB chip in the VNA) to see if they can create a 64 bit version of the driver (or help me find a workaround).  I’m not too optimistic.  But if they do, I will create another blog specific to the TAPR VNA and the driver solution, and will also pass along to the TAPR team and the designer.

Back to the SW40!

I revisited my blog from last year on analysis of crystal parameters.  Most of this information can be found by digging through books by people like Wes Hayward or some ARRL handbooks and other resources.  I went through that process last year, discovered the model for the crystals used in RF filters, and discovered the algebra and techniques necessary to find the values of the parameters of that model.  That previous blog has the gory details of that process and I won’t repeat it here.  I will pass along the results and some plots though.

Crystal parameters:
Rs = 5.5 ohms
Lm = 125 mH
Cm = 0.0127 pF
Cp = 4 pF
Series resonant frequency: 3.998944 MHz
Parallel resonant frequency: 4.006180 MHz

And a plot of the S21 for the crystal:

SW40 S21

These numbers, and the plot, compare fairly well to the ones obtained for the Bitx20A filter crystals.  The unloaded Q is higher in this crystal than it was for the other kit.  And the Cm was a little different.  But generally the numbers seem in line for published values typical for these crystals.

Next I built the filter on some clad and connected to the VNA.  The filter is meant for 4 MHz as you can probably tell by the plot.

SW40 crystal filter

SW40 filter tight zoom

The resolution of these snapshots is rather poor for some reason.  But the 3dB points are roughly 3.998922 and 3.999393 MHz for a bandwidth of 471 Hz.  The up and down wavyness in the passband is fairly dramatic and more than 6 dB down from peaks.  But I took the passband as being 3 dB down from these outermost peaks.  Beyond these two points the response drops off dramatically on either side.

The 6 dB bandwidth according to the specs in the assembly instructions from Dave Benson was 500 Hz, so this is spot on it looks like (which is more reassuring from the standpoint that my test setup is accurate than it is a validation of Dave’s work, at the moment at least).  This is much narrower than the Bitx20A filter, but that’s obviously by design — this is a CW rig and not an SSB rig.

This time around I was determined to use a pre-packaged filter analysis software instead of “Puff”.  Puff would work but it’s very basic and I thought it was high time to use some software meant for this purpose.  I tried the AADE package (V. 4.5) and after about an hour playing with it I was able to get a very nice S21-style plot out of the thing.  The general shape and the bandwidth lined up very closely to the output of the VNA, which was exciting.  The main discrepancy was the frequency of the passband was shifted about 4 kHz  downward, but I’m not sure I’m really that concerned for now.

Filter plot output from AADE

Filter plot output from AADE

I was very impressed that the shape of the passband was clearly repeated by the modelling tool, with that second dip being a little deeper and wider than the first.  I was also excited by the fact that it came up with a 3 dB bandwidth of 455 kHz, only 16 Hz different from measured!

I’m not sure if this is the best or most intuitive filter analysis software around.  I thought there was another package made by Wes Hayward or someone.  But given that I only needed about an hour to create a model and input my crystal parameters and generate a useful plot, I’d say that it did the job.

So I will continue on building this kit.  I wanted to refresh my mind about these formulas and the filter analysis generally, and at least I got to play with one of the software tools out there on the net.  I’m afraid I’m going to have to rapidly finish up the rest of this kit though with very little analysis, as I’ve run out of “play time” this summer and will very soon be inundated with studies and projects.  I hope to post a few more blogs on progress on that front.

Posted in crystal characterization, crystal filter, SW-40, TAPR VNA, Wes Hayward | Leave a comment

More op-amp fun; 6m opening

I’ve been playing with the op-amp circuits some more.  While the Radio Shack 741 chip more or less worked as advertised, and I was able to amplify audio out of my iPhone (although slightly distorted, for whatever reason), the NE5532 op-amp that came with my SW-40 kit was not playing along at all.  I looked at the spec sheets and checked and re-checked my breadboard connections but I only got complete rubbish from the output of that chip (using the same circuit as with the 741, for either 20 dB gain or unity gain).  This chip seemed to “sort of” work while in the kit board and connected to the soldered-in supporting cast of resistors and capacitors.  But it honestly didn’t sound nearly as good as the 741 did, and the amplification was minimal (and came with bad AC hum it picked up from my supply).

So, I concluded that this chip may be “fried”.  And I ordered some more through Mouser.  If it turns out that somehow the chip is fine, it won’t hurt to have a few spares, since I’m fully capable of destroying them later on, especially through wear and tear of possible field use (or static carpets someplace).

I decided to move on to the good ‘ole NE602 mixer and LO portion, a configuration I’m much more familiar with.  There’s also the IF crystal filter, which I will put on to the VNA and grab a snapshot of it’s s-parameters (just as I did with my Bitx-20 last year).  I plan to model the crystals and plot with Puff again, as I think this is a good exercise for me going forward with other rig designs.  This time I’d also like to try and use some pre-packaged filter analysis software courtesy of ARRL.  I have the disks and I might as well give them a try too.

My goal is to finish this project by the end of June.  There may still be some packaging issues dribbling into July but those shouldn’t be too awful.  The SW20 was a robust little kit and I’m expecting this one to fly well right off the bat too.

In other news, there was a nice little opening on 6m for part of the ARRL contest that was going on this weekend.  Sunday night I got in on some rare 6m CW contacts up and down the New England coast and into Nova Scotia.  I also snagged a station down in Bermuda and a few close-in grids near Roanoke.  It was fun to see the band come alive.  I heard some frustrated commentary from some saying that the band had openings but nobody was on, and they were getting tired of just sitting there calling CQ!  I guess even pressing the pre-programmed “CQ” button on the modern rigs gets tiresome after awhile.

Some minor news in the shack involved making up some jumper cables for the bench.  I now have a high current Astron 12V supply at the ready along with two analog multimeters (one just a voltmeter) and one digital multimeter.  This is the nicest bench I’ve ever had, by far.  Working on projects is much more efficient and fun now.

Coming up, I plan on doing some camping, and, hopefully, some operating from my old SW-20 kit rig.  I’d bring the Bitx-20 but with compromised antennas (and probably shaky propagation at night) I just don’t think this is the right time.  Ideally I’d run the SW-40, but, well, it’s still on the bench!

Posted in 6 m, op-amp, SW-40 | Leave a comment

The 741 chip and op-amps

From the late DeMassa's engineering textbook, Electrical and Electronic Devices, Circuits and Instruments

From the late DeMassa’s engineering textbook, Electrical and Electronic Devices, Circuits and Instruments

From spec sheet for the LM741

From spec sheet for the LM741

I decided I wanted a little bit more information about op-amps.  Just soldering in the audio amp into the SW40 kit board and moving on with only vague recollections from 20+ years ago about what these chips do wasn’t satisfying to me.  So, to the internet, and my old college textbooks, I went.

The top photo is of a generic op-amp circuit configured as a “subtractor”.  If you make all the resistors the same, you realize a difference function between the two inputs, va and vb.  The derivation for how this works isn’t all that hard and was found in the cited textbook in the caption.  You make shortcut assumptions which makes most of the algebra drop out and leaves you with simple expressions.  The main assumptions are things like input current into the op-amp can be neglected (it’s essentially zero), and the gain “Ad” is super large, and can be approximated as infinity, especially when on the denominator of an expression you want to go away.  Other important facts about op-amps are a very high input impedance (nice so as not to load down your inputs) and a low output impedance.

The all-important gain expression that I was struggling to find (to no avail) through random internet searches was the ratio RF/RS.  If all resistors are the same, the circuit will just give you the difference in input signals without any amplification.  But by manipulating this ratio you can boost the difference signal up to the point where you start clipping it (which happens pretty quick I found).

A couple of significant gotchas are lurking with hooking up this chip.  First, I went with an LM741 because I’d just purchased a couple of bags of these cheap items from the Radio Shack liquidation a few months back, and I figured if I blew them up it was no great loss.  But looking at the spec sheet and the little package pin description shown above, I didn’t realize fully what the terms meant.  It took some digging online but I learned that the offset null could mostly be ignored and is used to kind of “zero-out” internal chip discrepancies that might, say, cause two equal inputs to seem as though they were not equal.  And I thought I knew what V- and V+ were for — namely, ground and +12V connections.  Oops!

Although I found some possibly questionable websites showing the chip connected in this manner, I discovered that I could not get anything but nonsense results doing things this way.  I spent a couple of hours and tried different chips and resistor values but to no avail, the circuit was NOT finding difference voltages (I’d get things like a DC +8V signal even with AC inputs).

So, more Googling turned up some college physics lab notes on op-amps and lo and behold some used the 741 chip.  They clearly stated you needed to apply +12V to the V+ and a ** -12V ** to the V-!  I made the change (although clumsily, using a battery pack to deliver the V-), and it worked!  Wow!  Difference signals very clearly and crisply seen on the scope.  So if anyone out there in internet land has had similar struggles, I hope this helps (let me know if it did).

What was a little off though was I was still getting gain even though all the resistors were the same.  I can think of a few reasons for this.  With such high gain, the op-amp may magnify subtle differences between resistor values.  We know these cheap 1/4 W resistors are not precision, generally.  Another factor may just be that whole “offset null” thing (just speculating on that).  But, whatever the cause, I discovered quickly that if I made the feedback resistor lower, around 10 ohms while keeping the others at 100 ohms, I made the gain unity and the circuit worked just like the textbook equations predicted.

The next step was to see, could I make an audio amp out of this with 20 dB gain, just like the one in the SW40 kit?  Well, why not?  I already was showing gain when using all 100 ohm resistors.  I found that if I made that resistor around 70 ohms (with the others being 100), I got the desired gain.  With no load (ie, speaker) attached and just watching on the scope, I could inject a 1000 Hz audio tone into the input and see the 10x voltage gain on the output, clean as a whistle.  Note, this is actually better performance than I got with the SW-40 audio amp, which for some reason picked up a lot of 60 cycle AC noise.  Not with this simple 741 configuration.

The final piece of the puzzle was to hook up a load.  I tried a small speaker, those little 2″ type things you see on kit rigs.  Sure enough, the op-amp could not deliver to that load very well and distortions took over with very little input signal.  I fared much better when using iPod headphones.  I could tell the volume would be adequate and distortions really only got bad when cranking it so loud you’re hurting your ears.  Still it seemed reluctant to deliver much volume and so I can see that this is probably a limitation in using op-amps for audio amps.  I feel like I’ve heard or read that this is the case.  The trusty 386 chip seems much more willing to deliver audio oomph.

Still, I found the ability to scale gain up and down dramatically and getting clean outputs very satisfying and cool.  And it was also neat to finally play with these circuits, which had intrigued me for the past 25 years but which I’d never had the time or inclination to actually study.  It was high time to bust out the breadboard and fire these things up and play.

I’m finding I have this 20+ years of pent-up need to explore circuits in me!  These kit rigs really help serve as a guide down that path of exploration I think.  Each stage is something that opens a door into a wider understanding.  And once learned you can use that knowledge going forward into your own future projects.

Next I will probably look at the op-amp chip used in the kit again and see now if I can understand it’s configuration a little better.

Posted in LM741, op-amp | Leave a comment

The solder has melted! SW40 begun

Audio amplifier portion of board and scope probe

Audio amplifier portion of board and scope probe

It’s been almost a year before any real projects have been undertaken.  Finally, I dug out the SW-40 board about 2 weeks ago and began to play.

To refresh, this is a kit I bought back in the early 2000s from Small Wonder Labs, Dave Benson’s now shuttered kit company.  I’d already built a Rockmite 40 and an SW-20 and had a blast.  I’d just begun to understand the stages of the circuitry in these kinds of kits.  But I ran out of motivational gas and time and the 40 got shelved.  For a very long time!

I’m following Dave’s installation order so far, and installed the 78L08 8 V voltage regulator first.  I had to remember where all my hookup cables were but I managed to fire it up and measure spot-on 8V for a 12 V input.

Distorted 10 V p-p 660 Hz output signal from 1 V p-p input to op amp

The audio amp was next.  Dave used an NE5532 op amp chip for this purpose.  I injected 1 V p-p @ 660 Hz into pins 2 and 3 of this chip, in a haphazard manner, first passing through the installed capacitor and resistor network on Dave’s board.  There was a lot of AC hum and you can see the waveform reflects distortion.  I discovered, to my dismay, that it was coming through the power supply leads.  This is a problem I’ve had before with this supply.  But I’m also getting it from the leads coming out of my signal generator.  I put the circuit on batteries and used my iPhone as an audio source and the output of the amp cleared up nicely.

I recall having more trouble with AC hum pickup from Dave’s other kits as well.  The more recently completed Bitx-20 kit does not have this problem.  I may need to investigate a way to clean this up.  I know the hum isn’t there with my more pricey Astron supply.  But for now, at least I know the amp is working, and giving me 10x voltage gain (20 dB).

I tried to determine the theoretical gain from the circuit as drawn but my knowledge of op amps is more than rusty.  I suspect it’s a very simple ratio, but the app notes and info on the web only show very complicated circuits for audio equalizers and mixers and various audio equipment.  Maybe I’ll dig through my textbooks and figure it out.  Or, more likely, I will not worry much about it and move on.  If I ever desire to use this circuit or one like it for my own design though I’ll need to understand it better.

In other news, I ordered another enclosure like the one I used for the Bitx-20.  That one came from Hendricks kits, which has since changed to “Pacific Antenna”.  Although not currently listed on their webpage, the enclosure box is still being sold separately from the Bitx kits.  James, KA5DVS, at Pacific Antenna was very courteous and assembled and shipped everything to my QTH.

I really liked this little box and the accompanying KD1JV digital dial frequency counter display, which fits nicely into pre-cut holes meant for that display.  I decided this would be perfect for the SW40 kit.  The pair set me back a bit price-wise, and I’m sure there are cheaper ways to go.  In the future I’ll need to solve my boxing problems a little more cost effectively.  But for now I’m pleased with this.

Blue Pacific Antenna box and KD1JV digital dial in bag

Blue Pacific Antenna box and KD1JV digital dial in bag

For me, 40 m is really the king of the ham bands for CW operating.  So I want this rig to be one that gets used a lot.  In the future I may well have multi-band rigs which will make many of my current ones obsolete.  But given constraints on my shack time of late, I think I may be relying on this one for awhile.  So, I’m happy it will be in a good enclosure with a decent readout.

Probably nobody else is building these older SW kits now.  On the off chance that someone is, these upcoming blog entries might be of interest to compare notes.

I’m thrilled to be at the bench again.  I’ve got my 6 m running on my Kenwood TS-2000 in the background and I’ve snagged a few grids in the past month as well, as people break the squelch on the 50.125 watering hole, alerting me to an opening.  At the same time my homebrew 40 m receiver, the Mrad-40, is busy playing SSB rag chews, often on antenna topics (I recently learned what a “BOG”, or Beverage on the Ground, antenna was).  I find that soldering and measuring and building things goes better with that background accompaniment, along with of course the “beverage” of choice: coffee.

More to come soon, stay tuned.

Posted in Pacific Antenna, SW-40 | Leave a comment