Dave Richards AA7EE

March 29, 2013

Good Ops, Ben’s Best Bent Wire, and Some New Home-Made QSL Cards

Nearly every evening on 40M, I hear Bill Crane W9ZN for an hour or two coming in from Chicago. He’s a good op. I’m not sure what his top speed is, but I often hear him conversing easily with others at around 25wpm.  He always matches the speed of whoever he is talking to, which I think is one mark of a good op. I remember the first time I QSO’ed with KA7PUN a couple of years ago.  We were conversing easily at what was my comfortable speed back then (which was probably around 16-17 wpm).  I thought that was his regular comfortable speed until I heard him in QSO a few days later with another station sending much faster. I realized that he had matched my speed and felt very grateful to him for making me feel comfortable in that QSO.

Anyway, back to Bill. I first noticed him on the band for a style of sending that incorporates a variation on the “Ben’s Best Bent Wire” routine that commercial operators used to use in order to loosen up their wrists before a shift. At the time, I wasn’t familiar with this type of routine and only knew that Bill had a style that made him stand out on the band for me.  Here’s what I’m talking about.  This is Bill as recorded last night –

I imagine that a few decades ago, this kind of routine was more prevalent on the bands, but W9ZN is the only station I have heard doing it.  Some people would probably prefer to perform their warm-up routines off the air, but it sure is a good way for Bill to be instantly identifiable. A little online research seems to indicate that he was a Chicago radio personality in the 60’s and 70’s, going by the name of Bill “Butterball” Crane. I’d sure like to QSO with him, but he never hears my puny 5W sigs.  He’s running QRO, and a regular presence in the segment from 7031 – 7034 most evenings.

I’ve also been busying myself with making some new QSL cards, firstly for QSO’s I make with the CC1 beta.  I was inspired by NT7S’ CC1 beta card, and wanted one for myself. I’m lucky to have Photoshop (CS2) and to have finally figured out the importance of layers and how to use them.  The initial version of the CC1 beta card that I came up with looked good on the screen, but due to the fact that I didn’t have a profile for the printer at my local Fedex Kinko’s (they probably don’t have one), the card printed out a lot darker than it looked on my monitor, and some of the text ended up being buried in the background.  I did eventually come up with 2 versions, both of which look OK when printed. One, in my opinion, looks better in print than the other, but I’m waiting to hear back from NT7S as to whether he agrees before I print up a few of one of these two.  These are not scans of the printed cards, but jpeg renditions of the original Photoshop files. Bear in mind when you’re looking at these, that the printer in my local Fedex kinko’s prints files darker than they look on-screen, so if you’re thinking these images look a bit light, that is why –

I finally seem to be getting the hang of using Photoshop to do these kinds of layouts so, bolstered by the success of these cards, decided to make another one. It took me a while to scan the G-QRP Club logo and change it from black on a white background to white on a transparent background, but now I know how to do it, it’s a piece of cake –

Of all these cards, my favorite is my basic 2 color one.  The following image, unlike the previous ones, is not a jpeg generated from the original Photoshop file, but a scan of the final printed card.  I did this because the color of the card stock does a lot to make the card look good.  It’s called “Sawgrass” and unfortunately, my local Fedex Kinkos won’t be restocking it once their current stock is gone –

It’s simple, effective, and prints out well on a variety of printers – no complex graphics that need to be rendered in accurate tones. On top of that, if I need to make a lot and am feeling a bit skint, it doesn’t look too bad in monochrome either.

March 25, 2013

The First CC1 to CC1 QSO – and a QSL as a Memento

Filed under: Amateur Radio,Ham Radio,QRP — AA7EE @ 7:17 pm
Tags: , , ,

I made my first ever QSO with the CC1 beta a week or so ago. It was with Jason NT7S (the designer of the CC1 and proprietor of Etherkit) and on top of that, it was the first ever CC1 to CC1 QSO. Very exciting!  I was hoping to have been the first ever QSO Jason had with his CC1. That honor actually went to WA0JLY, but I did get to be Jason’s 2nd QSO.

He was recording video of the QSO, which is up on his website if you want to take a look, though I am rather embarassed by my sending.  For some reason, I hadn’t plugged the paddle into the jack on the CC1 properly, and when I came back to Jason, the paddle went nuts and wasn’t sending what I wanted it to at all. I finally discovered the error, plugged it firmly into the jack and continued with the QSO. Jason for his part, (due also to nervousness at our historic QSO I’m guessing, just like me,) wished me 71 at the end of our brief exchange.  I like that! As I pointed out to him, 71 is like 72, but even better.  From now on, whenever I QSO with NT7S, I am going to sign off by wishing him 71. Perhaps that could become the default sign-off for any CC1 to CC1 QSO’s in the future? He also told me that my mess-up in sending due to not plugging the paddle in properly is one of those things that help create a narrative to remember these occasions with. Well, I guess so 🙂

I don’t normally collect QSL cards, but some are special. This one from Jason is one of those in my collection that have great meaning.  In the early days of radio a QSL, instead of being seen as merely the final courtesy of a QSO, was the much-desired proof that a hard-worked for contact had taken place.  The early hobbyist would labor hard building his entire station, and spending many hours adjusting and tweaking in order to make contacts with other amateur stations. QSL’s were highly-prized pieces of proof that validated the work of the experimenter. I got some of that feeling on receiving this card from Jason –

This, in my opinion, is a QSL in the best time-honored tradition of amateur radio.  I’m running off to Fedex Kinko’s this morning to do a test-print of the custom QSL I’ve designed for my CC1 beta and hope to be spotting myself on QRPSpots later this week once I repair the final that I fried. I’m still not completely sure what I did, but it most likely had something to do with a stray clipped component lead or metal screwdriver 🙂

March 21, 2013

The Etherkit CC1 1st Beta – A Trail-Friendly QRP CW Transceiver

About a year and a half ago, I posted that I had completed the first beta version of the Etherkit CC-Series QRP CW Transceiver.  It was a neat little rig, with low RX power consumption (of the order of 50mA – a bit less, I think), full DDS VFO coverage of any one HF band, a built-in keyer with memories, RIT and XIT, as well as firmware that could be updated at will with a simple AVR ISP programmer (you can get them for around $20). It also used a lot of SMT devices, and was my first serious project using these tiny parts (the KD1JV Digital Dial was the first).

My CC-20 beta worked, and I made quite a few QSO’s with it, including some DX. It wasn’t perfect though. The DDS VFO had some in-band spurs, the TX/RX switching produced a thumping sound, the input and output of the crystal filter weren’t as isolated as they should have been, you could hear some low-level processor noise on the receiver audio,  and the sidetone sounded a little rough too.  Although that sounds like a long list of woes, I think that anyone who designs circuits is used to tackling these kinds of issues one by one, until the dragon is slain. We (by which I mean Jason NT7S, the man behind Etherkit) did manage to improve the isolation of the crystal filter by a fairly good amount during this beta build.

Then he came out with the OpenBeacon kit and the EtherProg.  I knew he hadn’t forgotten about the CC-series, but I’m thinking he wanted to get a few other kits up and running before coming back to tackle it again, which he duly did.

The rig has been renamed the CC1 and although it retains the same basic architecture, there are a number of changes and upgrades to the design. It is still a monoband QRP CW HF transceiver (available in your choice of band) with an output of 2 – 3W (depending on the power supply), and it still has a DDS VFO (tuned with a real knob!) that covers the entire band, as well as RIT and XIT (useful for working split), freq readout in morse code and a built-in keyer with memories. The firmware is still also upgradeable via an AVR ISP programmer.  Although at this stage in the development it has not yet been implemented in the firmware, Jason thinks it should be possible to include APRS functionality and WSPR too. That’s quite a lot for a rig that is not much bigger than a pack of playing cards.

The beta kit arrived in a Priority Mail flat-rate box (what a neat sight on top of my mailbox!)  The enclosure is to the left, in the middle was the bag of parts for the EtherProg (a separate Etherkit product which can be used to update the CC1 firmware). The big bag on the right is the bag of parts for the CC1 –

The CC1 parts bag opened up to reveal the inner packaging.  The bag containing the bigger parts has been opened and those parts dumped into a mint tin.  The EtherProg, as I mentioned, is a separate Etherkit product and is available now, but I’ve included it in this photo. You can see the board slid partially into the enclosure –

A view of the underside of the board. Our beta kits had the microcontroller pre-installed. Currently, this was the only way Jason could supply it to us flashed with the firmware, but regular production kits will not have this IC pre-installed (it will have the firmware already flashed though) –

In true Etherkit spirit (the phrase “Open Source Amateur Radio” is on their home page), the beta testing forums are open for anyone to view here, and the forum for the CC1 beta is here. Only beta testers can post in these forums, but anyone can post in the product support forums which are here (you have to register first.)  The CC1 beta forums include schematics and an assembly guide which, although not final of course, will be of interest to anyone who might have an interest in the kit when it becomes available.

A couple of days of soldering, and the receiver section (which is about 85% of the circuit) was finished. Alignment consists of peaking 2 trimmer caps in the bandpass filter, and adjusting the BFO so that the wanted signal is in the center of the passband.  The passband for my filter is not flat – there is a definite peak in the response,  so I adjusted the BFO to place the wanted signal at the peak of the filter curve.  I already had a noise source that I had built to adjust the filters for my K2, and Spectrogram on my computer (for the same purpose) so I used these to adjust the BFO frequency.  Both the noise source and the use of Spectrogram are detailed here. With the receiver aligned, I have now spent every evening since just listening to it. I keep looking at it and thinking, “That little thing is a radio?”

Here’s the CC1 board with the receiver section completed –

You can see the GPS connector at the left-hand side of the board (the rear) immediately under the green key jack –

The onboard connectors are really great. They save a whole lot of hassle with wiring, and make it a lot easier to run the rig on the bench before putting it in an enclosure. In the following picture of the underside of the board, you can see U4, the 50Mhz master oscillator and to the right of it, U5, the DDS VFO chip. On the right-hand side of the board in the center, is U1, the NE5532 AF amplifier (I just saw a cat hair lying on top of U1 – those things get everywhere).  You can also see the space for U2, the transmit buffer –

At first, I thought the receiver wasn’t functioning correctly, because on attaching an antenna, I heard only a very faint increase in background noise. I tweeted to Jason and informed him as such, as well as posting to the other beta testers in the forum.  My theory was that the AF amp had low gain.  As it turned out, it was a combination of the bandpass filter being way off it’s peak, and the initial BFO freq placing the signal fairly well outside the passband of the crystal filter. Had I thought to peak the trimmers before jumping to conclusions, I would have realized that all was well.

The receiver was sounding good. The DDS spurs that were present in my CC-20 beta are no longer an issue.  The crystal filter has better isolation – there is still some room for improvement, and that will be improved further before it comes to market – in fact, Jason just suggested a circuit change in this direction that beta testers are implementing as we speak. The TX/RX switching is very smooth and the sidetone sounds nice. There is a sharp leading edge on the sidetone waveform which gives a clicking sound, but that will just require some simple shaping, which, once again will be taken care of in the production model. EDIT – another blog, and also a discussion in a Yahoo groupo, seem to have misread my last statement as meaning that there are key-clicks on the transmitted signal.  This is NOT the case. The transmitted signal sounds nice. I was referring to the sidetone only, which is a simple thing to take care of.  I emphasize also that this is a beta,  and we will most likely be taking this little rig through another beta before it goes into production. The other issue, the processor noise that was present in the audio, is vastly reduced and by the time you read this, will most likely be cured altogether, as Jason just re-wrote the firmware, which I am waiting to apply to my beta.  Things are looking very good for this little rig.

A couple more views of the board at this stage, before we move on –

Having confirmed that the receiver is working,  the final push was on to build the transmitter and complete the rig.  It didn’t take long – just the installation of 12 parts and 2 more toroids to wind.

Here’s the completed board, before installation in the enclosure –

The world of SMT seemed like a closeted world of intrigue and mystery before I built my first project using them.  I had read web sites detailing the use of solder paste and hair dryers, or toaster ovens for soldering these tiny little parts.  It was a while before I realized that you can actually solder them the good old-fashioned way – with a soldering iron and a roll of solder.  I pick up resistors and caps and place them close to their final resting place on the board with a fine pair of needle-nosed pliers. Then, with a small jeweler’s screwdriver, I gently nudge them into their exact position on the pads. While carefully holding the part down with the tip of the screwdriver, I tack-solder one end in place. Then I solder the other end, and go back to the first end to properly solder it.  I use a 1/32″ chisel tip and 63/37 .02″ solder with a mildly active rosin core.  0.015″ solder would be even better, as it’s easy to apply too much solder (which is where a good-quality de-soldering braid, such as Soder-Wick, proves invaluable.)

IC’s with fine lead pitch are a little trickier. The NE5532 AF amp was relatively easy, as the leads are far enough apart to solder them individually. Needless to say, a very clean and well-tinned tip is vital. I wipe my tip on a damp rag and tin it before every joint – unless I’m soldering a number of joints in quick succession one after the other, such as with IC’s.  The AD9834 DDS chip has leads that are too closely-spaced to solder them individually. The technique that I learned from Jason involves soldering all the leads on one side with a big wodge of solder, paying no attention to whether the leads are bridged together with solder.  Afterwards, you clean up the solder bridges with de-soldering braid and a larger iron tip. A larger tip is useful here because you can wick up the excess solder more swiftly in order to avoid destroying the chip. Jason posted a good description of how to do this in the assembly guide.  Search for U5 on that page and you’ll find the description, along with a picture.  Flux is said to be very helpful here.  I managed it with no extra flux (other than that in the solder) , but plan on getting some for future use.

The CC1 is billed as a trail-friendly rig, and the kit will come complete with a pre-drilled enclosure with silk-screened front and back panels.  The enclosure we received with our beta kits is the exact same enclosure that will go out with the kits, with the exception that ours weren’t drilled or printed.  So the following pictures represent roughly what the final CC1 will look like, without the silk-screened panels. There might be a slight adjustment in the spacing of the controls before the final production model too.

Firstly, this one’s for size comparison with my CC-20 beta –

The board slides into rails in the side of the extruded aluminum case and is held in place by the nut on the BNC connector at the back.  Here’s a couple of front views without the front panel –

Man, is this thing a beaut or what?

I’m very fond of this little rig. I’ve only made 1 QSO with it so far (with Jason NT7S) but have spent every evening listening to it. It’s great to have the earbuds in, listening to 40M on this diminutive little transceiver while working.

I’m hoping to get some audio up at some point, but it may take a while. If you’re wondering when you can get one of these, well, it’s still in development but at this point I think it’s safe to say that it will be coming out. I do know that Jason NT7S is a perfectionist and won’t release it until he feels it’s truly worthy, and all issues have been thoroughly worked out. The design is already very close to where it should be and there’s a great momentum behind it, but we still have a 2nd beta to go through  Stay tuned and we’ll keep you posted.

March 10, 2013

W9RNK’s WBR Regen Odyssey

About a week and a half ago, I received an e-mail from Rich W9RNK.  After a long period of home-brew inactivity (about 20 years) he decided to pick up the soldering iron and start building again. He said that my post on building N1BYT’s WBR Regen Receiver (the most popular post on this blog by far) had inspired him and I consider that a great compliment. If one of my posts inspires someone to do something they haven’t done before, or haven’t done in a long time then in my mind, it completely justifies having and keeping this blog.

He did get his WBR receiver working after some initial setbacks   It seems that his problems were caused by using a core material for the inductor that wasn’t suited for the frequency. He used a toroid with a blue core, which is quoted as not being suitable for frequencies over 3MHz.  On substituting the recommended yellow color-coded toroid, the receiver started working.  In his write-up, which I link to below, he shows the schematic of 7N3WVM’s version of the tank circuit which includes a 0.22uH inductor from the center-tap to ground.  The QRP-Tech Yahoo Group run by Chuck Adams K7QO made the WBR the subject of an informal group-build not too long ago and I noticed that some of the members experienced problems with sensitivity. Steve AA7U found that adding a choke from the center-tap to ground alleviated the sensitivity problem. Based on his experiments, he determined that the optimum value is around 1uH.  I had no problems with sensitivity, so my recommendation would be to build the WBR as per the original QST article, and to experiment with adding an inductor if you do experience low sensitivity. However, I do wonder why others have had these problems when I haven’t?  In the original article, Dan mentions that the length of the stiff wire connecting the center-tap of the coil to ground should be about 1″.  I was careful to make mine about 1″, as well as to connect it to the ground-plane of the PCB, as opposed to connecting it to some other grounded point on one of the potentiometers or the enclosure. That’s all I can think of but hey – if an inductor works for you, that’s great.

The other main issue Rich had with his WBR was drift.  I hadn’t measured the drift on mine as it seemed to be quite good. However, prompted by his observations, I decided to take measurements on mine today.  From a cold start, it showed by far the biggest drift rate in the first minute (no surprise there) by drifting 120Hz downwards. In the next 14 minutes, it drifted another 190Hz down, for a total drift of 310Hz in the first 15 minutes. In the next hour, it drifted another 240Hz down, and the hour after that, 100Hz.  I would have been interested to see what the drift was in the 3rd hour but boredom, and the lure of other tasks to complete prompted me to stop!  My WBR (which has an AF preamp stage, unlike N1BYT’s original design) still only draws 13mA so when using it, I used to leave it on all day.  I’d find that I could set it on a net frequency, come back an hour or two later and hear little drift, so I’m thinking that had I measured the drift in the 3rd and 4th hours it would have been less still. Not world-class, but not bad at all for a circuit with no attempts made at temperature compensation, and intended just for general listening.

Here is W9RNK’s write-up detailing his odyssey towards a working WBR Regen.  It’s a pdf file, so make sure you have a pdf reader on your computer.

Rich W9RNK – WBR Build

Many thanks to W9RNK for writing this up, so that it can be shared with others. Hopefully it will serve as an inspiration to anyone else who hasn’t picked up a soldering iron in a while.

PS – I do, like Rich, think an AF pre-amp is a worthy addition. I took N1BYT’s advice and used the same pre-amp that he used in his OCR II Receiver (Sep 2000 QST). Here’s the schematic of the AF stages of my version of the WBR –

That’s it for now.

March 9, 2013

TX5K – A Well-Run Operation and an Interesting Island

Filed under: Amateur Radio,Ham Radio,QRP,Uncategorized — AA7EE @ 8:47 pm

Today is the last full day of the TX5K team’s operations from Clipperton Island and they have already dismantled site B.  Site B was mostly CW which meant I had to abandon my hopes of getting them on 40 and 30 in order to achieve a clean sweep on QRP CW for all bands 80 – 10M.  However, getting a single DXpedition on 6 bands is a first for me, and I’m pretty happy with it.

Clipperton Island is an interesting place. In modern times, 4 nations have fought for ownership, partially for it’s strategic positioning, and partially for it’s surface layer of guano (translation – the island is covered in bird-droppings.)

The last attempt to permanently colonize Clipperton ended in tragedy in 1917. I’ll leave you to read about it but as Joel KB6QVI said to me, a film could be made from the drama that occurred on this small island in the North Pacific. Indeed it could, and not much in the way of dramatic license would need to be taken in order to make the story compelling for a modern movie-going audience. The Wikipedia entry about Clipperton Island is here (opens in a new window), but you’ll find a more lively read in this article.

Two things struck me almost immediately about TX5K –

1) It was a well-run DX-pedition. Good operators with good ears.  I’m referring to the CW side of the DXPedition as I almost never use SSB when DX’ing,

2) The ability to see the QSO you have just made on a map on their website, as well as see your QSO in their online log is fantastic.  No waiting for a log to be uploaded – the website updates every 60 seconds and with each of the 6 QSO’s I made, I saw myself in the log no more than 2 minutes after making contact. No more wondering whether to make an insurance QSO.  This kind of technology can only have a positive effect on the QSO rate, although I did hear the op on 40M last night firmly talking to a station that had already made 5 QSO’s with him. Obviously that station wasn’t familiar with the near real-time online log. The software was written by expedition leader Robert Schmieder KK6EK, and sets a new standard for DX-peditions of this size,  IMO.

Here’s how I managed. Got ’em on 6 bands with 5W of CW to a 66 foot inverted vee with the apex at 47 feet (partially obscured and partially in the clear).  As Justin VE3XTI commented, I sure suck at SSB 🙂

John AE5X got a clean sweep on CW from 160 – 10 which, knowing John’s experience, came as no surprise at all.  I pretty much expect that kind of excellence from him!

Aerial photos of the DX-pedition site can be seen here.  Teams who mount these events go to great lengths so that we can have our DX QSO’s, so no whining on the cluster because the entity you want is not currently taking QSO’s in your preferred mode from your part of the world. In fact, don’t treat the cluster as a message board, period.

Cordell Expeditions, who mounted the TX5K operation, are planning an expedition to Heard Island in Antarctica next year. Definitely one to look out for.

In the meantime, I’ve got my sights set on 9M4SLL, which takes place March 10th – March 18th. Thanks for the tip-off Jason 🙂

Blog at WordPress.com.