SolderSmoke Daily News — Ham Radio Blog
Serving the worldwide community of radio-electronic homebrewers. Providing blog support to the SolderSmoke podcast: http://soldersmoke.com
http://www.hamradiobug.com/homebrew.html
Pete spotted this. Beautiful work. I noted that Byron hasn’t used the Manhattan style of construction. This makes his work look a lot neater, but it makes it harder to modify and debug the circuitry. On the other hand, OM Byron is obviously so good at this that his circuitry probably doesn’t require any debugging or mods.
The red S-meter and freq counter give it a slightly menacing appearance. Very cool.
Who out there can save us from further Dishal distress? Who has one of these fantastic bits of old test gear and is willing to donate it to the cause of accurate filter building?
Thanks to Tore LB4RG for alerting us to this gem.
Wow, Charlie has a lot of very cool ideas in this initial video. Some suggestions:
— With the IRF-510 in the final, beef up the heatsink, and follow Farhan’s lead by including an option for 24 volts on the drain. That would take you up to 20 watts or so (for those who are so inclined).
— Could that Teensy SDR board work at the IF frequency? If so, consider putting it ahead of the crystal filter. This would enable us all to join the Waterfall Signal Purity Police Force.
— I love the OLEDs too, but I found them to be noisy. I minimized the noise with shielding around the OLED and active decoupling on the power line to the AF amplifiers.
More of Charlie’s work here:
https://plus.google.com/107506245856154702088
It is not everyday that you get on 40 meters and run into a guy who has actually built a single sideband transceiver. But that is what happened to me today. I spoke to Dennis, K0EOO. He showed an unusual interest in my BITX DIGI-TIA. He went on to explain that he had done some homebewing himself. In the picture (from 1974), right next to his Vibroplex bug we see his homebrew, 80-10 meter, dual conversion, solid state (except for the 6146 final) SSB transceiver with digital readout. Wow. That’s a beautiful rig.
Off his right shoulder is a homebrew 700 watt amplifier using dual 4CX250s. And behind his left shoulder we see a homebrew tube-type receiver.
Note the look of pride and determination in OM’s eyes. You can just hear him saying it: “Rig here is homebrew.”
More pictures of Dennis and his rigs (including some amazing vintage gear) here:
https://qrz.com/db/K0EOO
http://www.isquare.com/millen/members/k0eoo.htm
http://www.vintagessb.net/k0eoo.htm
A few years ago my wife got me this nice little Sony ICF-SW7600GR receiver. On the front it proclaims that it is “AM DUAL CONVERSION” and “PLL SYNTHESIZED.” It has a BFO, and a filter of suitable width for SSB. It also has a synchronous detector — it generates an internal carrier that matches the frequency and phase of the carrier being transmitted by the SW broadcast (or ham AM) transmitter. This helps overcome the selective fading that often plagues AM signals. Sony advises switching to USB or LSB once the synchronous generator locks onto the carrier. Pretty cool.
The BFO is the reason I wanted this receiver. And wouldn’t you know, when I dropped it, it landed EXACTLY on the little BFO fine tune control pot. It was as if the Radio Gods disliked all the fancy digi PLL synchronous IC circuitry.
I tried without success to find the value of the destroyed pot. Finally, last week I just decided to have a look in there to see if I could just figure it out.
On the board I could see that the pot only connected at two places, so I figured it would be a varactor circuit on the BFO with one end of the pot to DC and the wiper to the varactor diode. I figured I’d try a 10K pot.
This seems to have been some good radio intuition. It works. I went with a small trimmer because it is less obtrusive and because once I set the BFO in the right spot, I think the de facto channelization of the 40 meter ham band will keep most of the SSB sigs in tune. And the Sony only tunes in 1 kHz increments. If necessary I can move the BFO a bit with a small screwdriver. I just glued the trimmer pot onto the back of the receiver — two wires covered by heat shrink run back into the circuitry.
10k might be a bit too small. Maybe 100k would be better? As it is, I can move the BFO above and below the “zero beat” point, and I don’t need more range. Mouser has a small trimmer pot with a tuning wheel that looks like it might fit, so I may try for a proper repair.
This is a rig that came together through a process of Spontaneous Construction. It started out with an innocent effort to get an Organic Light Emitting Diode display to work with an Si5351/Arduino combination. Then I figured I’d make a superhet receiver with it. Then Pete said I should make it a transceiver.
Most of my earlier transceivers so closely followed the schematics of Farhan or others that it didn’t really make sense for me to prepare a new schematic. This one was different. So I decided to prepare a proper schematic. I tried a few of the free-ware CAD or drawing programs, but each of them had a learning curve at the entrance. So I turned to LTSPICE. I have already climbed that learning curve.
The results appear above. Click on the image to make it bigger. I’m sure there are errors in there. And I think some of my parts choices might be less than optimal. But it works well.
The filter was deigned with the help of AADE software.
The idea of using two NE602s with a filter between them came (I think) from the Epiphyte.
The band pass filters were designed with ELSIE software.
The RF power chain is mostly from Farhan’s BITX40 module, with the pre-driver and driver modified for a bit more gain. Farhan’s amp is the most stable power amplifier I have ever used. It hasn’t given me any trouble, even at 20 Watts. Strongly recommended.
The microphone amp is derived from the 741 op amp circuit used in the (in)famous Wee-Willy DSB rig.
The receiver AF amp also comes from Farhan’s BITX40Module.
Please let me know if you spot errors or have suggestions for circuit improvement.
A link to the .asc LTSpice file appears below. Perhaps some brave soul more skilled in LTSpice than I am might want to turn this drawing into an actual simulation. Some of the parts (like the NE602s) have actual simulated components behind the drawings. Others (like the relays and the LM741 and LM386) are just drawings. But go ahead and flesh this thing out. Who knows, it might come to life in the PC and start making QSOs on 40!
Here is the .asc file: http://soldersmoke.com/NE602 Rig.asc
Allison KB1GMX alerted us to this rig. Pete and I had been discussing with her various designs for SSB transceivers. I recently completed an NE602 rig.
I really like the KN-Q7 (click on the picture above for a closer look). The T/R switching is much simpler than my use of three DPDT relays. It even seems simpler than the Epiphyte’s switching of the BFO and VFO frequencies. I also like the analog VFO and BFOs inside the 602s, but I think Pete would use an Si5351. The AF amp muting circuit is kind of neat.
I questioned the need for the extra IF amp for the receiver. My rig just uses the two NE602s and an AF amp circuit from the BITX40 module (2N3904 and an LM386), and with this I can hear the band noise floor on 40. But Allison points out that different bandpass and crystal filter have different losses, so with these components some additional gain may have been necessary. YMMV.
I’ve been hearing about this rig for many years. It first appeared in the September 1994 issue of QRPp, the journal of the NORCAL QRP club. A condensed version of that article appeared in SPRAT 81 (Winter 94-95). The designer is Derry Spittle VE7QK from Vancouver, British Columbia.
The name always puzzled me. Here is the explanation: It started with the Neophyte: A very simple direct conversion receiver that many of us built. The Neophyte was mostly an NE602 and an LM386. In the Epiphyte, a crystal filter and a second NE602 were added, turning the Neophytes into a superhet receiver and — with some additional circuitry — an SSB transceiver. The Oxford English Dictionary reportedly defines an Epiphyte as “a plant that grows on another plant”(see picture below). The Epiphyte grew out of the Neophyte.
And this plant grew in British Columbia, which seems — like Australia and New Zealand — to be fertile ground for simple phone rigs. I’m pretty sure the “Wee Willy” DSB rig also came out of BC, and it may have had a similar purpose: allowing for portable contact with the BC Public Service Net on 75 Meters.
There are many features of the Epiphyte that I like: There is a simple 455 kHz filter and a ceramic resonator BFO/Carrier oscillator. The original design featured a VXO-like circuit using a ceramic resonator at 4.19 MHz. And it ran off AA batteries (as did the NE602 DSB rig I took to the Dominican Republic).
Unlike my NE602 rig, the Epiphyte made an artful use of the fact that NE602’s can be set up to have TWO inputs and TWO outputs. Where I used DPDT relays to switch inputs and outputs from both NE602s, OM Spittle left all the inputs and outputs connected, and simply switched the VFO and BFO signals. Ingenious.
There were updates and improvements. The Epiphyte 2 and 3 featured increased power out (5 watts vs. 1 Watt). Version 3 has an IRF-510 in the final, driven by a CA3020A chip. That chip is capable of 70 db gain. Wow.
In 1996 NORCAL and G-QRP donated 50 EP-2 kits to radio amateurs in third world countries. Very nice.
In 2000 NORCAL did a kit of the EP-3 — it sold out in 24 hours. Here is a nice article on the EP-3:
http://www.norcalqrp.org/files/Epiphyte3Mnl.pdf
And above we have a video from Japan of an EP-3 in action.
Pete would call this a DiFX: a transceiver that is Different from a BITX. This started with my effort to get an Si5351 working with a little 1 inch square OLED screen. Tom Hall AK2B helped me with the software (thanks Tom). Once I got that done, I figured I could build a simple receiver with a homebrew 11 MHz crystal filter, two NE602 chips, and an LM386 AF amplifier. That was working great, then Pete told me to turn it into a transceiver. I used some of Pete’s boards (thanks Pete).
The Epiphyte transceivers also use two NE602’s, but they ingeniously switch the BFO and VFO between the two chips. I didn’t switch the oscillators — instead I switched the inputs and outputs of the two chips using two DPDT relays (thanks Jim). A third DPDT relay switches the antenna between T and R, and turns on and off the PA stage and the AF amplifier.
This is a DIFX, but there is some BITX circuitry in there. The power amplifier stages are right out of the BITX Module, as is the AF amplifier (thank again Farhan).
The only real problem I ran into had to do with the very low power out of the NE602 VFO mixer on transmit and the impedance matching between the NE602 and the PA chain. I had to increase the gain on the first RF amp (pre-driver) using ideas from Steve Weber’s 40 meter SSB CW QST contest rig (thanks Steve). I experimented with various connections between the NE602 and the BP filter. Finally I got it going.
The heat sink on this one is different too: it is just the chassis. The IRF 510 is bolted (insulated) to the aluminum box.
I fired it up this afternoon and in spite of horrible conditions on 40, quickly had a nice rag chew with KJ4ZMV in Indiana. I haven’t even built a mic amp yet! I am running the D-104 right into the NE602 balance modulator. There are no signs of unwanted modulation or spurs.
FB! TRGHS! VIVE LA DIFFERENCE!
SolderSmoke Podcast #195 is available. Link appears below (scroll down)
We’ve got a problem: Pete Juliano and the QRP Hall of Fame 🙁 PLEASE HELP!
That’s a beautiful graph, don’t you think? In addition to the very pleasing results, I liked Bob’s methods: the “by hand” collection of the data points using an AD9850, a ‘scope and a notepad (see below); the filing down of ceramic disc caps; the use of nail polish hardener — all this adds a definite artisanal element to this project and puts more soul in the new machine.
Like Bob, I too kind of bailed out on the Q calculation when I was doing this. But as I recall there is a variation on the G3UUR method that yields this parameter too, right?
Bill:
I’m reluctant to share this with you because the results appear too
good. I’ve attached a graph showing my four crystal, 11Mhz
filter measurements. The graphed points are read values using my
AD9850 DDS VFO feeding to a TEK scope.
A TIA amp was used for input and output of the filter. My development software was the Steder-Hardcastle software as presented in November 2009, QEX.
I am now practiced in the black art of filing off the tops of disc
ceramic caps to “adjust” their values. This black art also involves
the mysterious qualities of Sally Hansen Nail Hardener.
Let me share my method for developing the filter.
I’ve built CW filters before but this was my first effort at SSB
bandwidth which is less forgiving.
The G3UUR oscillator method (see page 3.19 in EMRFD) is a simple and
effective filter design method. but it does not provide a measure for
crystal Q, a value which impacts filter insertion loss.
So I decided to choose crystals with proven pedigree. These were 11 Mhz crystals from Mouser, part number 20-HCA1100-S. A lot of ten costs $5. These were the crystals selected by Jim Kortge, K8IQY, for use in his 2N2/20 rig.
For software, I use the Dishal package that can be downloaded from the ARRL and other sources. This package was the basis for the
Steder-Hardcastle article in November 2009, QEX. The “Xtal” pull down
menu provides entries for an individual G3UUR oscillator.
Simply put, all critical filter input values are generated by reading
the change in crystal frequency as an additional capacitor is added
into the oscillator circuit.
I suggest starting with a 4 crystal filter. Only two capacitor values
were required for my filter. Five capacitors were required–two series
and three shunt.
Start by reading the “open switch” frequency for each crystal. Sort
the crystals into increasing frequency order and choose the four with
the most narrow frequency span.
Using the pull down menu measure the individual crystal measures for
Lm, Cm, and series frequency. Average these across the four crystals.
The Lm or Cm and series frequency are placed into the Dishal software
main menu. Also enter the average Cp which is the measured capacitance across the crystal leads.
Finish up the main menu entries by entering the number of crystals (4), and the desired bandwidth–generally 2.4 to 2.9 Khz. Finally enter the acceptable ripple, which is often 0.1db.
Let the software calculate the filter values. Expect some odd
capacitance values. By changing the filter bandwidth–say from 2.4 Khz to 2.35 Khz I can move one of the capacitance values to a standard value.
The software also displays the input and output impedance. If the
filter is centered between two TIA amps, this filter impedance must be
transformed to 50 ohms in the amps. This provides the transformer
winding ratios.
The Dishal software has always given me good results. But I haven’t
compared its results to Ladpac–especially GPLA.
Bob -N7SUR
It happened again today. Conditions were good and I was BOOMING into the NYC area on 40 meters. 40 over. Everyone liked the signal and said it sounded great. Except for one anonymous grump who chimed in to say that I was “9 kc wide.” I imagine he was basing this on a quick look at his super-dooper SDR waterfall, without any consideration of signal strength or the characteristics of his own receiver. Sigh. The Waterfall Police had struck again.
OM W8JI gives a great description of the pitfalls of this kind of “you’re-too-wide-because-my waterfall-says-so” reasoning. Check it out. And keep it handy in preparation for your next encounter with the 40 meter Waterfall Police.
https://www.w8ji.com/checking_bandwidth_with_receiver.htm
Pete WB9FLW alerted us to the work of Peter DK7IH, a very talented homebrewer who recently followed the lead of Pete N6QW in building some really small SSB transceivers. Here is his Micro QRP SSB rig:
You can see more of his fine work here:
https://radiotransmitter.wordpress.com/
https://www.qrz.com/db/DK7IH
Have you guys noticed how many Peters there are among homebrewers, especially among SSB homebrewers? Just from recent mentions on this blog:
Pete Juliano N6QW
Pete WB9FLW
Peter Parker VK3YE
Peter DL3PB
Peter W1UO
Peter GW4ZUA
Peter G6GNR
Peter VK2EMU
Peter VK2TPM
Peter HA5RXZ
Peter DL3JIN
SolderSmoke Podcast #192 is available:
http://soldersmoke.com/soldersmoke192.mp3
Farhan alerted us to a very interesting presentation on the history of single sideband: It was in episode 81 of Bob Heil’s “Ham Nation” show. It starts at minute 22:
https://www.youtube.com/watch?v=xSF0WBdK1IQ&feature=youtu.be
Two things really caught my attention:
— Note how OM Carson, way back in 1915, had figured out how to get rid of the carrier, but needed some way of eliminating the unnecessary sideband. He did it by using his antenna tuner as a filter. FB OM!
— In the early days of SSB, when it was an exciting new technology, hams had regular “sideband dinners.” At these events an award was presented. Kind of like an Oscar or an Emmy I guess. The award was the “Sideband Suzy” (see above). Kind of a classic figure… but half of Suzy was missing!
That, my friends is an extreme example of what we mean when we use the word “rig.” This magnificent machine sent Pete’s melodious voice across the mighty Pacific several times during the recent CQ WW contest.
Pete wrote to Jun:
