A NEW Audio Board for The Planker

After some discussions with a fellow named Francis Flynn on Facebook about better ways to mute the RX audo during TX, and then subsequently looking at my original design and trying to come up with a better way to implement CW and Digital Modes (I'm diggin' JT65!), I decided to scrap the thing and start over.  It took a few days, but I installed the board diagrammed above in the rig today and it's working great!

As I mentioned in the post from earlier this week, my Linux PC does not have a serial port to support PTT, and I had no luck using a USB dongle, so I'd built a simple VOX circuit to provide T/R control when operating digital modes.  This worked well (I kept the VOX board), but, since the microphone and soundcard audio used the same amp, I had to remember to unplug the mic when operating digital, else my digital tones would be accompanied with my occasional bursts of unsavory language "What the F*&^ is this A$$ H01E doing!??"  Obviously the FCC frowns on that sort of thing, and lacking self control, I felt that addressing it through technology was the prudent course of action.

So - This board, well, to me, it's simpler:  The Audio PA uses the same LM380 as the original board, but the 2N3904 preamp has been replaced by one section of an TL082 Op-Amp (I know, the schematic says LM324 - it's pin compatible..)  This amp and the '380 are powered in both RX and TX modes, as they are necessary to provide an audible CW sidetone.  However, the 4066 CMOS gate at its input disconnects it from the product detector/balanced modulator while in transmit.

The microphone audio and digital audio sources each have their own TL084 amps, and the gain of each is optimized for the two applications.  The remaining three gates of the 4066 are configured to, when the microphone PTT initiates transmit, gate mic audio to the balanced modulator, and, when the Digital Vox is active (low) the microphone is switched out of the circuit and digital audio gets routed to the modulator.

Not shown is the CW sidetone interface (forgot to document it...) - The local audio (speaker) is injected at the "hot" side of the volume control, while the audio to be transmitted is sent directly to the microphone amplifier.  In this case, one still must be congniscent of speaking while sending CW - the mic will be live - but I don't worry about that as I seldom say anything while talking with my hands.

I'm pretty happy with the performance of this board: Microphone and digital audio sound perfect, while CW still needs a bit of work as it is only driving the rig to 3-4 watts.  This should be an easy fix - The CW filter module terminates in an attenuator network, so I just need to pare back a few dB worth of attenuation.  Souldn't be a big deal.

Anyway, Thanks Francis Flynn!  Your comments got the grey matter stimulated, and this board is a definite improvement over the original.  Also, thanks for sending the link to the W7EL rig - That was a nice walk down memory lane, and seeing how Roy used the FET as a mute gate gave me the idea to use the 4066 CMOS chips for audio switching and routing.  FB OM!!

73 and go make something!!!

Steve N8NM

Planker Circuit Description: Audio Module

I'm cleaning up some of the Planker docs in preparation for an article for QRP Quarterly, and will pop them up here with brief descriptions and notes.  Since I like to start every project from the "back end", I'll start here with the audio board.

Planker audio board Rev 1.0

This board is fairly straightforward: On the left side is the transmit audio circuit, starting with  the mic connector,followed by a two-stage mic amp using a J310 and 2n3904.  

Toward the center, you'll see the common connection to the Product Detector/Balanced Modulator board, to the right of which is a 2N3904 amp driving an LM380-8.  To produce an audible CW sidetone in the speaker/phones, the output from the sidetone circuit is introduced at the top of the AF Gain pot. The LM380 is powered on in both transmit and receive, but power to the driver transistor is disconnected while in transmit.  Nonetheless, enough voice audio leaked through to the 380 that it was clearly audible, so I added a 2N7000 to shunt the output from the driver to ground while transmitting. Even though this places the 1uF at the top of the volume control at ground potential, enough sidetone still get's throug to be useful, and I see no reason to improve on it.

The circuit above delivers enough mic audio (using a cheap condensor) to drive the rig to full output, and the coupling caps were chosen to minimize low-frequency response.  All stations worked so far have made positive comments about the audio, so I'm calling that a success!

The receive audio will easily deliver a couple of watts to an 8 Ohm speaker, and is pleasant to listen to.

The next board that I plan to document is an adjunct to this one, providing separate inputs for a computer soundcard interface for digital modes, and a functional VOX circuit.  Maybe I'll have that up tomorrow, maybe not... We set goals, not deadlines!

73- Steve N8NM 

Planker Circuit Description: Audio Module

I'm cleaning up some of the Planker docs in preparation for an article for QRP Quarterly, and will pop them up here with brief descriptions and notes.  Since I like to start every project from the "back end", I'll start here with the audio board.

Planker audio board Rev 1.0

This board is fairly straightforward: On the left side is the transmit audio circuit, starting with  the mic connector,followed by a two-stage mic amp using a J310 and 2n3904.  

Toward the center, you'll see the common connection to the Product Detector/Balanced Modulator board, to the right of which is a 2N3904 amp driving an LM380-8.  To produce an audible CW sidetone in the speaker/phones, the output from the sidetone circuit is introduced at the top of the AF Gain pot. The LM380 is powered on in both transmit and receive, but power to the driver transistor is disconnected while in transmit.  Nonetheless, enough voice audio leaked through to the 380 that it was clearly audible, so I added a 2N7000 to shunt the output from the driver to ground while transmitting. Even though this places the 1uF at the top of the volume control at ground potential, enough sidetone still get's throug to be useful, and I see no reason to improve on it.

The circuit above delivers enough mic audio (using a cheap condensor) to drive the rig to full output, and the coupling caps were chosen to minimize low-frequency response.  All stations worked so far have made positive comments about the audio, so I'm calling that a success!

The receive audio will easily deliver a couple of watts to an 8 Ohm speaker, and is pleasant to listen to.

The next board that I plan to document is an adjunct to this one, providing separate inputs for a computer soundcard interface for digital modes, and a functional VOX circuit.  Maybe I'll have that up tomorrow, maybe not... We set goals, not deadlines!

73- Steve N8NM 

The Planker goes digital.

I'm finding CW and SSB activity on 60m is pretty limited, but "channel 3" is hopping with JT65 signals throughout the evening.  So, what the heck - might as well jump in!

Running WSJT-X on my Linux laptop, I ran into a challenge in getting the rig to transmit:  The laptop doesn't have a serial port, and it recognizes my USB<>serial dongle as ACM0.  Unfortunately, WSJT-X only supports Sn and USBn devices, and while I'm sure there's a symlink work around, it was simply easier for me to deal with it in hardware by adding a simple VOX circuit.  So, I kludged one together and made my first JT65 QSO!

This is going to lead to a redesign of the RX/TX audio board, not only to incorporate VOX, but to provide separate RX & TX audio I/O (with isolation transformers) for the PC interface. I'd planned on redesigning the board anyway, there are a few things on the current board that I'm not entirely happy with, so it's not a big deal.  I'll put up the diagram and some pics once I have it operational.

73!  Steve N8NM 

On the air with The Planker!

After a few months of casually melting solder, The Planker made it's maiden voyage yesterday afternoon when I enjoyed a brief chat with Lauren, WD5HIO, in central Ohio.  I'm very pleased to say that nothing exploded or otherwise malfunctioned, and Lauren commented that I had a very nice signal with "excellent audio".  Yes!!!!

So, what is all this Planker nonsense?  It stems from an email exchange I had with Pete, N6QW, earlier this summer.  Pete had been working on a rig covering 40 and 60m, 60 being the only amateur HF band that I'd never operated on.  At that time, I was putting the finshing touches on my "all-band" HF rig, and was getting a little frustrated because I'd built myself into a corner and was running out of space on the chassis.

Since I'd never been on 60, I became intrigued by Pete's rig, and thought that it'd be a good "rainy day" project for the summer if I kept it simple enough.  At the same time, I wanted to avoid my bad habit of establishing a "form factor" for the chassis and trying to fit all of the circuitry within those confines, and, since I've had this 24" chunk of pine board kicking around for years, everything just clicked.  As N2CQR says: TRGHS - The Radio Gods Have Spoken.

The architecture of the rig is very similar to Farhan's (VU2ESE) BITX: It's a single-conversion superhet SSB transceiver using bilateral IF stages.

Each individual stage is built as a module, the input and output impedance of which is 50 Ohms.

 From left to right: The small board is a 7 pole LPF, behind which is the power amplifier (based on Farhan's RF386) and forward of which is the antenna relay and pre-mixer bilateral amplifier (based on W7ZOI's "TIA" circuit, using six 2N3904s.)

Moving to the back of the plank, the two upright transistors to the right of the PA handle the T/R switching.  These are actually some bogus 2SC1969s that I bought on Ebay; they don't work at RF, but are adequate DC switches.

In front of the switch board is the mixer, a homebrew DBM using "matched" 1N4148 diodes.  Forward of the mixer is the 20 MHz IF amplifier, another W7ZOI "TIA" module.

Underneath the crooked shield (right of the switch/mixer/IF amps) is the crystal filter, which I made from seven 20MHz CPU crystals.  The bandwidth of the filter was set, largely experimentally, to 2.4 KHz, by selecting the appropriate shunt capacitors.  I first modeled it on the computer and then adjusted it using a spectrum analyzer.

Moving to the right again, what you can't see (because it's behind the display panel) is the product detector/balanced modulator, which is the 2 diode single-balanced circuit stolen from the BITX.

The display panel is, well, a display panel... A small board behind the 7 segment display contains the 4511 driver IC, and the pushbutton steps through the five 60m channels and WWV at 5 MHz.

The next board is the synthesizer and MCU.  I'm using an Arduino Pro-Mini to control a SI-5351, which provides the local and beat frequency signal sources.  And no, phase noise isn't an issue.

Then, finally (if you've made it this far) is the transmit and receive audio board.  The transmit audio amplifier is a pair of cascaded JFETS (J310), and the receive side uses a 2N3904 to drive an LM380-8.

If you're wondering about the handle:  I added that to be funny - it technically makes it portable.

 The transmitter outputs, eh, about 7-8 watts PEP using an IRF-510, though I just received a package from RF Parts and will be changing that to a proper RF device (Mitsubishi RD16HHF1) shortly.

The picture shown above is how the rig looked at the time of my QSO with WD5HIO.  Later in the evening, I tidied-up the wiring a bit, but the circuitry remains the same.

One thing that I still need to do is implement a tone generator for CW:  When I started this project, I wasn't aware that the FCC had revised the regulations allowing this mode, so it wasn't on my radar.  That'll be handled in software, using the Arduino to generate the tone and cleaning it up with a RC LPF network.

Otherwise, there you have it.  I'm debating whether to mount it on a chassis and put it in an enclosure, or leave it as it is... The crudeness of the breadboard is growing on me!

Update: Added a few lines of code, LC LPF and an input line for the CW Keyer.  The Arduino did a marvelous job - told it the output a 1.5KHz square wave, and it did exactly that.  So, with the proper attenuators, I'm feeding the tone into the mic circuit to put the CW signal right in the center of the channel, and a little back into the RX amp to provide a side-tone.  Works great!

73 - Steve

Back to the shack!

We've had some abnormally nice weather here in the Detroit area over the past few weeks, so I haven't been spending much time melting solder in the dungeon.

I did discover something new a few weeks ago: About 10 miles from me is an outlet of the computer chain Micro Center.  What I didn't know was that they started carrying Arduino stuff and other electronic parts and supplies - cool!  Better yet is that they've been running some sales: Uno boards for $5.00 and Pro Minis for $3.99.  So, I've stocked-up!

With my newly acquired boards, I've done some clean-up on the all-band rig and planker: The haywired breadboards have been replaced by (relatively) neatly-wired Nano and Pro Mini controllers on perf board, so the rigs look a little less scary.

I ran into something noteworthy when installing the Pro Mini in the Planker:  The "Raw" power supply input on the Arduino board is claimed to be good for up to 16 Volts, but nope... Good thing I bought extras, because the first one smoked at 12 Volts!  Not a big deal, I've had a bunch of LM78L08 regulators that I've been looking for something to do with.

About the only other thing I've accomplished is building the PA for the Planker.  Last year, I'd built Farhan's RF-386 amplifier, and abandoned it because I couldn't get it to behave.  The design looks solid, so I figured it had to be something in my layout - possibly the Manhattan construction?  I dunno.  Anyway, built another version and it's very well behaved and delivers a solid 5+ Watts on 60M with about -20dB/m input.  Cool!  Getting close to being QRV!

Right now, my bench looks like a disaster, so I'll post some updated pics after straightening things up a bit.  Hot dogs are tasty, but nobody wants to see the slaughterhouse.

73 de Steve N8NM