Sunday, December 29, 2019

CQ CQ CQ!

I've been trying to figure out what transmitter I want to build next - spending a lot of time modeling it in LTSpice and trying to figure out how it will work.

But I wanted to actually get on the air.  I've been practicing my Morse a bit lately, so thought I should just see what I could throw together.

The things I had that were is somewhat working state (or could be made to be...) were my direct conversion receiver and a NS-40 CW transmitter.  First thing was I needed a way to connect them.  I wanted to make something with a relay and have the Arduino I was using for VFO switch it.  But I didn't have any lying around I could find.  But I did find a DPDT switch, so I made a little thing in a nuts tin


When the switch is in the transmit position, the transmitter is routed into the dummy load (I didn't want risk accidentally powering it with no load).  It doesn't give good isolation - the receiver screams when I key - but seems to somewhat do the trick.  Here it is spread over the table:




I tried to send with my homemade key but wasn't getting any responses.  I wasn't sure if it was my antenna setup, the low power, or my bad sending.  So I thought I should try using a keyer.  I found the k3ng Arduino keyer which looked like it would do the trick.


It went together pretty quickly.  The ns-40 had some interface circuits on the site that worked well.  Only thing is the circuit works on active being low, so I had to switch high and low in the software.

It was getting late, so I wanted to just do a quick test.  The keyer has a CW keyboard mode, so I sent CQ from there.  I didn't get anyone coming back.  But I did get spots on the reverse beacon network!

Very exciting!

Knock-off Nano Note

Quick note about how I got the random Nano I found in my junk box working.


  1. It wasn't showing up in list when I plugged it in.  Found answer here to install new drivers for OSX so it could see the board (links to this github repo).
  2. The upload wasn't working.  Kept getting a "not in sync" error.  Found answer here.  Turns out I needed to use the "old" bootloader under the "processor" section in the Arduino IDE.

Saturday, December 14, 2019

Decoding CW from WebSDR with Fldigi on OSX

I got the urge to play around with WebSDR this morning.  After a bit, I got the further urge to try to use Fldigi with it.  I figured I already had Loopback from some music things I do, so I'd have a go.

Worked really well.  I just have a simple setup where I have the output of Safari into an output channel, and then am monitoring that.  Then in Fldigi, under "PortAudio" I have it use "Loopback Audio" for capture.  

Couple of notes:
  • It didn't show up as a soundcard at first - probably want to setup Loopback before opening Fldigi
  • It was initially doing some mode *not* cw.  It looked like I was supposed to change that in the lower left (where it said the mode), but you actually go up in the menu under "Op Mode"

Sunday, December 8, 2019

Activities - 8dec19

Have been trying out WSPR with the Si5351 this morning.  I am using the sketch here.

Couple of learnings:

  • Can list serial ports on a mac with `ls /dev/tty.*`
  • The "serial" library it is talking about is pyserial (not serial).  I used pip to install the first and was confused why it wasn't working (the gist wasn't returning the error)
  • You can't program the arduino while the script is running (as they both use the serial port)
  • This script is set to run every 10 minutes.  Confused me why it wasn't working, but then it suddenly did.

No luck so far, though it isn't helping that the WSPRnet website doesn't seem to load very well.

So then I changed tact and decided to get my DC receiver back up and running.  Before even turning the VFO, I was getting in some local religious broadcast as strong and clear as could be.  Possibly need a stronger band pass filter...  Turning on the VFO, it was working, though I couldn't really hear anyone (some digital modes and someone speaking maybe spanish).  But it was 11am, so probably not the best time for 40m.  May try again tonight.

One thing I noticed this morning: powering my Arduino from my Mac works great.  Powering it from any power plug I can find is very noisy on the Si5351 output (either the 12v or usb).  Strange.

Friday, December 6, 2019

Si5351 VFO

A few years ago I made a VFO with an AD9850.  Now I changed it to be a Si5351.  I am using the same awesome setup from Jan again - he updated his for this chip.  I even still had the lcd/wires on a breadboard, so I just had to wire it back to the Arduino!

Several things I ran into:

  • I had a hard time convincing it to use the right libraries.  I had some out of date versions of some, and it kept wanting to use the Adafruit library instead of the Etherkit one.  I finally deleted the Adafruit one to get it to work.  
  • It didn't work at first, and I thought I had fried another Si5351.  I put another one in, and it didn't work, either.  I suspected my wiring, but then tried the Adafruit sketch I had made, and it worked.  Turns out it was the i2c address - I had to update it to be 0x60.
  • The calibration was off.  It was actually off by about the same as the default BFO frequency, so I went down all sorts of rabbit holes trying to figure out why it was adding that in (which it wasn't).  Finally figured it was actually the calibration, and ran the calibration tool the Etherkit library has (which is kind of awesome) to fix it.
  • I can't seem to get the Arduino to acknowledge changes I make to the 'definitions.h' file.  Or at least when I change those and reupload, they don't seem to have taken effect.  Maybe something with the eeprom, not sure.  I ended up changing the calibration in the .ino file for now while I look some more.
I had originally planned to use this with a Tayloe mixer.  But I screwed up my mouser order and the IC I got is the *smallest* I think it could possibly be.  It was one of the last things I put in that order and I was pretty tired at that point, and I swear I checked that it was larger than that...  Anyway, I decided I would try this with the EMRFD direct conversion receiver I had made a while back, but it is now pouring outside so don't want to go out in the rain to pull in then feedline.  So that is a project for another night.

Sunday, December 1, 2019

Power in LTSpice

A quick note of something I found tonight.  I've been playing with some IRF510 designs, and for one of them, it looked like I was getting crazy good efficiency.  Like, not possible good.  I was evaluating this by running a transient simulation and then option-clicking (in osx) on the load resistor and on the IRF510 to get the power on those components, and then control-clicking on the labels in the graph to see the average power.  The power dissipation on the IRF510 was *way* too low (like 290mw when the output overall was 3.8w, which would be like 93%).  Turns out that when I was doing this, the current was negative some of the time, which was showing I guess as negative power.  So the average was much lower.  Wacky.  I guess I'm still learning this program.

Friday, November 29, 2019

More with the 74AC240

Continued this morning with the 74AC240.  I am skipping input 0 (as it seems to be shorted), so have output 1 going into all the other inputs, and have all those outputs connected together and going through my low pass filter into a dummy load

Am measuring 32.4 V pp, which should be somewhere around 250mw.  Which seems about right.  Some people claim to get half a watt out, but that is driving the chip with higher voltage, and I don't want to mess with that right now.

I am noticing that the voltage of the signal out seems to jump up and down a bit.  Not sure if that is due to the stray wires of my test setup, the fact I'm getting my power from the Arduio's 5v pin at the moment, or something else.  But will work for the moment.

Thursday, November 28, 2019

Long time, no post

It has apparently been something like 2.5 years since I've done anything here.  I wasn't having any time to work on electronics, and I really needed to do something non-technical outside of work anyway, so I put away my gear for a while.  But my hobby interests seem to be cyclical, and I'm firing up the iron again.

This past week I've done a few things.  First, I cleared the desk and got the gear back out.  Then I've been reading over the blog to try to remember where I got to the last time I worked on anything.

I got out my NS-40 and powered it up.  Still works great.  So I got out my version of the NS-40 circuit and did some measurements on it for a bit.  But at some point I keyed it and there was a spark, and it doesn't seem to be working anymore.  Whoops.

Then a few nights later I got out another Si5351 and wired it up with my Arduino (I blew the last Si5351 at some point).  I downloaded the sample Adafruit library onto it except changing clk0 to put out something near 7Mhz (I think it is 6.8 something, but the clock designer software doesn't run on Mac and I'm being lazy).  Looked at that output into the dummy load a bit (through the LPF network).

Today I tried the 74AC240 experiment again.  I'm again trying to wire the outputs in parallel to drive more.  Taking my cues from this page (the TTL version).  But going slowly, I tried just driving a single input and going into a high load (20k or something).  Seemed to work on the breadboard, but crazy ugly.  So moved it to a perf board.  I tested all my connections as I went, and found that the 1M to ground link was showing 25 ohms.  Super confused.  Finally moved from input 0 to input 1, and it seemed to work.  So maybe I accidentally blew up something on this 74AC240 the last time I was playing with it.

Next time I'm going to try to parallel the remaining lines (skipping input 0, as it seems sad now) and seeing what that looks like into the dummy load.

As an aside, I'm going to look into ordering some proper mosfet drivers.  The '240 experiment was kind of to see about using it to drive an IRF510, but I think even if I get the power out, I'd need to put a transformer between it and the IRF510 to get the voltage I want.  I *think* I can get some gate drivers that would work at 40m for not much, and that would be far easier and probably more predictable that what I've been playing with so far.