Raspbian USB disconnects and USB keyboards

I was looking for a solution to a USB soundcard that would randomly disconnect from my Raspberry Pi (it’s a Signalink USB soundcard/radio interface for amateur radio). Turns out, the issue I was seeing was caused by RF getting into somewhere when my radio was keying up – when I moved the HT radio further away from the Pi (only as far as the length of the connecting cable, a few feet) then my problems stopped. This could probably be avoided better with some snap-on ferrite beads.

In this post (and others), there is a suggestion about adding dwc_otg.speed=1 to your /boot/cmdline.txt. I tried this for a while and this didn’t seem to make any difference to my USB disconnects for this particular USB soundcard. It did however stop a USB keyboard from being recognized (a Gear Head Mini USB). Remembering I had added this param and then removing it solved my keyboard issue.

Lesson learnt: if trying out solutions to problems by trial and error, if something doesn’t work, remember to remove it afterwards in case it breaks other stuff 🙂

 

Tuning Direwolf packet parameters for use with HT radios

Direwolf has a  number of tuneable properties – by default after installing and configuring the basics, your callsign etc, it normally ‘just works’. I’ve been experimenting setting up Direwolf and PiLinBPQ as a BBS on a Raspberry Pi, and using axcall or LinPac on another Pi with Direwolf to call into the BBS. I have a Rigblaster and Yaesu FT60 HT on one Pi, and a SignaLink and Wouxun HT on the other radio.

The problem I ran into was that either Pi seems to work ok to call into a Packet node nearby, but calling from one Pi to the other was pretty unreliable in terms of not decoding packets, and the number of retries before a packet would get received and decoded. I tried a number of options, and feedback in the Yahoo Groups Direwolf group give me some suggestions to try. The tuning parameters are all covered in the excellent Direwolf docs.

Possible problems with using HTs for packet could include a combination of these factors:

  • Receive audio too loud into Direwolf (aim for around 50)
  • Receive audio not loud enough – adjust soundcard volume (Rigblaster/Signalink) and alsamixer levels
  • The radio doesn’t transition quick enough between transmit and receive (can be fixed by adding DWAIT)
  • Latency from the soundcard audio either in receive or transmit – can be fixed by adding TXDELAY for delay at the start of the transmission, and/or TXTAIL for a delay at the end

The settings I finally wound up with on both Pis that worked for me are:

DWAIT 20

TXDELAY 60

TXTAIL 60

 

 

 

Making a Raspberry Pi based Amateur Radio WSPR Go-Kit

A couple of weeks back I got a Raspberry Pi setup with TAPR’s QRPi low pass/band pass filter board, running WsprryPi, providing a low-cost, low power, WSPR transmitter. For my first attempt everything was hooked together temporarily. I wanted to see if I could get the Pi and a USB battery pack packaged in a weatherproof box with connectors for an antenna so I can take it out and leave it outside in the yard transmitting for a day (or at least as long as my USB battery pack will last for – I currently have 10000mAh USB pack that seems to last at least 8 hrs or so with a 1/4 charge remaining).

This is what my second attempt looks like:

  • 10″ Tactical Weatherproof Equipment Case, from MCM
  • on the sides of the case I drilled through and added banana jacks to connect each side of the wire dipole on the outside of the case. On the inside I soldered a short wire from each jack which connects into the QRPi board connector
  • the battery is a 10000mAh Anker USB battery pack (hard to see in the photo but it’s at the back of the case)
  • battery and the Pi are velcro’d into the case

There’s plenty of free space on the inside of this case, I probably could have gone with a smaller one. I have some other Pi related projects in mind though, so will probably use this size case again for another Pi + Packet radio related project.

So, the results, how does it perform. Well the first couple of weekends I ran this I didn’t get any spots, so I suspected something was up with my antenna connections. I had already run with the same wire dipole connected directly to the board, so I knew it would radiate a signal. The first test I roughly cut the wire dipole using the 468/MHz formula but I didn’t bother to check it on an antenna analyzer. So next step, where is the resonant point of this antenna?

Hooking up my YouKits analyzer, the center line is showing 14.1 MHz, but the low point is clearly to the left, so the wire is too long:

 

 

 

 

 

Trimming off about 6 inches from each side got us more in the ballpark – now we’re looking good: wp-1464669440438.jpg

Getting the Pi back out in the yard, I hung the antenna from a couple of bushes at approx 5ft off the ground. This is very low for a dipole on 20m, but I know from previously running with this antenna at the same height that I still got spots all the way out to the East coast, so I’m not too worried about getting it up higher.

To keep things in perspective, remember we’ve only running 100mW, so we’re running low power, definitely QRP by any definition. At this low power, any spot is a good spot for me.

So, what about the spots? I didn’t get anything all afternoon, but then I got a number of spots from Texas, so if anything, there was a good propagation path between California and Texas early evening, but the spots dropped out shortly after sunset:

Since every day on HF is different, this is definitely an ongoing experiment 🙂 So far, this works pretty good and I’m happy with the setup!

NOTE: to transmit on Amateur Radio frequencies requires an Amateur Radio license, with the appropriate privileges for the frequency/band you are using. Also, per recommendations for the WsprryPi software, ensure you are using appropriate low pass/band pass filters so the Pi is not generating unexpected harmonics on other frequencies. 

For future reference, current solar conditions (from http://www.hamqsl.com/solar3.html ):

SFI=86, SN=25, A11, K=2. 20m day/night was fair

Common dipole antenna dimensions for Amateur Radio HF bands

When building your own antennas, you can either look up a design or dimensions online or from book, or you can use a common formula. 468/MHz is commonly used to calculate the total length of a wire dipole in feet. I had a few dimensions jotted down for reference, but it’s easy to stick the formula into a spreadsheet and just enter the freqs you’re interested in. For example, here’s a Google Doc with the common HF bands:

I hope I’ve got these right, but if not, drop me a comment below.

Here’s a link direct to the Google Sheet if you’d like to bookmark it: https://docs.google.com/spreadsheets/d/1d3Mmnyzvj8AG1lFPd59RSMJhE9ihbE2M4w5noDzyedo/edit?usp=sharing