Digital Voice – WB3GCK QRP Amateur Radio

An Extra Helping of Pi

Back in 2020, I started fooling around with digital voice modes. I don’t really use them a lot, but there are a couple of DMR nets I like to check into each week. I recently started using the WPSD software instead of the Pi-Star software I had been using on my two hotspots. Just for the heck of it, I wanted to try upgrading the old Raspberry Pi Zero board on one of them to something more capable.

According to the documentation, WPSD was never written for single core processors, like the old Pi Zero W board in my hotspot. The developer recommends a Pi Zero 2 W or better. While it will work on a single core Pi, it runs slower—I can attest to that. I also noticed that CPU utilization was well over 100% at times. With its quad core processor, the Raspberry Pi Zero 2 W is supposed to be five times faster than its single core predecessor. So, I was hoping to see some general performance improvements and faster boot-ups after this upgrade.

In the interest of full disclosure, I’m no Raspberry Pi expert—far from it. I had to do some online research before launching into this. It looked simple enough; the Pi Zero 2 W board is a drop-in replacement for the original Pi Zero board, so it seemed simple enough.

I bought a Pi Zero 2 W board on Amazon for about $28 (US). The board didn’t have the header pins installed, so I bought some. I bought a box of 40-pin headers for less than $10 (US). I now have more of them than I’ll use in my lifetime.

I took advantage of a rainy Sunday to do the upgrade. First, I needed to install headers on each end of the Raspberry Pi’s GPIO area. The two headers had two rows of five pins each. With the help of an Xacto knife, I snapped the pieces I needed from one of the (many) 40-pin headers I now own. With my aging eyes, I needed to use a lighted magnifying glass to do the soldering. To avoid melting the plastic, I allowed some cooling time after I soldered each pin.

The new Pi Zero 2 W board installed in the hotspot case. The two headers I soldered in are towards the front.
The MMDVM board installed on top of the Raspberry Pi board.
The is the fully-assembled hotspot. This pink section to the right is a piece of paper I used to cover some overly-bright LEDs.

From there, it was a simple task to disassemble the hotspot, swap boards, and put it back together. I estimate that the hardware part of the upgrade took all of about 20 minutes.

I had previously burned the WPSD image on a new Micro SD card, so it was ready to go. When I inserted the card and powered up, the hotspot booted up in a minute or two. After importing a backup WPSD configuration file and doing a couple of tweaks, the hotspot came to life. Success!

The performance improvements were immediately obvious. The dashboard and configuration web pages were much more responsive, and the CPU load typically stays under 20%. It’s been running for a week now, and it’s been rock-solid. Since this upgrade went so smoothly, I went ahead and upgraded my other hotspot. That one was just as easy and is also working great.

For you Raspberry Pi power users out there, this is pretty trivial stuff. Yeah, I guess it is. But it was an easy and fun little project, and I learned a bit about Raspberry Pi boards in the process.

73, Craig WB3GCK

My Hotspot is Back From the Dead

While trying mobile DMR last month, my (relatively) cheap hotspot experienced an untimely death. I set the hotspot aside as a future troubleshooting project.

While the Pi-Zero board appeared to be working, the MMDVM board was dead as a doornail. The OLED display was blank, and the RF side of the hotspot was non-responsive. It also wasn’t interacting with the Internet at all.

The other day, I came across a posting on a Facebook MMDVM page that shed some light on my problem. Some users reported shorts between the internal boards and the aluminum case. This tidbit of information prompted some further investigation.

I removed the top of the case, reinstalled the antenna, and applied power. To my surprise, the hotspot booted up and came back to life.
I let it run for a while without the case to verify that all was well. It was.

I noticed a solder connection that extended a little beyond the edge of the MMDVM board. It appeared to be a power connection, so I suspect that might have been what shorted. I also noticed that a single screw and the header pins are all that secure the top board. I’m guessing the board shifted a small amount while mobile causing the short.

My disassembled hotspot. If you look closely, you can see the electrical tape I added inside the top cover.

To remedy this, I applied electrical tape inside the top cover of the hotspot. Not being 100 percent certain where the problem was, I covered everything except the vent openings. I used a razor blade to trim the tape around the OLED display and antenna jack openings. I also took a file to the little solder blob I had noticed. (I exercised extreme caution in doing this since I’m sometimes prone to creating new problems.)

I reassembled the case and powered it up. Voila! It still worked. Hopefully, this fix will avoid any reoccurrences and make it withstand the vibrations of mobile use.

Although you’re far more likely to find me on CW on the HF bands, I do monitor Brandmeister TG-3142 (Pennsylvania State-wide) on DMR. And, on occasion, I listen to D-Star REF20A.

73, Craig WB3GCK

One of Those Days

Did you ever have one of those days where just about everything seems to go wrong? Yesterday was one of those days for me.

I went out to my daughter’s property yesterday to do a little portable operating. The temperature was in the low 30s (F), with a wind chill temperature down in the low 20s. Earlier in the day, my daughter cautioned me about the strong winds on the hill, my usual operating location. Despite all that, I headed up there anyway. I figured I’d hunker down in the truck and be fine.

As I was putting my vertical on the back of the truck, a strong gust of wind had me starting to regret my decision. I pressed on and got everything set up. The wind was whipping the antenna around, but everything was working fine. So far, so good.

I started on 40M and quickly put a Canadian station in the log. I moved up to 30M and soon worked SKCC Straight Key Month stations, K3Y/5 and K3Y/9.

About that time, I looked in the rear-view mirror and saw my antenna leaning over. The movement from the wind got the top of my antenna stuck in a tree branch. I removed the 20-ft pole and laid it on the ground. I pulled my truck forward a few feet and re-installed the antenna.

My 20-foot Black Widow pole blowing in the wind. It eventually got stuck in a tree branch.

I got back in the truck and fired up the KX3. I spent the next half hour or so trying to reach stations but not being heard. When I went out to the antenna to change bands, I found that I had neglected to connect the ground. Doh!

After correcting my oversight, I had much better luck on 20M. I worked N4CD who was activating a park in Texas. I finished up with an SKCC QSO with a station on the West Coast in Washington.

I was mentally celebrating my coast-to-coast contact until I looked in the mirror and saw my antenna wire flapping in the breeze. Sometime during that last QSO, two sections of my pole had collapsed. That was enough for me; it was time to pack up.

But, as they say on TV, there’s more…

I’ve been using DMR since last year, and I wanted to try going mobile. So, I brought my AnyTone AT-D878UV HT and my hotspot. I had the hotspot in the back seat, using one of the truck’s USB ports and built-in Wi-Fi hotspot. On the drive out, it worked fine. When I started the engine to head home, I noticed that the hotspot’s display was blank. It was connected to the Wi-Fi but not sending or receiving anything.

When I plugged it in at home, I heard a brief snap. So, I think something in there got fried. I suspect there might have been a voltage surge or something when I started the truck. Doh! I guess I’ll have to start learning about Raspberry Pi and try to rebuild it one of these days.

Looking back, I guess it wasn’t a total disaster. Despite Mother Nature and my self-inflicted issues, I still managed to put a few decent contacts in the log. I’ve had worse days, I suppose.

Here’s to better days in the future!

73, Craig WB3GCK

Delving into D-Star

I mentioned in a previous post that I was going to venture into the digital voice modes. Well, I did. I was successful in getting a hotspot up and running, although I did hit a snag along the way.

Yeah, I know; as usual, I’m late to the party on this stuff. Up until now, I’ve had only limited experience with D-Star. Years ago, I used a borrowed Icom ID-1 for 1.2GHz data during a couple of ARES-RACES exercises. I hadn’t used D-Star again until I bought a Kenwood TH-D74A six months ago. Even then, I had only used D-Star to check into a local ARES-RACES net over a nearby D-Star repeater.

My local ARES-RACES group here in Chester County, PA, held one of our monthly Tech Rallies using Zoom. Jim WA3NOA gave an interesting talk on digital voice hotspots. Inspired by Jim’s talk, I went ahead and ordered an MMDVM hotspot from Amazon.

While waiting for the hotspot to arrive, I started studying up. I found lots of great articles and videos out there. W6GPS’s videos on the TH-D74 and D-Star were particularly helpful. Following one of his videos, I picked a hotspot frequency and programmed it into my radio.

The hotspot arrived a few days later. I was immediately struck how tiny this thing is. The instructions that came with it were sparse but clear enough to get started. I was able to successfully connect my laptop to the hotspot’s internal WiFi hotspot and access the Pi-Star configuration screens.

Long story short… Before too long, I had the hotspot connected to my home WiFi network and had the D-Star section configured. I keyed up my radio and received the expected D-Star response from the hotspot. Then, I tried to connect to some reflectors. No matter which one I tried, I got a “reflector is busy” message on my radio. Hmmm…

I sent screenshots of my hotspot configuration screens to Jim, but he didn’t see anything out of wack. I next focused on my WiFi router. I tried connecting the hotspot to other WiFi networks. No joy.

Finally, I went out and checked the status of my D-Star registration. That showed that my registration from way-back-when was missing a Terminal ID. An email exchange with WA3NOA confirmed that was likely my issue. My D-Star registration password no longer worked, so I reached out to Jim W3BIF, the admin for the W3EOC D-Star repeater.

My faulty D-Star registration. This turned out to be the cause of my hotspot problem.

The next morning I had an email from W3BIF saying he had straightened out my incomplete D-Star registration. Within minutes, I was connected to a reflector and having a QSO with a ham in Georgia. He reported that my audio sounded great.

I used to be one of those guys who thought this stuff isn’t real ham radio. I guess I’ve come around a bit. I won’t be giving up CW any time soon, but I will admit it has been fun playing around with this new (to me)mode over the past week.

Over time, I might branch out to try some other digital voice modes. In the meantime, you can often find me monitoring D-Star reflector, REF20A or REF30C.

73, Craig WB3GCK

My Activities of Late

I haven’t been posting much here lately. The COVID-19 pandemic and other family obligations have been cutting into my ham radio activities. Nevertheless, I do have a few projects in the works.

A few weeks ago, I started another project in my ongoing series of speaker wire antennas. This one will be a variant of the bi-square antenna. This antenna has the potential to be a little more field-friendly than the delta loop I tested last month. It’s all built; I just need to get out somewhere to set it up and see how it works.

I’ll file my next project under the category of Old Dogs/New Tricks. Back in December, I bought a Kenwood TH-D74a HT. That gave me the ability to reach a nearby D-Star repeater. This week, I purchased an MMDVM hotspot to go along with it. I plan to spend some time in the coming days getting it set up. I’m hoping to be able to eventually connect to the DMR talk groups used by my ARRL section and local ARES-RACES groups. Fortunately, my local group has some experienced hotspot users I can consult if I run into any snags. Wish me luck.

Sadly, our camping season with our little QRP Camper is off to a late start. State park campgrounds in our area have been closed due to pandemic. We have reservations at a state park in Maryland next month, however, and it looks that might be our first trip of the year. I’m looking forward to a little QRP-portable operating from the camper.

My local QRP club has started making plans for Field Day. We have a set of social-distancing guidelines we’ll be following this year. We’ll be limiting the number of participants, keeping our tents at least 10 feet apart, and eliminating common eating areas. Also, we won’t be sharing stations and equipment. This year’s Field Day will be different, for sure.

Other than that, I’ve been active on our local ARES-RACES nets, and I have been checking into the Pennsylvania NBEMS Net on Sunday mornings.

You can also find me on 40M or 80M CW in the evening. I usually hang out around the SKCC watering holes.

I’ll be posting more on all of this stuff in the coming weeks. Until then, stay safe, and I’ll see you on the air.

73, Craig WB3GCK