Some Details of Station Remote Control


Introduction

 

My first experiences in remote control were related to the cottage in Simpsiö mountain of Lapua. Maukka OH2BYS built there a system based on a Kenwood TS-2000 radio and W4MQ software. There was no computer controlled amplifier nor any good low band antennas. That’s why I got no new ones on any band or mode using that system. For me, it was basically just a tool to compare propagation between my home QTH and Lapua. There were differences. Usually, because of the stack of beams there and because of the clear view to everywhere, the high bands were better in Lapua. However, sometimes the signals were better in Espoo. 400 km surely is a long way what comes to radio wave propagation.

The usefulness of remote control changed when I teamed up with Markku OH2RA whose station has pretty good antennas on all bands. At first there was no remote controlled amplifier there, so for a few years it was receive only. His QTH is much quieter than mine, so just to be able to receive better provided me new ones especially on topband. Then we begun transmitting experiments. The very first test QSOs were made so that Markku tuned his conventional manual amplifier for a band in which I needed a QSO. It worked, though you can’t possibly call it practical. Most of our real problems were software related. Finally we understood that the radio control software we had was basically useless. It worked just fine locally, but not remotely, even though it was adverticed for remote use. The issues were solved by running the radio control software locally, and routing audio via a different program. Naturally, a remote desktop program is used to control it all. Network latencies were also a bit of a nuisance. That all became quite good when our free VPN provider LogMeIn Hamachi decided to begin charging for their services. Instead of paying we quit using it and a miracle happened – network latencies were not a problem any more. At the end of 2013 I bought myself a Christmas present – a used Icom IC-2KL amplifier. It was then promptly installed to Markku’s station. Well, remote receive capability is surely a nice thing, but to be able to transmit remotely is super. At least when it happens to be a station with big high band stacks and directional low band antennas.

Despite of my own station being quite minuscule compared to Markku’s station, I also wanted to make a remote control system for myself, as it might be that Markku’s station is not always available for me. Besides, it was fun building all the bits and pieces. The following chapters are about my own station.

To make it remotely controllable turned out to be a rather complex task due to historical reasons related to station automation. An overall block diagram of the remote control aspects of the station is shown in figure 1. Antenna selection is done automatically but it is not shown in the picture as it is completely independent from remote control systems.

Mona Lisa

Figure 1: Generic View

There are three different operating modes in Espoo. DXDoubler is used to select audio. Local radio is usually on both channels. In addition, one can listen to OH2RA’s station on one channel or on both channels. There is a computer to control my station and another computer devoted just for controlling OH2RA’s station. At first there was only one computer, ie. the local control pc that controlled the whole system. A Windows 8 update made that a bit unstable, though, and that’s why another computer was added. I toggle between the computers using a small KVM switch. By the way, to make computers quiet takes some effort. One computer was reasonably quiet right from the factory. Replacing the standard graphics card with a fanless one lowered noise level even further. Some rubber-like insulating material that I added seems also to have some effect. The other computer became bearable by replacing the original disk drive with a SSD and changing a few fans to a quiet one. If I wanted to make these completely quiet then the nexts steps would likely be fanless power supplies and fanless CPU coolers.

Other standard components used in the station are K1EL WinKeyer for sending CW, and MFJ-1275 Sound Card Radio Interface for isolating the local control pc from other station equipments. PTT functionality of WinKeyer is sometimes used with preset delay to make sure that the leading edge of CW keying is not lost, but usually I simply let the VOX of the radio to do the job. It is a bit easier so.

 

PTT Functionality in SSB

I originally wanted to avoid using VOX for SSB and use PTT instead, though VOX works ok too. There is a Vellman K-8055 based relay controller that is used to switch station powers on and off. It is used to control my antenna rotator (a Yaesu G-1000DXC), too. One of the relays of the relay controller acts as PTT when the remote operator wants to transmit his own voice in live. Most practical DX chasing situations, however, call for a voice keyer. I am using a free VoiceKeyer software package called SM3WMW 1.7 for that. It controls the PTT line via RS-232 connector’s RTS signal. 

The following opto isolator based box is used to change the RTS signals into PTT relay signals.

 Mona Lisa

Figure 2: PTT circuit for VoiceKeyer

.


Safety Timer

Remote control relies totally on data connection working ok. If the connection fails, the station is out of control. This is clearly not a good thing. If some component fails, like if a keyer locks into transmit state or some software package halts at a wrong time, some equipment damage is bound to happen. One approach to solve this is to have another command channel that is independent from the ordinary data channel. I have for example seen quite many people to use a GSM relay to switch the system off in failure situations.

In principle, this solution should be good enough. However, for it to work properly, one assumes that one’s primary network connection and GSM network do not fail simultaneously. Maybe so, maybe not. Mobile operators have indeed invested quite significant amounts of money in reliability so one could assume that there is certain degree of independence. Unfortunately, my earlier experiences have proved that something else can go wrong. It has for example happened that because of not using the GSM connection for a certain period of inactivity time, the operator has simply shut down the whole GSM subscription. Another hiccup was that the mobile phone that was used to control a safety relay box had failed. And then there is also one more smallish problem. Even if everything works as planned, to send a message through GSM to switch off the station in an orderly manner may just take too long a time. I wanted the safety shutdown to happen independently of anything that is outside of my station. The resulting devices are shown in figure 3. Their circuit diagrams are in figures 4 and 5. 

Figure 3 Safety devices in Hammond enclosures

 

Mona Lisa

Figure 4: Safety Timer

Figure 4 shows a simple timer that is based on an Arduino Uno microcontroller. It samples the PTT output line of the radio (TX Ground) in an infinite loop. When the PTT line is grounded, a 60 second timer starts to run. If it is run to the end without a PTT line release, a warning light is lit. If the PTT line is not released in 20 more seconds, then the PTT line is opened so that the amplifier goes to receive mode, and a fault light is lit. If the PTT line is not released in even more 20 seconds, all radio and amplifier related powers in the station are switched off. For local operations, there is a mechanical switch that can bypass the above mentioned functionality. Usually I don’t use it, as I really can’t make any reason for longer than 60 second transmissions without a PTT release. This was the first of my independent safety devices.

Figure 5 shows a network watchdog which is the other independent safety device. Hardware is very similar to the safety timer. The important difference is that the network watchdog is connected to the station control pc via usb port. This is how the watchdog hardware communicates with the station control pc which then communicates with the remote operator pc.

 

Mona Lisa

Figure 5: Network Watchdog

Arduino’s software runs in an infinite loop waiting for an alive message from the station control pc. When an alive message is received, the Arduino board enables station powers by switching on a relay and starting a timer. A small LED blinks quickly to show that an alive message has been received, a text describing what is going on is shown in the text box that is between the NET and Save Settings buttons, and a large LED is kept lit as long as the relay is connected. If there is no new alive message within the timer’s interval, the relay is simply disconnected, which switches off all station powers regardless of any other device trying to do whatever. And where does the station control pc get its idea to send the necessary alive messages? Well, there is a simple MS Visual Basic program that sends those messages whenever it receives an alive message from the operator pc. The user interface of this simple program is shown in figure 6. The same program is used at both ends, i.e. at the station control pc and at the operator pc. Only some software settings are different.

Figure 6: Network Watchdog’s PC software

I could have built just one box to contain all functionality of the safety timer and the network watchdog. I could have saved some euros doing so. The reason for building two hardware boxes instead of one was to simplify the software. I believe that having two completely separate simple systems in series is significantly more reliable than having one complex system. This is especially so because when doing the finishing touches of this remote control system it has been some fifteen years since I was a professional software engineer. I do not regret a minute for getting into a bit different career, but it is a fact that I no longer have the programming skills I used to have. Note that both boxes are built in a fail-safe style. If for example the power is lost in either of the boxes, it immediately fails in such a way that no control power is given to the radios. 

 

Software

The following software packages are used:

Hamradio Deluxe for radio control

SM3WMW VoiceKeyer for voice keying

Wktest for controlling the WinKeyer CW keyer

RemAud for audio transmission (IP Sound does not work ok if the operator pc does not have a public IP address)

Self-made C++ program for Velleman K8055 card (powers, rotator, ptt)

Self-made Visual Basic program for amplifier control

Self-made C program for Arduino Uno based safety timer

Self-made C program for Arduino Uno based network watchdog

Self-made Visual Basic program for network watchdog

A remote desktop program for controlling the local control PC

A VPN program for secure communications.

 

Conclusion

I surely could have made my station’s remote control system less complex by simply buying the necessary stuff from commercial suppliers. It is essentially my linear amplifier that kind of decided the direction of this project. Had I bought some of the widely-supported amplifiers, like an Expert 2K-FA, I could have used ready-made devices that operate well together without any fancy add-ons that I had to build. I am not really sure if I even saved any noticeable amounts of money by building it all myself. I am absolutely one hundred percent sure, that if I count any monetary value to my time that I spent building and debugging my station, I would have been better off using commercial solutions. But it was heck of fun doing it myself. I remember the good old days, when I was young and I built things by my own hands, when there were new things coming to radio, and when there was new DX to work on the bands. Hey, these days nowadays likely are still good times. They are the good new times.

 

To main page