Steve Hicks N5AC has put together a rather informative article on the GPSDO option:
What is a GPSDO and Why Would I Want One?
– by Steve Hicks – N5AC, VP Engineering
All radios need one or more oscillators that are ultimately used to set the frequency where you will be listening or transmitting. If these oscillators are off frequency then you will be listening or transmitting off frequency. How far off frequency? You can tell how far you are likely to be off by looking at the stability of the oscillator and which frequency you are on. For example, if the oscillator is rated as five parts-per-million (5ppm) and you are listening on 1MHz (one million Hertz), you could be off as much as 5 Hz. A typical temperature controlled crystal oscillator (TCXO) for an amateur HF radio is 0.5ppm. To calculate how far off you might be on 20-meters, again we just multiply the accuracy in ppm by the frequency in MHz, or 0.5 x 14.2 = 7.1 Hz. Over time, all oscillators will also drift. These drift specifications are sometimes forgotten, but they are additive to the stability specifications. Aging in a 1-5ppm oscillator are often in the same range of 1ppm or so. So 10 years after purchase, your radio can drift an additional 5-10ppm (70-140Hz). If this drift is the “bad” about TCXOs, the good is that they have very good short-term stability. So even though they can drift a lot over a long period of time, in the short term they are very stable.
Most amateurs know that a Global Positioning System (GPS) receiver can help out with this problem. A GPS receiver is designed to solve a system of equations that have variables of time and position. A GPS receiver becomes “locked” when it finds a solution to these equations. The result is that the receiver knows precisely what time it is and where it is. If the GPS receiver continues to track satellites, it will continue to maintain good time and position information. But it is interesting to note that as the GPS solves the equations, there is short-term drift caused by a number of factors. This jitter presents as short-term instability that, should we rely solely on a GPS for frequency data, would be continually moving around our transmit or receive frequency. So in a sense, a GPS receiver is the opposite of a TCXO — it has bad short term stability and good long-term stability.
The GPS Disciplined Oscillator (GPSDO) is the melding of these two worlds to create something with both good short term and good long term stability. How does this work? Because of the aging problem, most TCXOs have either a voltage steering line that allows the frequency to be adjusted with a voltage or a variable control that can be adjusted with a screwdriver. What if we used a microprocessor to compare the TCXO frequency output to that from a GPS and made adjustments over time to the TCXO to keep it in line for long-term stability? This is exactly what a GPSDO is. The GPSDO module for the FLEX-6000 achieves 0.005ppm stability as long as the GPS has an antenna connected and at least one GPS satellite is in view. Comparing with just a TCXO as we did before, we might be off by 72 milli-Hertz (0.072Hz) on 20-meters. So this is the first key benefit of a GPS receiver in the radio: you will always be on frequency.
The second benefit from a GPSDO is the ability to get precise time information. Having the current time is important for a number of ham applications such as contest logging or digital modes and a GPS can help with this. It can also be used to time-stamp data coming out of the radio. Why would you want to do this? Let’s say that you are a net control station on HF and you are coordinating information from a large geographic area. Even with a good station, propagation can work against you and make it difficult to talk to certain geographic areas. Often there are others that are in a HF net that can hear other participants and these other operators can perform the relay service: they offer to relay the message from the station the net control cannot hear to the net control. But what if this could be done in a more automatic fashion?
What if we could take the RF signal from that operator and send it, timestamped, to the net control and his radio could either select between or combine the RF stream from his radio and the remote radio in order to hear better. Performing these operations is easier if you have good information about the relative time differences in the data and have data that is at the same exact sampling rates. There are three pieces of information required to achieve this: the current time, a precision oscillator and an indication for the start of each second of time (1PPS), all of which are provided with the internal GPSDO from FlexRadio. By using a GPS to synchronize the oscillators at both stations and to time-stamp the data, we are able to make better decisions about how best to combine these two RF streams. While this capability (optimal combining) will not exist with the first release of SmartSDR, it is something that will evolve over time.
Though the GPSDO will not be supported until sometime down the road, should one have the option factory installed from the get-go?
Food for thought,