RC boat skippers have told me that they will take the starting times from their GPS. They assume this is an accurate source of time. After all, the system needs very accurate time to function. So times taken from your GPS should be as accurate as you will ever get. It is logical. Is it true? Is a GPS really a good source for time?
That is a complicated question. The short answer is no. Your GPS is not the most accurate source of time. It may well be as close as you care to know the time. However, if you need to really know what time it is, you will probably need a better source. To understand why, we need to look at the two questions in the title.
Before one can really know what time it is, one must decide on which time to call real. There are two basic ways of keeping accurate time, and they keep different time. Traditionally we have kept time based on the rotation of the earth. This gave us Greenwich Mean Time, which has been refined to what is called UT1. We have since learned to tell time by atomic decay and have developed Atomic Time, TAI. The two types of time differ. This led to Coordinated Universal Time, UTC, which blends the two.
Most people would call UTC real time. Your GPS does not. The GPS system uses GPS Time (well, the satellites in the system use space vehicle, SV, time, but we are keeping it simple). GPS time started midnight January 6/7 1980 and then matched UTC. In the years since UTC has added 13 leap seconds and will add more, to keep UTC with in 0.9 seconds of UT1. GPS time does not add leap seconds, so GPS time is about 13 seconds ahead of UTC now and getting further ahead.
Your GPS does not need to know UTC. It may tell you GPS time. It will then be many seconds off. Or it may convert GPS time to UTC for its time display. This will get you closer, how much closer depends on the sophistication of your unit. It can be very close or seconds off. To know how accurate the time dispensed by your GPS is you need to know something about its software. Or you need to compare it to a time tick.
However even if your software is the best, you will not get truly accurate time off a moving boat. At least not as long as the system is being degraded. Selective Availability plays with the time signals. This is how it introduces error in the position the unit gives you. So the GPS time the unit gives will be off a little from UTC, even if the conversion to is perfect. Degraded information in / degraded information out.
Does anyone really care? 15 seconds off at most, maybe within a second, is close is enough to get you to the race on time, or to use a tide table. Why would a sailor need to know time more precisely? Celestial navigation. Celestial navigation needs accurate time to get an accurate longitude. Even a fraction of a second will make a significant difference when reducing sights.
Further we find ourselves again facing the question of what time is real. To the celestial navigator UT1, the time that best matches the rotation of the earth, is real time. Not UTC. UT1 is the time you need for accurate sight reductions. Even if you could get accurate UTC from you GPS, there is no good way to convert that to UT1. WWV and WWVH do broadcast every minute that minutes correction to convert UTC to UT1. This comes in coded form — right after the time signal. That makes the GPS redundant.
Bottom line: Your GPS makes a good wrist watch. It is not a good source of time for celestial navigation. However should you want TAI, Atomic Time, and can get unmodified GPS time from your unit, just add 19 seconds. You will only be off by the amount that the GPS system is degraded by our military.
Steve Worcester, Sugar Magnolia
Dear Steve:
After reading your provocative article on GPS and the different time standards, I e-mailed Garmin tech support as follows:
I just read an article about the different time standards employed by different authorities. I own a Garmin 48. Is the time displayed UTC, UT1, GMT or GPS time? And how much time error does SA typically cause? There is some curiosity around here, as our sailing club has started using time taken from GPS for starting races. The assumption is that everyone will thus be synchronized, but I do not think anyone has compared different brands or models of GPS receivers to make sure they all use the same standard. Also if the GPS is used to time finishes, will the results possibly be skewed due to SA?
Not to worry. We score the races on relative time: the difference between start and finish times. There is no need to know the time of day to score a race. We could score races with a stop watch; though we, the computer and yours truly, would prefer that you did not, unless the race starts at 1000 hrs.
The key is that all the boats are finished on the same time piece. Two GPS units are two, different, time pieces. Ed.
Garmin replied:
“You are correct in your assumption that the GPS units might be a little off, but it is not because of SA. The software in the unit prioritizes the functions and time display is not high on the list. Your error could be + or - 5 sec. The time standard is GMT.”
This does not completely answer the questions, and in fact raises more questions. I wonder what Magellan, Trimble or Lowrance might say.
Carl Applebaum
Let me correct and expound on what the tech said.
First, there is SA error. However as long as we are only trying to be within 5 seconds of correct, it is not worth thinking about.
Second, the display time standard probably is not GMT. I am pretty sure that they use the same time standard as the rest of the world, UTC. GMT is similar to UT1, which is what they would be using if the were syncronising time to the earth. The GPS can not keep this time. The difference between UT1 and UTC, as well as GPS time, changes daily.
But if we are talking 5 second errors, this is irrelevant. UTC is kept within 1 second of UT1. The larger error is the conversion of GPS time to UTC.
The machine has a much more accurate time, which it uses to navigate. That is a GPS time.
Some GPS units will let you read just that: the GPS time. This will be off UTC by 13 seconds now and more later.
Others will apply a one time correction based on when the GPS was designed / built. As the difference between GPS and UTC changes with time, these will have different errors.
As the changes are predictable, the GPS maker can probably program around them for accurate UTC; but that is complicated. It may require more hardware and cost money to get it perfect.
To repeat what the tech said, a time the user can read is a low priority.
You did not buy the GPS because you wanted a watch, did you?
Let us put things in priority. I mean, how many times have you laid awake at night worrying if your watch is set to UTC or UT1? The only time I would worry about the difference between UT1 and UTC, or about the time errors because of SA, is when reducing sights. There are not many more times that I have cared if I am even thirty seconds off all universal standards.
People have navigated, around the world, with less accurate time. The clock on the Spray is reputed to have had only one hand. The explorer for which one brand of GPS is named had no idea of his longitude. He just sailed a latitude till he found some place new. And why not, he did not know where he was going anyway? One can navigate, if less accurately, without knowing the time.
But you do not have to try to do so. You can get accurate time, UTC, off the radio ( WWV or WWVH). They will also broadcast the corrections to give you UT1. Why settle for less?
Further, those early navigators knew the limits of their tools. This is important knowledge still.
Ed.