cancel
Showing results for 
Show  only  | Search instead for 
Did you mean: 

GPS Time differing among cameras

JerseyPix
Apprentice

THis is driving me nuts. I have a 1DXII and 5D4 both using GPS to set the time. Its vital that I have both cameras synched as I use both to shot events, often going from one to the other and back very quickly. 

 

I have my time set via the GPS in both (set to "auto update"). Here's the kicker - the time is off by about 40 seconds between the two bodies. Any suggestions?

12 REPLIES 12


@Waddizzle wrote:

@jrhoffman75 wrote:

An hour apart almost seems like an issue with daylight savings time setting.


It is.  My cameras do not automatically switch to and from daylight savings time.  I have to go in and check/uncheck the box.


It is, of course, unreasonable to expect Canon to issue a firmware update every time the Government decides to change the DST date, What does work, though, is to do all your time setting via EOS utility, because your (Internet connected) computer does know what time it is.

Bob
Philadelphia, Pennsylvania USA

TCampbell
Elite
Elite

There are some rational reasons why this can happen but you need to know a bit about how the GPS system works.

 

There is a difference between UTC time and GPS time.  It's technical... but basically due to the way the GPS navigation system works, it cannot tolerate the injection of "leap seconds" ... and UTC time periodically gets a "leap second" injected.  The GPS time does NOT inject leap seconds (there are other techniques they use to deal with this problem... more on that in a moment.)

 

All GPS satellites have their own notion of current time and they simultaneously transmit the time signal.  To determine your location, a receiver picks up these radio signals... but each signal will be received at fractionally different times.  This information is used to perform a calculation by the GPS chip in your receiver to determine that it's a little closer to one satellite, then to another... and it triangulates your position (there are other factors such as compensating for the effects of General Relativiity ... but we can ignore those factors for purposes of this explanation.   I'll just mention that the GPS system DOES, in fact, compensate for the effects of General Relativity and, if it did not, your receiver would not be able to get an accurate position fix (the error would be so large as to render the system useless) and leave it at that.

 

 

What is of the utmost importance is that all GPS satellites broadcast precisely the same time at the same moment ... and the fractional difference noticed by the receiver is how your position is determined.

 

 

The satellites are not in fixed positions in the sky ... they are constantly criss-crossing the globe.  So in addition to knowing the time, your receiver has to know where the satellite *should* have been located in space at the moment when each time pulse is transmitted.   To deal with this problem, your receiver is loaded with the orbital data for each satellite.  It's programmed into the chip.

 

Unfortnately the satellites do actually shift a bit in thier orbit (for a lot of reasons).  This means that a satellite might not have been where you think at the time it transmitted a time pulse... based on using the orbital data for each satellite.  

 

To deal with THAT problem, the orbital data is constantly being updated.  The master station tracks the real positions of each satellite and adjusts the orbital data so that the system can be used to determine accurate positions.  But this information needs to be updated in each GPS receiver.

 

You may recognize that you don't recall updating firmware, etc. in your GPS receivers and this is because that orbital data is constantly being updated as part of the GPS protocol.  The data is referrred to as the "GPS Almanac" and every satellite is constantly transmitting the almanac (inter-spersed with the time-pulses).   But it's a lot of data and it takes a while to download the entire almanac.

 

It turns out that it's a well-understood amount of time and it never changes... it takes precisely 12.5 minutes to download the whole almanac.

 

 

So back to those "leap seconds"...

 

 

Embedded in the almanac... is the official number of leap seconds that have occurred.  This allows a local GPS receiver to get the "GPS time" ... and when it finally learns how many "leap seconds" have been injected into UTC time... it can combine the GPS time, with the count of leap seconds, to derive the UTC time (even though the UTC time itself isn't what the GPS satellites broadcast.)

 

 

 

 

Back to your problem.

 

That's a LOT of data I just shared about how the GPS network functions.

 

But here's what it boils down to.  If you haven't used a GPS in a while, it will not have the correct positions of all the satllites (which is why getting a position fix can (a) take a while and (b) not be very accurate until the receiver has been running for a while).  

 

But it's ALSO why the time will be wrong.

 

To update your GPS... you need to power it up in a place where it has a clear view of the sky... and just leave it running for a minimum of 12.5 minutes.

 

At the end of that 12.5 minutes, you should notice you have a fairly accurate position fix AND you should notice the time is updated to a more accurate representation of time.

 

You would need to do this with both cameras.

 

 

 

 

One more thing.

 

When I photographed the total solar eclipse across the US last August, I was part of a research project that wanted my image data.  But the data was worthless unless they also had my precise location (not just latitude & longitde... they also needed my accurate altitude) AND they needed my precise time in UTC.

 

I had to use a GPS on my laptop and use that to set the clocks on my camera ... but it changed the times on my camera to UTC time (not local time). 

 

I noticed that after the project, I tried to set the clocks back to local time and it didn't update.

 

I think I had to disable and re-enable auto-update to get the change to take effect.

 

That *might* be a Canon bug and might only be related to the fact that I was just trying to update the time-zone and not the "time".  But I do recall having to switch the feature off and then back on to get it to update.

 

 

 

Hopefully something in all of that will be helpful.

 

 

Tim Campbell
5D III, 5D IV, 60Da


@JerseyPix wrote:

THis is driving me nuts. I have a 1DXII and 5D4 both using GPS to set the time. Its vital that I have both cameras synched as I use both to shot events, often going from one to the other and back very quickly. 

 

I have my time set via the GPS in both (set to "auto update"). Here's the kicker - the time is off by about 40 seconds between the two bodies. Any suggestions?


Sure, just don't use GPS. (It's pretty much superfluous for event photography anyway.) Synchronize your cameras via EOS Utility before going into the event. I always used two cameras when I was doing event photography, and that worked fine for me.

 

EDIT: How did I end up replying to a post that's so far back in the thread (and to which I had already replied!)? Maybe I'm not as awake as I thought I was. Actually, though, my latest reply is still almost relevant. Oh, well ...

Bob
Philadelphia, Pennsylvania USA
Announcements