GPS Module PPS accuracy?

JPB33

Senior Member
I see quite a few clock designs use the PPS output from a GPS module (VK15E module in stock) to replace the crystal timebase when a GPS signal is available. I'm looking for a stable accurate 1Hz signal as part of a Picaxe based clock timer but trying to decoding anything else from the module is way beyond me!

Read varying reports and data sheets I've found but dont seem to cover it. Question is anyone know how accurate and stable is the PPS output?

Thanks
 

JPB33

Senior Member
Thanks for the links, very useful, lots of issues raised but it looks like other folks use the PPS for serious work in spite of evidence of jitter.

The clock code is interesting but I think I'll try the 1Hz output first as it should do what I want. I was going to put it through a schmitt trigger before I use it. Put a scope on the output and the pulse is 200ms high and 800ms low, exactly what I want, great start!
 

AllyCat

Senior Member
Hi,

Or you could just use the time from the module
It rather depends if you want to make a very high accuracy "short term" measurement, or know the "Absolute Time" very precisely. The ASCII strings from a GPS engine/microcontroller are not likely to be very good for short-term measurements.

However, the GPS system is fundamentally much more "accurate" than the rotation of the Earth, to the extent that I believe that "GPS Time" is around 10 seconds ahead of GMT or UTC. Presumably any commercial GPS engine corrects for that in calculating its data stream.

But if you have a project to land on Mars it's worth checking whether your calculations are in Metric or Imperial units, or when developing a much higher-powered rocket launcher that an integer calculation doesn't overflow. ;)

Cheers, Alan.
 

Pongo

Senior Member
From the Mildy Interesting Department: VisualGPS sells software that conditions a PC clock to GPS time. Good if you are running a seismometer or similar instrument.
 

PhilHornby

Senior Member
However, the GPS system is fundamentally much more "accurate" than the rotation of the Earth, to the extent that I believe that "GPS Time" is around 10 seconds ahead of GMT or UTC. Presumably any commercial GPS engine corrects for that in calculating its data stream.
When I got my first Garmin, back in 2006 ish - you could see that the clock was different to that obtained from other accurate sources. I seem to recall an error as high as 13 seconds.

However, commercial units don't seem to exhibit this behaviour these days, so presumably they must (somehow) be allowing for it now.
 

eggdweather

Senior Member
I see quite a few clock designs use the PPS output from a GPS module (VK15E module in stock) to replace the crystal timebase when a GPS signal is available. I'm looking for a stable accurate 1Hz signal as part of a Picaxe based clock timer but trying to decoding anything else from the module is way beyond me!

Read varying reports and data sheets I've found but dont seem to cover it. Question is anyone know how accurate and stable is the PPS output?

Thanks
The gps.gov webpage and specifications gives pps accuracy as '1. The normal operations performance standards are consistent with a Block II/IIA clock stability of 1x10-11 at a tau of 3 seconds for either Rubidium clocks or Cesium clocks.' which you can take as very very accurate :) or in normal parlance 0.1ppb, by comparison a normal xtal is about 10ppm not bilion. I didn't look through all of the (huge) specification but I'm sure they cover jitter, my guess is it will tend to zero.
 

AllyCat

Senior Member
Hi,

.. commercial units don't seem to exhibit this behaviour these days, so presumably they must (somehow) be allowing for it now.
I wasn't aware of the issue with my first Garmin (GPS45), back around 1994, but a quick Google produced the following from here.

"GPS time was set to match UTC in 1980, but has since diverged. The lack of corrections means that GPS time remains at a constant offset with International Atomic Time (TAI) (TAI – GPS = 19 seconds). Periodic corrections are performed to the on-board clocks to keep them synchronized with ground clocks.

The GPS navigation message includes the difference between GPS time and UTC. As of July 2015, GPS time is 17 seconds ahead of UTC because of the leap second added to UTC June 30, 2015. Receivers subtract this offset from GPS time to calculate UTC and specific timezone values. New GPS units may not show the correct UTC time until after receiving the UTC offset message. The GPS-UTC offset field can accommodate 255 leap seconds (eight bits).

GPS time is theoretically accurate to about 14 nanoseconds. However, most receivers lose accuracy in the interpretation of the signals and are only accurate to 100 nanoseconds."


Cheers, Alan.
 

eggdweather

Senior Member
My understanding is for navigation purposes GPS time is the important factor not UTC i.e. where is the Earth at any moment in time especially as the earth rotation slows, the leap second is used to ensure GPS time tracks (the slowing of) earth time so that GPS UTC time difference will continue to grow.
 

JPB33

Senior Member
Thanks for all the info. Not knowing much about GPS my initial concern even after some googling, was that the PPS was just an arbitary timed flashing led to confirm lock and not linked to any standard.

My application is a simple clock timer using the minute hand to start the timing process for one rotation of the hand. From that enable signal I've previously used a TTL gate to run 1hz pulses through a counter to check the rate of the clock after a repair and I wanted to simplify the process using Picaxe. I've always (been doing this for 40yrs) found difficulty in finding a stable drift free 1hz source and the GPS module PPS looked too good to be true!
 

AllyCat

Senior Member
Hi,

I've always ... found difficulty in finding a stable drift free 1hz source
I don't know if they're still available, but not long ago I bought a pack of two "Analogue Travel Alarm Clocks" from Poundland (guess the price). The nearest I can now find with a quick Google is from Asda. ;)

They contain a 32.768 kHz crystal and a chip that divides by 65,536 to drive an H bridge output with an exact 0.5 Hz square wave, from a single AA cell. Each second the bridge reverses the current in a small magnet to produce one "tick" of the second-hand (the minutes and hour hands are connected by simple gears of course).

So it's an easy source of an accurate 0.5 Hz square wave and not too difficult to get to 1 Hz pulses by detecting the edges of the square-wave.

Cheers, Alan.
 

JPB33

Senior Member
Thanks Alan for reminding about using a board from a clock, I'd a used one previously and it worked well but it seems to have vanished somehow hence the look at GPS now that the modules seems to be at rock bottom prices. Peter
 

fernando_g

Senior Member
How things have evolved!
Previously one would have had to design a large board to receive and decode WWV or a similar radio signal.

In the late 80s an acquaintance actually did that. The large board was the size of a letter, full of LSTTL circuits and a 680x processor.
 
Top