mx5cartalk.com

Has anyone using GPS LAP timers, had issues with the data accuracy? by that I mean, drawing out lines traveled that sit well away from the track resulting in G’s ,speed etc. being off

I’m using the QStarz Q818X with Racechrono and guessing its most likely the Qstarz GPS unit becoming faulty, but apparently GPS satellites can at times suffer from ionospheric interference or issues from solar flares,
but I have no idea how this interference may affect readings if at all

I’ve used this setup for about 6 months now on 3 separate occasions at EC with no errors or distorted data in any sessions.

On Monday at Wakefield, 7 out of the 11 sessions were distorted even when rotated between 3 different cars,

It went from working fine in the 1st session to slightly off in the next 3, then 2 fairly distorted readings followed by 2 that were severely distorted,
then magically it worked perfectly for 3 in a row


Good




Slightly off




Fairly distorted




Severely distorted

Woh! That's weird! My PerformanceBox worked fine for the entire day on Monday. I'll post some circuit maps tonight. Have you checked the number of satellites your GPS was picking up? Looks to me like there might be a problem with the sensor? Might also be worth posting this on the Racechrono forum if you haven't already.

Good to know yours worked fine (same place same time).

The sat number was between 7-11 all day, I can't see a pattern related to it as it worked fine on 7 or 10 and some sessions that had errors where also fixed to 10.

How many sats did yours fix to?

Some possible causes of your GPS errors:

Multipath effects
Interference (natural or artificial)
Small angular separation of satellites
Satellites too low to the horizon

JBT wrote:Some possible causes of your GPS errors:

Multipath effects
Interference (natural or artificial)
Small angular separation of satellites
Satellites too low to the horizon

Drinking and driving...

JBT wrote:Some possible causes of your GPS errors:

Multipath effects
Interference (natural or artificial)
Small angular separation of satellites
Satellites too low to the horizon

I’m wondering if these sorts of interference are the cause, and either the GPS unit or the software is unable to decipher between good sat data and bad sat data(sats having Interference issues), whereas the Performance Box may deal with it better.

It has worked fine previously so most likely only a fault in my GPS unit, but will be interesting if others using a similar setup have come across it.

I just had a closer look at my PerformanceBox data. There was one brief period where the position data was off quite a bit, but nothing as bad as yours was. I was only getting 4 satellites during the warm-up lap of my 3rd session and the position was way out but then came good for the timed laps (see attached map).



Here's a summary of the number of satellites I got throughout the day:

1 - 8 satellites
2 - 8 satellites
3 - started with 4 then latched onto 6-8 satellites (mostly 8 )
4 - 6-8 satellites (mostly 8 )
5 - 5-7 satellites
6 - 8 satellites

It doesn't seem to latch on to any more than 8 satellites. Unfortunately my fastest lap was the first one in the 5th session, which explains why my GPS timing was a few tenths off the official time.

35 lap session at Winton last weekend using a Nokia E51 and Qstarz BT818EXtreme. I only looked once or twice during the day, to notice that there were 8 satellites.

They're all spot on this time but on some tracks on some days the whole lot can be displaced by 10 or 15 meters. I've never experienced the sort of variation you got on a single day. Also, actual times were all within 0.1 sec of official times.
The fat bits indicate the overtaking zones , except for the main straight which is the being overtaken zone .

I suspect your Qstarz is a bit suss.

It doesn't matter if you have 8 or 18 satellites. If the angles are all wrong and there is some interference about, you will get some inaccurate information. It will vary from day to day and even during the day.

GPS units will also report a parameter called 'HDOP' (horizontal dilution of precision) which is an indication of how good or not the fix actually is...

It's in the GPGGA NMEA sentence, what your logger does with that information I have no idea.

JBT wrote:It doesn't matter if you have 8 or 18 satellites. If the angles are all wrong and there is some interference about, you will get some inaccurate information. It will vary from day to day and even during the day.

I think we may have solved it and may be related to what your saying JBT.

As a sat moves on its own orbit the signal travells a longer path through the earths ionosphere, this creates dispersion effects similar to the change in colours seen in the sun at different angles, so the signal suffers a slight distortion,
As it flies further away the path through the Ionosphere gradually increases with a similar increase in the distortion.

So essentially if the unit remains fixed to a sat moving out of range, then the data will slowly begin to distort, as seen in the pattern of increased distortion over time in the track data, until the Sat was dropped and it all went back to normal.

The reason it may have remained fixed to a sat. With a 35hr battery life I left it on all day

Sound about right?

I'm not sure if it's true, but I heard that the civilian GPS doesn't give anywhere near as accurate as a reading as is actually possible (i.e. false data put in to stop terrorists doing ... things??)

edit: On upon a time...

Initially the highest quality signal was reserved for military use, and the signal available for civilian use intentionally degraded ("Selective Availability", SA). Selective Availability was ended in 2000, improving the precision of civilian GPS from about 100m to about 20m.

Rotary wrote:As a sat moves on its own orbit the signal travells a longer path through the earths ionosphere, this creates dispersion effects similar to the change in colours seen in the sun at different angles, so the signal suffers a slight distortion,
As it flies further away the path through the Ionosphere gradually increases with a similar increase in the distortion.

So essentially if the unit remains fixed to a sat moving out of range, then the data will slowly begin to distort, as seen in the pattern of increased distortion over time in the track data, until the Sat was dropped and it all went back to normal.

The reason it may have remained fixed to a sat. With a 35hr battery life I left it on all day

Sound about right?

Ionospheric effects are one source of error, but it's usually averaged out by the receiver. It was more likely a combination of some or all of GDOP (geometric dilution of position), multipath effects and interference. It doesn't take much to throw it out by 10 or so metres - which is still very accurate for a relatively cheap receiver.

JBT wrote:It doesn't take much to throw it out by 10 or so metres - which is still very accurate for a relatively cheap receiver.

Rotary's worst logs are displaced by almost 150m!

150m is quite possible too.