logging Setup for Garmin GPS 18x5hz
Moderators:R!C0, JonMan, RickS
OK, will try that maths on some raw data, thanks RickS. No rush on this, whenever you have some free time. Err..... right... 
Re: logging Setup for Garmin GPS 18x5hz
Hi there again!
Right! I've managed to massage the data you sent, with a couple of simple maths functions and produce a 'spike-free' trace, using the entire log, of Rockingham!
As mentioned before, I produced new Log and Lat traces from your raw data using:-
combine16('GPS LAT HW','GPS LAT LW') / 10000000
combine16('GPS LON HW','GPS LON LW') / 10000000
This produced our 'spiky' traces that I posted previously in channels GPS Lat and GPS Lon.
From measuring the difference between the real data in the plot and the extremes from the spikes, I set a 'Declick' and GPS Valid channel function to those figures:-
declick('GPS Lat', 0.005, 5)
and
'LAT ADJ' > 0
Hey-presto the plots 'line free' for the whole log! Have a look at the screen print and tell me what you think. It shows pits to track, back to pits and out again to finish, as does the log. Of course, this is really 'shutting the stable door after the horse has bolted' as all I've done is tidy up the data once logged. This would of course mean that a new math calculation would have to be done each time the venue changes and the Lat/Lon figures alter. Not ideal. The answer is not to have spiked data to start with (which I notice happens at the same point in most laps). I'm wondering if signals were erratic due to external features of the circuit. I had one or two signal losses in the Golfs MkII's we race when we came to Rockingham and discovered that if I didn't pass under the Start/Finish gantry, the problem went away..! Go figure!
I still reckon that the i2 is smoothing the data when logged and I will follow this up to see what improvements can be made from this end.
Right! I've managed to massage the data you sent, with a couple of simple maths functions and produce a 'spike-free' trace, using the entire log, of Rockingham!
As mentioned before, I produced new Log and Lat traces from your raw data using:-
combine16('GPS LAT HW','GPS LAT LW') / 10000000
combine16('GPS LON HW','GPS LON LW') / 10000000
This produced our 'spiky' traces that I posted previously in channels GPS Lat and GPS Lon.
From measuring the difference between the real data in the plot and the extremes from the spikes, I set a 'Declick' and GPS Valid channel function to those figures:-
declick('GPS Lat', 0.005, 5)
and
'LAT ADJ' > 0
Hey-presto the plots 'line free' for the whole log! Have a look at the screen print and tell me what you think. It shows pits to track, back to pits and out again to finish, as does the log. Of course, this is really 'shutting the stable door after the horse has bolted' as all I've done is tidy up the data once logged. This would of course mean that a new math calculation would have to be done each time the venue changes and the Lat/Lon figures alter. Not ideal. The answer is not to have spiked data to start with (which I notice happens at the same point in most laps). I'm wondering if signals were erratic due to external features of the circuit. I had one or two signal losses in the Golfs MkII's we race when we came to Rockingham and discovered that if I didn't pass under the Start/Finish gantry, the problem went away..! Go figure!
I still reckon that the i2 is smoothing the data when logged and I will follow this up to see what improvements can be made from this end.
- Attachments
-
- CMW003.gif
- (346.97KiB)Not downloaded yet
Re: logging Setup for Garmin GPS 18x5hz
Just received an e-mail notification that there was a new reply to this (old) topic, but I see nothing new here. Is the server playing up? Thanks, and happy New Year.
Re: logging Setup for Garmin GPS 18x5hz
Yes apologies, we're getting a lot of new year spam!
Re: logging Setup for Garmin GPS 18x5hz
I guess it must be. There have been a few topics added and replied to today (usual busy '1st day back at school..!'). I'll get it looked into.
Thanks for the 'heads-up'.
RickS
PS. Happy New Year.....
Thanks for the 'heads-up'.
RickS
PS. Happy New Year.....