Synchronising TWO DL1 run files possible?

[Kris]

Hi

I'm using two independent DL1 data loggers to record vehicle accelerations on a railway vehicle to ascertain relative body bogie accelerations. I have both loggers fitted to each end of the vehicle, and have external accelerometers connected to bogies to measure three dimensional accelerations. They are set to record at a GPS speed of >3km/h and are 50MB files that aren't in synch from the start.

I'm also not recording RAW GPS data to save card space as the units are recording for several days at 100Hz sample rate (could be pursuaded if this data is useful though).

I'm having difficulty superimposing the accelerometer accelerations on the analysis software (7.3.38). The speed/ acceleration data appears to be out of synch by 13 seconds between each data logger run file.

When I set the graph X axis to use GPS time the accelerations then appear synchronised but the acceleration data becomes very distorted (looking like spikes and squares instread of sinusiods), as does GPS speed. This distortion changes but doesn't reduce when GPS time is then smoothed over 1000 seconds.

Is there any way to synchronise the time between different files to a GPS time stamp and have synchronised data that isn't distorted by the process?

If this isn't possible, is it possible to manually shift both time bases on the x axis so the accelerations are correctly overlayed (i.e use longitudinal shock accelerations to synchronise the time base)?

Thanks!

Kris

[tristancliffe]

Could you trim the run file that starts first (Run Manager, Reduce Run Length [I think]) by 13 seconds? Might take a bit of trial and error, but should leave you with good data, shifted to where you want it.

[Kris]

Could you trim the run file that starts first (Run Manager, Reduce Run Length [I think]) by 13 seconds? Might take a bit of trial and error, but should leave you with good data, shifted to where you want it.

An excellent idea Tristan, I'll give it a go. It's just a shame the data would synchronise with GPS time as the trimming accuracy is as good as my ability to measure the time shift between the two signals, and I'll be looking at the phase of the external accellerometer signals for each end of the vehicle to identify how it's riding on the track (currently not that well as bits tend to fall off (air reservoirs, twistlocks etc).

Next job will be to try out a VIDEO4 and a couple of bullet cameras...

[faraday]

There are simpler ways to do this.
Email me: rotoflow at optusnet dot com dot au and we can discuss