GPS groundspeed

[Genghis the Engineer]

Interesting problem here somebody might be able to help me with.

Flew two sorties on the same day, using two different GPS sets (one a Garmin panelmount, not quite sure the other but it was a handheld) to generate groundspeed as reference data.

It appears that the two sets gave markedly different (16%) groundspeed values under what seem to be identical circumstances. Now some of this is inevitably repeatability error - but not that much.

The first thing that springs to mind is that possibly there are more than one method - for example differentiation versus some kind of doppler measurement, used in determining speed in a GPS set and that this has led to different GS indications.

I may yet simply go and do a cross-calibration of the two sets against each other, which will give some clue, but has anybody else any experience to lend to the problem?

G

[compressor stall]

I have trialled a Garmin 12 against a KLN900 and got the same, even at speeds around 400kts groundspeed.

Was your handheld one of the old 8 channel ones perchance? That might have made a difference - it might have had to use satellites lower to the horizon as they are the only ones they could see. Lower to the horizon means more error.

Mind you 16% sounds a lot!

CS

[Onan the Clumsy]

16% does sound like a lot and it probably wasn't split evenly either. Sounds to me as if one of those unit is 'broken', because if its calculation was routinely in the region of 16% off then I don't see how it could get certified in the first place.

I guess you'll have to put them side by side to eliminate all the variables.

[bfd777]

I work with a lot of different GPS units doing survey work and we see variation between sets all the time. Some variables are how many channels (8-12), type of antenna (both brand and whether it is active or passive), quality of installation, satellite geometry (of those being received) and then a variety of more technical variables.
Cheers

[411A]

One of the many reasons the FAA does not allow any of the handheld/column mount sets to be used for IFR navigation.
Have two KLN89B (IFR approved) sets installed in my private aircraft, and the GS readout is always within 1-2 knots.

[TopBunk]

Genghis

16% is the difference between mph and kts. 1nm = 6080ft, 1sm = 5240 ft. 6080/5240 = 1.16.

Are you sure that they were both displaying in the same units?

[Genghis the Engineer]

The mph .v. kn error is a possibility I shall check that.

I wasn't flying in the aircraft, it was a report from a full-time TP. But in discussion with him and reviewing the data it seems that the errors are primarily happening at high speed (we use GPS as truth data for ASI calibration, using a geometric method). At high speed the aircraft was in a fairly steep dive (1500 fpm RoD and at around 1.4Vh) - it appears possible that at the attitude (steep) that went with that he may have blanked a number of the handhelds satellites with a fair chunk of the back-end of the aeroplane and thrown it's output.

I shall report back further, thanks for your thoughts.

G

[Onan the Clumsy]

so maybe one was in 2D mode and one in 3D mode?

[EchoTango]

I'm with Topbunk. Well, almost. (There are 5280 feet in a statute mile)

Another possibility is that your speed was not constant. GPS units have different methods of speed averaging to give an output speed. My hand held GPS appears to output a moving average velocity over 3 or 4 position fixes, at one fix per second. An aircraft GPS may well average over a different period.

ET

[Chuck Ellsworth]

One was set to display Knots, the other was displaying miles per hour.

chuck

[RatherBeFlying]

Typically groundspeed is thought of as speed over the ground.

But in a dive your speed can be divided into a horizontal and vertical component. Perhaps one unit calculates groundspeed on the horizontal component while the other calculates it on the actual distance travelled between the two points in space.

You might want to check algorithms with the manufacturers.

[aviate1138]

The Garmin 196 is WAAS capable and is the current dog's gonads of portable GPS devices. IMHO
Lot of bang for the bucks. Can be used in your car or floating gin palace as well. All the data is built in - tide tables - etc.
Waterproof too. Hell, I should be a Garmin salesperson!

B J

[arcniz]

So next time maybe do it the old-fashioned way: with three sensors.

[erikv]

I've carried my handheld (Garmin 92) many times and have never seen substantial differences in indication with other GPS systems.

The only time it ever disagrees with the FMS is during extended over water ops, where my handheld seems to deserve more faith.

Nautical versus statute has my vote.

Erik.