knot meter vs. gps

[Scott]

Bill, why worry about it?

I was going to calibrate my knotmeter until I figured that on a lake where I'm not going anywhere it's just a relative measurement anyway. I've noticed a change as the season wears on. In the spring, when first installed, we were reading higher speeds by about half a knot than we get now. I blame it on accumulated scuz from the water. If I take the transducer out and clean it, I bet the readings increase slightly. But what difference does that make? I can pretend we're going 8 knots when I know we're going 6. Maybe there is a slight amount of current, induced by wind or drainage, that sets up at different locations on the lake. That would seem to be the cause for some variability assuming both systems are accurate.

 

[Fred Ficarra]

Bob, I think the receivers that surveyors use sit

in one spot and average their location. That is extremely accurate. If you can interface to a chart you can see your receivers accuracy on the computer screen. I run a laptop with Navtech carting. The GPS shows accuracy to within 3 feet. The location on the chart agrees. I'm right where it shows in my slip. And you can also see interference. The location is all over the place. The CG is working on that. The first of a new generation of GPS satellites was launched last week to help correct that problem.Also, if each satellite starts its' update every six seconds, lets remember most receivers track 12 satellites. That's really accurate no matter what.

 

[Fred Ficarra]

Ah, Bob,

a couple of things. First, that link is old. It's 10 year old stuff. Selective availability is history.  The other point is that the CG acts only as an agent of the mariner when it comes to GPS. The Department of Defense owns GPS. As such, we all have to go calling with hat-in-hand. Except me. I emailed President Clinton and one of his staff colonels wrote back and said the President agrees with me and has ordered SA turned off. True!

 

[Jon W.]

Measured course

I sail a lot on a lake with minimal current. I have calibrated my knot meter using the GPS. I have also noticed the discrepancy between the GPS and knot meter. It varies about +/- .3 knots, and doesn’t stay the same for long. I just chalk it up to variables in both the GPS and knot meter combined.I did want to remind people that calibrating a knot meter over a measured course can also be affected by any currents, so an out and back run should be done.

 

[Scott]

Fred!

I would sleep much better if I knew that you had the same influence with the current president! I tend to agree with David's explanation. I have talked with surveyors about GPS and they indicate that the handhelds are good for accuracy to only about 25 to 50 feet. Not nearly good enough for pinpointing such things as property corners and such. So this would seem to indicate that the accuracy variable in location would affect the instantaneous velocity reading.Better technology is available at a price, however. Say $10,000 for the best handheld units? It's a pretty good investment for a small survey company or the wealthiest boat owners.

 

[Bob]

And Fred!

It's good to hear Clinton made those colonels earn their pay. I always thought that having them serve canapes was a touch on the slack side.

 

[Don12364]

Hey Guys and Bill

Why debate the accuracy of GPS. Boat logs of the thru-hull variety are notoriously innacurate. You may get them right at one speed with a non-heeling boat but you will be lucky for this calibraion to hold throughout the speed range.Influences are:-1) Condition of hull affecting boundary layer thickness and its turbulence, 2) Water flow not being dead in line with the impeller. This varies on each tack. 3) The influence of local obstructions such as keel etc.4) Depth of impeller below surface.

 

[BobW]

Cause it's fun, Donalex ;D

Fred, I knew that spec was old, but the parts of it I quoted haven't changed - I know because my GPS is a Garmin 75, circa '95 and it still works, which it certainly wouldn't if the navigation message format or frequency had changed.Also, SA doesn't affect the navigation message - it messed with the actual time that was sent.And even +/- 3 meters accuracy only gets us +/- 20% accuracy when traveling at 5 knots. Bottom line: neither the log nor the GPS is going to be very accurate in the short term. The only way to accurately measure velocity is over a longer period and distance.Cheers,Bobs/v X SAIL R 8

 

[Fred Ficarra]

Bob, that's for sure.

My old Magellan has a estimated accuracy readout on the main screen. It's a handheld. Before SA was turned off it read between 30 and 300 feet for estimated error allowance. Now it reads 3 feet almost all the time, even when interference is bouncing our location all over the marina. It doesn't know there's interference.

 

[Reudi Ross]

my 2 cents on GPS accuracy

I have noticed that on land my handheld accuracy is relative to a benchmark and not random. By that I mean if I set a waypoint at a benchmark and walk precicely 300 feet according to the gps and set another benchmark, when I measure the actual distance it is within 1-2 feet. When I come back a few days later it may be off as much as 40 feet from the original benchmark, but the second benchmark is off an equal amount in precicely the same direction and the actual distance between is still within a foot or two. I believe that when surveyors use GPS they tie into a known control point and correct for this discrepency constantly.

 

[Bill C320 Papillon]

Wow

Thanks for all the input. I had no idea that my question would generate so much discussion.I think I'll take Scotts' advice and not worry too much about the discrepency however now I feel challenged to somehow plot a nuatical mile on a map and transfer it to known points on the shore to check and see if either instrument is in the ball park. I'll do this on a day when I'm really bored because there is no wind.Thanks again

 

[Calif. Ted]

Umm, If you sail, walk or bicycle away from

a set waypoint till the GPS says it's 1 mile away I think you've measured your distance. As Stu noted, you gotta do it going both directions and average. Then it's just a speed, time, distance problem or am I being simple-minded here ? Course it would be much more accurate to do 100 mile legs but we're talking bout an "instrument" that relys on a bronze (?) paddlewheel with crusty lttle white worms always trying to bring it to a stop.

 

[Jim Seamans]

Accuracy ?

We use a garmin 176 on Java and last summer while off the coast of Delware we had any interesting experience. I had set in two waypoints that were buoys off the coast. At 2 in the morning I had to steer around the first marked buoy because it was dead on and would have it hit. Later that morning after sleeping my wife had the same experience with the second mark and was complaining to me that I should have told her about it. I was not expecting the gps to be that accurate. Lesson learned.JimS/V Java

 

[Fred Ficarra]

Jim, that reminds me.

When we were hanging around south Florida I did the same thing. There's a sea buoy on the way to Ft. Jefferson. I used it for a way-point. Lucky it was daylight. The autopilot was aimed right at it. We would have hit it on the nose, I mean bow. And that was with SA I think. That was '94. Was it turned on then? Maybe not. *o

 

[rsn48]

GPS should be more accurate

The GPS as has been noted is your speed over ground, whereas your knot meter is speed through the water, taking into consideration current. I find the GPS to be more accurate in terms of "dead reckoning" etc.

 

[Fred Ficarra]

Dead reckoning is just that,

everything is dead as far as on board instrumentation. It's a very educated position guess. Kinda like celestial navigation.