GPS accuracy?

[Bill Huson]

Is GPS speed dead on or questionable? This weekend I hooked up a new fangled GPS thinger for a trip to Charleston, SC. Got it for Milady a few months ago and she sez it's the cat's meow. It did not warn us of the BIG HONKING BRIDGE near Charleston, and Milady hates bridges an darn near caught the vapors and fell out. Took her a good half hour to get over the shakes. Anyway, I noticed a cute thingy in the lower right corner of the GPS screen - a speed limit sign icon with the posted speed limit, and my speed - black number if legal, red if over. Of course I'm busting into the red zone, but I also noticed my truck speedo was reading 2 MPH+ slower than the GPS readout. Saw truck needle is at 70, GPS sez 73.

I'm thinking the GPS is more accurate - true???

[David_L6]

I've compared the speeds shown on my Garmin nuvi 2555 to the speeds shown on my old Garmin eTrex and they match up. My truck speedometer reads the same as GPS. Do you have larger tires than what came on your vehicle?

[Guest]

Gps

GPS accuracy varies with the number of satellites that the unit is receiving to undertake its calculation. Normally a Garmin Etrex will be accurate to 16 feet when it has 8 satellites so position accuracy is .003+/- miles. Assuming that the software to calculate the MPH is robust and the refresh rate of the unit is adequate you can expect the unit to be accurate to better then .1+/- MPH. I also have a GPS that is certified for use in an aircraft and it reeds identical to the Garmin. My truck speedo is accurate after 50% of the tires have been used up so I have a low error indicated speed with new tire and a high error with well used tires.

Alan

[Racers Edge]

A few years ago I put 3 Garmins (three different models) in my boat for a day of testing. Two of them were the same and one was consistently .2 mph slower. When I contacted Garmin, they had no explanation why they may not all be identical because they were all based off the same software. So all GPS units are not identical even if they are designed to be.

[bill van steenwyk]

Never had the opportunity to use a GPS when testing or racing, as I stopped driving prior to their widespread use, so I know absolutely nothing about accuracy of the units against other means of checking speed.

We always found that speed was nice to know and brag about if you were fast, but nothing beat the stopwatch and tach when testing as the fastest boat was not always the one that got to the finish line first. Knowing RPM is invaluable when testing props.

That being said, there is much to be pleased about when testing or racing when not having to drag a pitot tube around for speed information. LOTS of difference sometimes in handling and acceleration without one compared to one in the water, more than you would think. If you want to really find out what dragging something around does for you, try raising your sponson fin when coming past the last turn buoy.

Although I think it is illegal now, I had a "power" raise and lower on a sponson fin at one time and you cannot believe the difference in acceleration when passing the last buoy and you are able to lift it out of the water while straightening the boat out at the same time.

[modracer7b]

I also ran the retractable sponson fins in B and C stock hydro in the 60's. They were only legal for a few years as they were out by the time I ran a D stock hydro early 70's when I was running part time. Never used a gps either as I went by the sound of the engine revs, elapsed time around the course, and speed reached coming out of the turn and the first quarter of the straightaway to get my competition setups. I always left some top end off
the final setup. I did blow myself over once when I set the retractable fin down too fast.

Bill Rosado

[ZUL8TR]

Speed, testing and racing

Back in the 70's I would test different setups, props, etc with a pitot speedometer in a South fla canal that was very stable with hardly any wind and no waves. Really didn't care about the speed only the Relative Difference in speed and rpms in a straight line. At the race course the setup was always different and I would judge the setup by being able to keep up with the top runners and eventually get ahead of them. Acceleration, keeping the revs up, fin depth and other race conditions more important than top speed especially since most of the courses were only a mile around. The only course where top speed was most important was Lake Hollingsworth in Lakeland Fla. as it was a long course back then. There you never let off the throttle unless a black flag was thrown up.

I ran setups where the fin was clear of the water in the straights and set it down by body position in the turns then moved back before turn exit to clear the fin drag. Worked very well once technique mastered.

[Matt Dagostino]

Testing

A few years ago I put 3 Garmins (three different models) in my boat for a day of testing. Two of them were the same and one was consistently .2 mph slower. When I contacted Garmin, they had no explanation why they may not all be identical because they were all based off the same software. So all GPS units are not identical even if they are designed to be.

We always use multiple GPS units while testing and rarely do they both show the identical speed. One or the other is sometimes off by 1-3 tenths of a mph.

We always use NEW batteries in the units every time we use them as they seem to draw a lot of power to operate. Don't be cheap and buy off brand batteries!! Use top shelf Energizer or Duracell!!

And just because a certain propeller may be a couple tenths slower or faster doesn't mean anything until it is raced!! Often times a propeller will not perform as well on the race course as it does in perfect test conditions.........

[krazy karl]

I used a Garmin handheld to check max speed when testing. This year I used a Keller speedometer and a tachometer. I was able to watch the speed no matter what I was doing. It was a real eye opener to see what was happening in the corners and coming out of the corners. I have 3 props for my ASR, new last year, and was unable to do any testing last year, so most of my testing was under race conditions. I found that my main prop from last year didn't reach max speed until I was at the end of a long straight. Another which I suspected was a dog, was, slow in the corners, slow on acceleration and down on top end and low on RPM. I had suspected it to be low on pitch but now appears to be too big. The last prop which I never even tried last year because I suspected it to be unsuitable turned out to be my best. Held a higher speed in the turns, came out quicker, ran the fastest and carried the highest rpm.

My point is GPS's are nice and accurate but unless you can watch the speed all the way around the course probably not the best way to go. Also if you can watch it full time, is it giving you the best info around turns? It tends to measure distance form point to point, probably on the scale of accuracy, ie accurate to say 16' or whatever it comes up with depending on the satellites. So on a turn it may not be recording the true arc thus true speed. For example two summers ago I had a part time job transporting railroad crews. We were paid by the mile which was measured by a GPS unit in the van. On a 400 mile trip we were cheated about 20 miles because of the point to point measuring used by the GPS. It was especially bad when the unit lost signal momentarily due to environmental conditions. True on the speedometer, the pitot tube will take some speed away but it should be equal for each run and we are only comparing setups anyway. My
kk

[David_L6]

I agree with you KK. GPS can't compete with a speedometer for testing unless all you want to know is max speed.

[john527]

garmin engineer

I talked with an engineer from Garmin awhile back, because I was getting amazing speeds from my snowmobile, he reassured me that the testing and feedback that they have done with there products, the gps was ready alittle happy because of the vibration that the machine was producing.
Snowmobiles, dirtbikes and boats always seem alittle high because of the horizontal and vertical motions at the same time.
At least he seemed to know what he was talking about!

[ZUL8TR]

error

I used a Garmin handheld to check max speed when testing. This year I used a Keller speedometer and a tachometer. I was able to watch the speed no matter what I was doing. It was a real eye opener to see what was happening in the corners and coming out of the corners. I have 3 props for my ASR, new last year, and was unable to do any testing last year, so most of my testing was under race conditions. I found that my main prop from last year didn't reach max speed until I was at the end of a long straight. Another which I suspected was a dog, was, slow in the corners, slow on acceleration and down on top end and low on RPM. I had suspected it to be low on pitch but now appears to be too big. The last prop which I never even tried last year because I suspected it to be unsuitable turned out to be my best. Held a higher speed in the turns, came out quicker, ran the fastest and carried the highest rpm.

My point is GPS's are nice and accurate but unless you can watch the speed all the way around the course probably not the best way to go. Also if you can watch it full time, is it giving you the best info around turns? It tends to measure distance form point to point, probably on the scale of accuracy, ie accurate to say 16' or whatever it comes up with depending on the satellites. So on a turn it may not be recording the true arc thus true speed. For example two summers ago I had a part time job transporting railroad crews. We were paid by the mile which was measured by a GPS unit in the van. On a 400 mile trip we were cheated about 20 miles because of the point to point measuring used by the GPS. It was especially bad when the unit lost signal momentarily due to environmental conditions. True on the speedometer, the pitot tube will take some speed away but it should be equal for each run and we are only comparing setups anyway. My
kk

GPS units work best in straight line and if the receiver is a fast processor, I preferred pitot speedo back when I was racing in the 70's (had no choice gps wasn't around and even if it was I still prefer the real time view of the speedo as you note). Plus differences in changes are what are more important that absolutes.

In the world the piecemeal calculations and accumulations (as in you note in your example of a 20 mile difference over 400 miles) they are subject to truncation error that accumulates with the number of pieces being accumulated. This becomes a real issue in the modeling of the real world where differential equations are approximated thus truncation error of the differential function and the time step size become very important and if not properly adjusted for minimizing the error the model could very well be useless for analysis and prediction (example global warming predictions and what portion is man made? - will never be known with reasonable acceptable accuracy).

[john527]

if you guys find something that works and holds up let me know, our DMH keeps breaking all the speedometers we use, we have to go to a mach meter!!!

[blueskyracer]

if you guys find something that works and holds up let me know, our DMH keeps breaking all the speedometers we use, we have to go to a mach meter!!!

Thats a good one John.. Its kind of like our 79 Corvette.. The spedo only goes up to 85 so I can never get a ticket for over 85MPH right??

[Tim Weber]

I've had instances where a GPS will give me wacky numbers.
I made a run and I was watching it in the boat for a second or two and it was showing 65 ish and when I came in it was 97 mph.

I look at a gps as a tool for comparison. I seem to win a lot of races even though I'm the slowest guy on the beach.

I learn more from the tach in reality.

Tim