posted 09-05-2006 08:58 AM ET (US)        

Yesterday I was conducting a little drag race with Steve and BACKLASH. We both ran our boats up to wide-open-throttle in relatively flat water, and found there was only a slight difference in speed. After the test we compared the speed measurements we obtained from our GPS receivers. I observed on my GPS that we were going about 38.5 to 39-MPH. Steve observed on his GPS that we were going about 44-MPH. I was quite surprised at the 5-MPH difference in speed deduced by the GPS receivers. (Curiously, the YAMAHA pitot tube speedometer on BACKLASH was in very good agreement with my GPS reading.)

My GPS is a current model Standard Horizon CP-150. It is a 12-channel receiver with WAAS enhancement. During the test the Horizontal Dilution of Position error (HDOP) was about 1.1-feet. [With the benefit of seven years of addition study, I now realize that the HDOP is a figure of merit, not a dimension in feet--jimh]

Steve's GPS is an older LOWRANCE unit with differential GPS (DGPS) enhancement. I believe it is a 5-channel unit. I am not certain if the DGPS receiver had a correction signal or not.

A speed of 39-MPH is

39 MILE/ HOUR X 5280 FEET/MILE X HOUR/3600 SECOND = 57.2 FT/SEC

A speed of 44-MPH is

44 MILE/ HOUR X 5280 FEET/MILE X HOUR/3600 SECOND = 64.5 FT/SEC

Thus the differential speed between the two boats would be about 7.3 feet per second. I do not believe that there was that much difference in speed. We were more or less keeping pace with each other, although BACKLASH may have had a slight speed advantage. I would judge the difference to be less than one foot per second. That is a difference in speed of

1-FOOT/SEC X 3600 SECONDS/HOUR X MILE/5280-FEET = 0.7 MPH

From this brief test I was made aware of a potential for error in speed measurements by GPS receivers. It was quite surprising to see such a large variation between two devices. This brings to mind Mark Twain's observation:

"A man who owns two watches never knows what time it is."

[This citation may be in error. Many sources now cite a different author for this familiar quotation.--jimh]

For many years I have been observing that my boat could only reach 40 to 41-MPH, and, to tell the truth, feeling a little bad about that. Now I find that perhaps all this time what was holding me back was my GPS!

posted 09-05-2006 09:05 AM ET (US)            

This is really going to throw out any speed data we have all taken for gospel.

posted 09-05-2006 09:38 AM ET (US)         

--
Moe

posted 09-05-2006 12:20 PM ET (US)            

I have often wondered about the accuracy of GPS speed measurement. Not only in the marine environment but also on navigation instruments for automobiles. My buddy has a built in navigation system in his car (Becker)and I have a mobile system in my car. I once took mine along with me when driving in my buddies car and compared the displayed speed on both units. Much to my surprise there was a difference of 8% at 120km/h (about 75MPH). I have no idea who's was correct. The speed on the speedometer was in-between the difference. (4% off of both measurements).

Erik

posted 09-05-2006 12:52 PM ET (US)            

It is also possible that the DGPS was intermittent or not working at all. That should have produced erratic readings on Steve's unit.

There is another possibility. GPS measure location in 3 dimensions, thus vertical as well as horizontal distance and speed. Could one of the boats have been on the edge of porpoising, thus moving up and down as well as forward?

If none of those possibilities is the case, I am baffled, too.

posted 09-05-2006 01:14 PM ET (US)            

I have previously checked my WOT speed with my Garmin handheld, 12-channel, with WAAS enhancement and observed 42 MPH at 5700 RPM with a light load. Yesterday with a medium heavy load I was only able to reach 5500 RPM. I did not play with the engine trim at all. At a steady 3500 RPM, my speedometer was showing 25 MPH, while the GPS showed 26.5 to 26.7 MPH.

I can only conclude from this test that Jim's 12-channel GPS is probably much more accurate than my old 5-channel unit.

Steve

posted 09-05-2006 01:27 PM ET (US)            

The manual for my LMS-350A shows under CALIBRATE SPEED:

The speed display on the sonar screens comes from the optional speed sensor - not the GPS receiver.

So the speed is actually coming from the paddle wheel!

posted 09-06-2006 10:58 AM ET (US)            

Guys,

posted 09-06-2006 01:46 PM ET (US)            

A pitot tube is a small rubber or plastic tube, maybe 1/8" inside diameter that runs from the backside of a hole that faces forward, either on the front of your motor or on the front of a bracket that extends below your transom, to what is effectively a pressure guage that is calibrated as a speedometer on your dashboard. More speed, more pressure, higher MPH readings on the speedo.

A paddle wheel is exactly that, a paddle wheel that is mounted on the back of the transom, low enough so the water running under the hull turns it, and the MPH readings on the speedometer are based on the number of revolutions of the paddlewheel.

posted 09-06-2006 08:46 PM ET (US)            

I have had the opportunity to use many different handheld GPS units since the early 1990's and I have taken all of them out on the road to check them out for speed accuracy relative to my vehicle's speedometer. My original Magellan5000 was not very good (too long ago to remember the details, but it was truly a POS and it cost me over $1000US back in 1994).

More recently my old trusty Garmin48 (circa late 90's) has always tracked very close to what my speedo said. Ditto for a GarminMap76, a Garmin etrexVista, and a Garmin72 (all circa 2003). There is (as someone mentioned previously) an update lag with the G48 (it is quite outdated compared to the new models) but once you hold a steady speed for about 5 seconds the GPS computer catches up. The lag is not quite as noticeable on the newer units.

posted 09-06-2006 08:47 PM ET (US)            

Paddlewheel speedometers usually are electrical. The paddlewheel contains magnets which induce a pulse in an attached coil. The faster the paddlewheel spins the higher the pulse frequency. Various electrical techniques are used to derive a speed reading from the pulse frequency.

posted 09-07-2006 01:42 PM ET (US)            

Also, somewhere I recall reading an accident report on a airplane near disaster in which the culprit was found to be a blockage in a pitot tube sensor. It was throwing off the plane's computer controls.

As for the GPS, I am actually somewhat pleased to learn that there was some non-GPS problem on BACKLASH contributing to the variance in the speed readings. I do feel sorry for Steve--losing 5-MPH off your top speed is never a good outcome! :-)

posted 09-07-2006 09:15 PM ET (US)            

http://www.rvs.uni-bielefeld.de/publications/Reports/aeroperu-news.html

posted 09-08-2006 12:28 PM ET (US)            

Not to be picky- ;-)

posted 09-08-2006 01:42 PM ET (US)            

The pitots are heated and yes on rare occasion one can be plugged thus the redundancy. Techinically, airplanes don't care about ground speed. The wing needs airspeed and that is the data provided by the pitot and is super critical to the pilots. Imagine landing with a 50 kt tailwind and the stall speed is 110 kts and approach is usually 1.3 times that and trying to use a GPS to provide AIRSPEED--no way----the pilot would see a approx 150 MPH ground speed but the wing would see only 100 knts and --can you say STALL? BUT--the good old pitot does not care about any that--it only sees the speed of the aircraft via pressure in relation to the air through which it moves and therefore --by it's nature and operation and simple physics---only provides AIRSPEED, the speed of the aircraft through the air. The DME, GPS and on some aircraft Inertial Guidence are used to calculate ground speed--for navigation purposes but for flight purposes--airspeed by pitot/static system.

posted 09-09-2006 12:02 PM ET (US)            

"They deduce speed from measurement of a pressure differential."

Inanimate objects do not deduce, "to deduce" is a function of intellect and reasoning.

posted 09-09-2006 10:15 PM ET (US)            

The two boats are travelling at different rates of speed at any given moment.

Now, if the two speed measuring devices were on the same boat, we might have a conversation about the accuracy of the two devices. When they are on different boats there is no possible correlation between their readings, because at any given moment they are indeed travelling at different speeds. That fact that one or more of the participants thinks they are going the same speed is irrelevent.

Did I miss something here?

Roy

posted 09-09-2006 11:11 PM ET (US)            

This is one terrific thread!

It began with 2 similar Whalers drag racing with an observed 5 mph difference in speed both who thought they were using GPS's to measure SOG (sorry Jim for using abbreviations). Then the plot thickened with the possibilty the boats might be using different speed measurements (knots -vs- mph). But thanks to the truthfulness of one the drag racers the difference was found to be a paddle wheel sensor that had been used to measure SOG (mystery solved, BUT). The story got even better, a Whaler that could do 63 knots (followed by a BS call) which was soon followed by a plane crash and finally a military issue brought about by some poor sucker who posted some inaccurate info from another board. Thank God someone pointed that out.

Can't wait to see where this thread goes next. Maybe a classic Whaler armed with cruise missles & a GPS thats positional data is only good to plus or minus 11 feet.

WoW (semi-shout) ..... one last idea. Maybe someone can find a way to incorporate the ongoing Boozin & Cruisin Foodfight Thread into the story!

posted 09-09-2006 11:15 PM ET (US)            

You can say that a anemometer measures the wind speed by rotating in proportion to the speed of the wind. When that rotation is connected to a gauge and a pointer that show the wind speed on a scale, it is the intellect of the person who designed the instrument that measures the speed.

That is how the speed is deduced from the measurement of pressure differential. It is not a direct measurement of speed at all. The speed has to be calculated or deduced from the pressure differential.

posted 09-10-2006 09:06 AM ET (US)            

"Let me see if I have this straight: two boats are travelling at different speeds measured on different speed measuring devices and someone is at a loss to explain why the readings are different?"

This analysis was going well right up to the point where the critic wrote

"...and someone is at a loss to explain why the readings are different?"

No, there is no lack of understanding why the measurements are different. This is the entire focus of the inquiry. Let me explain:

The underlying assumption has been that measurements of speed by GPS techniques were very accurate, so much so that it was folly to question them. An example was provided in which there was a gross error in measurement of speed via GPS techniques. Further investigation allowed for the discovery that the error was not due to variation in GPS accuracy at all, but was caused by another situation. It remains for further testing to discover if two GPS units will measure the same speed for the boats involved in the test.

Next we have a summation of the thread which speculates on where it will go next. This critic correctly summarizes that the discussion of the thread has been changing, however, in terms of his prediction on where the next topic of the thread would turn, this critic is in error. He fails to anticipate that the next topic of the thread would become a discussion of its critics.

posted 09-11-2006 01:06 PM ET (US)            

We should have tested this with the four boats we had on Lake Erie on Saturday. I'm sure the results of that test would have certainly introduced a new dimension to this discussion!

posted 09-11-2006 01:42 PM ET (US)            

Dave--It was too rough Saturday for testing. We want to pare this accuracy down to the 0.1-MPH range. There was so much vertical motion on the boats that it would have affected the speed computations. With the wave height on Lake Erie (2-4 footers) we'd have needed a SPORT 150 to take on those monsters and run at 25-MPH and stay on an even keel.

posted 09-12-2006 04:12 PM ET (US)            

What is most important here and not discussed as I scan this thread is absolute accuracy of TROLLING SPEED!!! I swear my buddy is messed up because he trusts his Lowrance GPS for trolling speed read-out and the last two times we been skunked while rods were a -poppin all around us.

What about trolling speed? THAT is much more critical than accuracy at WOT.

Ya gotta win that race - it's who crosses the line first, Jack!

posted 09-13-2006 02:20 AM ET (US)            

You need to be careful using a GPS speed-readout (directly) for trolling as you are measuring the speed over ground. You need to know the speed through water to find the correct trolling speed. You can compensate for this by adding or subtracting the drifting speed to the GPS readout and also check you paddlewheel speed to confirm.
I.E. if you are drifting along with the water flow at a speed of 2 knots your GPS will show a speed of 2 knots. The paddle wheel should show you a speed of 0 knots. Throwing out a lure in this case would have no effect and most likely not catch a fish. If you want to troll against the current (which is always better than with the current)at a waterspeed of 3 knots your GPS should be showing a speed of 1 knot and the paddlewheel speed should be 3 knots.

The accuracy of the paddlewheel measuring device will most likely be less accurate at slower speeds, often making it difficult to find the sweet speed. Checking out the lure action beside the boat is the best way to find the speed you need for a certain type of fishing lure.

posted 09-14-2006 09:17 AM ET (US)            

Even if jimh and Steve didn't have WAAS or DGPS on,
ionosphereics wouldn't account for the difference they saw.

posted 09-14-2006 10:25 PM ET (US)            

What can I deduce from your experiment? That Steve's boat is faster than your boat.

"I would judge the difference to be less than one foot per second"

Be for real. Exactly how did you manage to gage that while going at WOT? Could it have been half a boat length per second? It doesn't take long for half a boat to go by when your blasting along on the water at 44 mph. If it took "ONE-ONE-THOUSAND" for ten feet if Steve's boat to go by, you have the following result:

10 FEET/SEC X 3600 SECONDS/HOUR X MILE/5280-FEET = 7 MPH

Wha-La, mystery explained!

Steve spanked your monkey and you're trying to muddy the waters with all of this mumbo jumbo, all the while avoiding the real pickle: that Continuouswave needs a repower, or rather, more power to keep up with Backlash.

The truth ain't pretty

Roy

posted 07-22-2007 12:14 PM ET (US)            

Water skiers have traditionally used pitot speedometers. As pointed out these have a small bracket with a hole facing the direction of movement. They measure the dynamic pressure of water, which is 1/2 Ro v square, with ro being the water density. A good rule of thumb is that the dynamic pressure of water is about 700 pounds per square foot at 20 mph. The size of the hole in the pitot is immaterial, as is the depth of submersion. Because dynamic pressure increases as the square of speed, these devices can be quite accurate at high (>20) speed, but bad for lower speeds. If you look at a cheap pitot speedometer, the distances between displayed speeds gets larger as speed increases, making it less prone to error at higher speeds.

The pitot unit of choice for water skiers was the Airguide Jeweled Contralog speedometer. It had a "movement" within the head that rectified the exponential spacing of of cheaper units brought on by the physics of squaring the speed. They had in addition an "averaging tube" which was an air tank that kept the movements of the dials dampened and smooth. The contralog units were very easy to read in the critical 15-36 mph range where all skiing occurs. They were in additional adjustable and were in fact always adjusted so that the trap times through a measured distance (the slalom course usually) agreed with the display. Adjustments were necessary for water temperature and air pressure. Obviously, current would affect trap times over the bottom, but the discussion of how to compensate for current when skiing a slalom course is beyond this discussion. Two units were used in most boats. In practice, even after adjustment, it is not uncommon to see a 1 mph difference between these units.

Pitot speedometers are especially vulnerable to debris fouling, which may affect response time and reading. Pitots almost never read high, except when they are plugged and the boat slows down and they keep their old reading.

Lately, high end skiers have been using paddlewheel speedos. They have the advantage in not being exponential in read out and being more resistant to fouling. They are also electronic, and their output can thus be used in speed control servo devices. I have never seen a paddlewheel speedo read high but I suppose anything is possible with electronics.

posted 07-23-2007 04:13 PM ET (US)            

On our Albin, I've noticed quite a bit of variation between the SOG (taken from the GPS) and the Speed (taken from the paddlewheel). Being in Puget Sound, it's pretty much a given that anywhere you go, there will be some amount of current. It's often hard to determine exactly how much of the difference is due to current and how much is due to the paddlewheel's inaccuracy. Since current here rarely flows in a straight line, you don't know when you're really going with or against the current, as opposed to going in some tangent direction to the current.

What I have done is take the boat out in open, relatively steady water and run speed trials in large circles. I've come to the conclusion only that the Speed reading is consistently lower than that of the GPS. By how much is hard to predict, since the variation seems to itself vary by how much growth there in on the bottom of the boat and the paddlewheel itself. Further complicating matters is the fact that the boat is cleaned bi-monthly by a diver, and it's often difficult to tell if the bottom has been cleaned since your last use. You also have no way of telling whether the diver bothered to clean the paddlewheel properly on that particular visit.