Speed Through GPS Versus Old Fashioned Paddle Log

[chala]

Originally Posted by transmitterdan

So GPS speed pucks are very accurate

I know for a fact that that if I drive up or down a hill with my car the car GPS speed will be inaccurate. Also when the boat is ashore, due to the relative accuracy of the GPS, the COG will be everywhere and if Tracks is on, the track distance after 24 hours recording can be as high than 10 nautical miles with the boat not moving at all.

[noelex 77]

Originally Posted by chala

According to this thread:

If you go to the quoted thread Dockhead in post 4 gave a great description of the difference between true and ground wind:
http://www.cruisersforum.com/forums/...ion-73563.html

In my view most cruising sailors would be better if they switched to ground wind. The main reason is ground wind is derived from SOG . The GPS does not have the systematic errors (such as under or over reading, reading slightly differently of different tacks) this means that ground wind is much more accurate and consistent.

I would also argue that the combination of apparent and ground wind is at least as useful to a cruising boat as apparent wind and true wind. We have been stuck with true wind only because traditionally instruments could not display ground wind.

However, this is largely academic as few instrument systems allow the user to switch between true and ground wind. This is a pity as the instrument receives all the data to make the calculation and it would be an easy software fix to give the option.

Next time you are at a boat show talking to instrument representatives badger them to incorporate the option. Even if you are a racing sailor and prefer true wind, a switch to ground wind gives you a useful wind display if the paddle wheel gets fouled.

[Seaworthy Lass]

To save hunting, here is Dockhead's post from 2011 describing the difference between ground, true and apparent wind.

It is a nice clear description. The only thing I disagree with is the comment I have highlighted in bold and underlined. Dockhead, didn't you mean you wouldn't care if you had a 1020 knot ground wind in the same direction not true wind?

This example does illustrate the importance of knowing ground wind (rather than lack of importance) when sailing in areas with strong currents. If the current changed or you decided to secure yourself to land, then true wind info would be very misleading on its own.

Originally Posted by Dockhead

Apparent Wind = the wind speed and direction in relation to your boat. So in a dead calm motoring ahead at 10 knots, you would have a 10 knot apparent wind at 0 degrees.

True Wind = the wind speed and direction in relation to the WATER. In the example above, you would have Apparent Wind of 10 knots and True Wind of 0. Your instruments calculate True Wind by taking Apparent Wind (the only thing your wind instruments measure directly) and taking out the vector for speed through the water. Which is why you get no true wind readout if you have not got STW data -- speed through the water data -- in your network.

Ground Wind = the wind in relation to land, not water. In the example above, if there is no current and no leeway, Ground Wind = True Wind. If you are motoring directly against a 5 knot current, however, then you have Apparent Wind of 10, True Wind of 0, and Ground Wind of 5. To calculate Ground Wind, your instruments take out the vector for SOG, rather than STW. Ground Wind is of no relevance to sailors unless we are sailing in a strong tidal current and are trying to anticipate what the wind will be like when the tide changes.


Before GPS, it was impossible to calculate Ground Wind. That is because the only speed data we had was STW. Possibly the idea of True Wind versus Ground Wind arose because instrument makers who suddenly had access to SOG data on the network needed some way to differentiate true wind calculated that way versus True Wind calculated using STW.

True Wind, as calculated with STW data, is actually what we really need, since our boats sail in the interface between water and air. If we were sailing in a 1000 knot current, we wouldn't give a damn that we have a 1020 knot true wind in the same direction. We would only feel or care about the 20 knots differential between the water and wind -- that's what we sail in. That is why no one bothered, I guess, to change the definition once instrument makers had access to SOG data.

This will be rather academic for anyone who sails in waters without strong currents. For them, SOG and STW, and therefore True Wind and Ground Wind, will be more or less the same most of the time, but for perhaps some effect of leeway. For others, like me, who sail in the English Channel where the tide rips at up to 10 knots (Alderney Race), the disinction is not academic.


Don't be confused by the fact that meteorologists use True Wind to describe the difference between wind and land. Of course they do -- they are sitting on land, not on water. That in no way contradicts our usage of True Wind as being the difference between wind and water.


You asked which kind of wind is expressed in relation to North, what kind of North, or your bow. Don't get confused by this. "True" in "True Wind" has nothing to do with True North. It means true as in -- without the effect of the boat's motion. Any wind can be expressed any way you want -- in relation to your bow, in relation to either kind of North. But your instruments will generally tell you wind in relation to your bow, which is what you need for sailing. Ground Wind, which you don't need for sailing -- which is irrelevant for sailing here and now (but may be relevant for planning), would be useful to know in relation to North, to the extent you care at all. Which North -- why that depends on how you think, and how your instruments are set up. I always use Magnetic for everything, so that is what I would want to know, personally.

When floating, the simple way that I think of it is that the vector sums of these apply with the direction of the vectors reversed where appropriate.

TW = GW + current
AW = TW + STW + leeway = GW + SOG

Where:
TW = true wind
GW = ground wind
AW = apparent wind
SOG = speed over ground (from GPS)
STW = speed through water (from log)

[Dockhead]

Originally Posted by Seaworthy Lass

. . . The only thing I disagree with is the comment I have highlighted in bold and underlined. Dockhead, didn't you mean you wouldn't care if you had a 1020 knot ground wind in the same direction not true wind? . . .

Yes, correct. I am going to go back and revise it. Thanks for catching it.

[Dockhead]

Originally Posted by noelex 77

If you go to the quoted thread Dockhead in post 4 gave a great description of the difference between true and ground wind:
http://www.cruisersforum.com/forums/...ion-73563.html

In my view most cruising sailors would be better if they switched to ground wind. The main reason is ground wind is derived from SOG . The GPS does not have the systematic errors (such as under or over reading, reading slightly differently of different tacks) this means that ground wind is much more accurate and consistent.

I would also argue that the combination of apparent and ground wind is at least as useful to a cruising boat as apparent wind and true wind. We have been stuck with true wind only because traditionally instruments could not display ground wind.

However, this is largely academic as few instrument systems allow the user to switch between true and ground wind. This is a pity as the instrument receives all the data to make the calculation and it would be an easy software fix to give the option.

Next time you are at a boat show talking to instrument representatives badger them to incorporate the option. Even if you are a racing sailor and prefer true wind, a switch to ground wind gives you a useful wind display if the paddle wheel gets fouled.

Well, Maretron instruments display Ground Wind. I want (and have) Ground Wind at the nav table for longer-term (longer than one tide) planning.





But I don't think it's a very good idea to use Ground Wind for sailing. It's not what you sail in. It will lead to big mistakes in areas with strong tidal streams. You can be 5 knots off, in a place where the tide regularly reaches 5 knots, like in the Channel, and the direction may be significantly different, too.

Maybe otherwise if you're sailing somewhere with very weak currents, such that STW and SOG are more or less the same. I guess where the currents are never more than a knot, many sailors will get better results with Ground Wind just because SOG measurement is so much more accurate than STW with the typical not frequently enough cleaned paddlewheel log.


STW measurement is tricky. I have the same expensive ultrasonic log which David M has on his research vessel -- Airmar CS4500. Even this device is often out more than a knot. I don't really understand why, except that since it depends on tracking particles in the water stream, maybe some water is too clear, or something.

[Seaworthy Lass]

If STW is via a simple paddlewheel then as I understand it, leeway is not being taken into account when true wind is being determined (unless angle of heel is taken into account and estimates made, as some sophisticated systems do). This can lead to significant errors.

SWL

Edited to add:This also raises the issue that STW from a log is not actually speed through water, it is just the forward component. It is not as valuable info as it seems.

[Dockhead]

Originally Posted by Seaworthy Lass

If STW is via a simple paddlewheel then as I understand it, leeway is not being taken into account when true wind is being determined (unless angle of heel is taken into account and estimates made, as some sophisticated systems do). This can lead to significant errors.

SWL

Edited to add:This also raises the issue that STW from a log is not actually speed through water, it is just the forward component. It is not as valuable info as it seems.

Yes, that's a very good point.

One advantage of ultrasonic logs is that they DO give STW in any direction, unlike a paddlewheel.

However, the True Wind calculations assume that you are actually moving through the water in the same direction as your heading. As you point out -- that does not account for leeway.


Now what would be really cool would be an ultrasonic log which could also determine direction of travel through the water. I would think that technically this would be trivial. This could give you some really fantastically useful data, like a direct measured output of leeway, and would give you a more precise calculation of True Wind, overcoming the problem you mention.

This would be the actual water-referenced counterpart to COG. Heading is NOT an exact analogue to COG, as you insightfully point out. It is assumed to be by our instruments, but it's not true.

[Ericson38]

Here is a spread sheet for the course to steer problem in the Chesapeake, CTS was 244.8 degrees, not the 250 estimate I used. Uses law of sines.

Coming down from Reedville Va, heading back to Lake Jackson Creek, after sunset. Tide is running out of the bay, as predicted by the chart plotter and confirmed by the set/drift out of the B&G Hyrda, which is using the GPS SOG and COG, the flux gate, and the STW sensor. Chart Plotter agreement is nice, haven't not seen the content and process using its internal current/tide database ! New Garmin 741, had it a month.

Pass Windmill Point Light 1 mile to starboard, steer directly to Red Nun '2R', water is deep between these two, no channel to maintain. Having a nice 8 knot SOG ride down with 6.5 STW until then. Now taking a line to Jackson Creek inlet, turn to 230 and directly towards R '2'. Plotter gives me bearing to R'2', but that is not the course to steer, as current is going to set me south. Down to 5.5 knots STW, current drift according to B&G is 1.5 knots, set is 160 true, which is a 90 degree cross current heading to R'2', which marks just about the 1/2 way point to the creek entrance. Course to Steer (CTS) is in the 250 deg range at this STW, as this is now effecting by the drift rate and my STW rate. B&G applying the law of sines for me. The width of this section of the channel is 100 yards. I pass directly south of R'2' and observe the plotter's course for flashing Red '6' post. Current still running out of the Rappahannock River here, as well as the Bay itself, so set and drift are about the same on B&G display. Knowing distance to Red '6' and my STW and Drift to the south, maintain the same course to stay in the channel. At arrival of R'6', turn to about 340 true, drop STW down to 4 kts to head to day board R'2', between two duck blinds on shore and a couple of clam farms on either side. Current at 'R' is down to 1 kt, running out of Jackson Creek. At this dayboard, the usable channel width is around 25 feet (daymarkers either side at 50 feet separation) and is shaped like a snake for the next 1/3 mile, still we get to Deltaville Marina, where it widens out to an anchorage in front. At this point, just steer between dayboards the whole way, note how close their shadows look to where the Garmin places them, make mental notes for future reference. Garmin shows current out of Jackson Creek close to my STW-SOG difference, so again, Garmin is getting a high grade tonight.

The whole purpose in the above was to stay close to the right side of the channel, and even if I could have drifted out of it, resulting in a last minute hook to starboard to regain it, the practice itself is important.

Could have just plugged in the waypoints of these nav aids and put the ST70 in track mode, which will keep XTE within 300 feet or better, but that is too much lateral distance for these particular confines.

BTW- ST70 Control Head Manual claims the AP does compensate for tidal drift in track mode to attempt to minimize XTE error, using STW and SOG, HDG and COG.

[Ericson38]

Originally Posted by chala

I know for a fact that that if I drive up or down a hill with my car the car GPS speed will be inaccurate. Also when the boat is ashore, due to the relative accuracy of the GPS, the COG will be everywhere and if Tracks is on, the track distance after 24 hours recording can be as high than 10 nautical miles with the boat not moving at all.

GPS receiver is directly calculating your position in ECEF coordinates, the same frame as the satellite positions and orbits. Then the receiver converts the ECEF X,Y,X into your local reference frame, say LLA. So the calculated speed is in 3-D. Going up a hill or down, your 3-D speed accuracy is not effected. Going up a 45 deg incline, your speed on a horizontal plane is reduced by square root of 2. ( or .707).

Depending on where your boat is now, there can be a lot of wander in the static position calculated, from multipath induced by nearby structures (masts and corner reflectors are good sources of this). Over 24 hours, this could end up as quite an accumulated trip log distance, but all the moving was within a 10-20 foot radius of your boat.

[valhalla360]

Originally Posted by Ericson38

Here is a spread sheet for the course to steer problem in the Chesapeake, CTS was 244.8 degrees, not the 250 estimate I used. Uses law of sines.

Coming down from Reedville Va, heading back to Lake Jackson Creek, after sunset. Tide is running out of the bay, as predicted by the chart plotter and confirmed by the set/drift out of the B&G Hyrda, which is using the GPS SOG and COG, the flux gate, and the STW sensor. Chart Plotter agreement is nice, haven't not seen the content and process using its internal current/tide database ! New Garmin 741, had it a month.

Pass Windmill Point Light 1 mile to starboard, steer directly to Red Nun '2R', water is deep between these two, no channel to maintain. Having a nice 8 knot SOG ride down with 6.5 STW until then. Now taking a line to Jackson Creek inlet, turn to 230 and directly towards R '2'. Plotter gives me bearing to R'2', but that is not the course to steer, as current is going to set me south. Down to 5.5 knots STW, current drift according to B&G is 1.5 knots, set is 160 true, which is a 90 degree cross current heading to R'2', which marks just about the 1/2 way point to the creek entrance. Course to Steer (CTS) is in the 250 deg range at this STW, as this is now effecting by the drift rate and my STW rate. B&G applying the law of sines for me. The width of this section of the channel is 100 yards. I pass directly south of R'2' and observe the plotter's course for flashing Red '6' post. Current still running out of the Rappahannock River here, as well as the Bay itself, so set and drift are about the same on B&G display. Knowing distance to Red '6' and my STW and Drift to the south, maintain the same course to stay in the channel. At arrival of R'6', turn to about 340 true, drop STW down to 4 kts to head to day board R'2', between two duck blinds on shore and a couple of clam farms on either side. Current at 'R' is down to 1 kt, running out of Jackson Creek. At this dayboard, the usable channel width is around 25 feet (daymarkers either side at 50 feet separation) and is shaped like a snake for the next 1/3 mile, still we get to Deltaville Marina, where it widens out to an anchorage in front. At this point, just steer between dayboards the whole way, note how close their shadows look to where the Garmin places them, make mental notes for future reference. Garmin shows current out of Jackson Creek close to my STW-SOG difference, so again, Garmin is getting a high grade tonight.

The whole purpose in the above was to stay close to the right side of the channel, and even if I could have drifted out of it, resulting in a last minute hook to starboard to regain it, the practice itself is important.

Could have just plugged in the waypoints of these nav aids and put the ST70 in track mode, which will keep XTE within 300 feet or better, but that is too much lateral distance for these particular confines.

BTW- ST70 Control Head Manual claims the AP does compensate for tidal drift in track mode to attempt to minimize XTE error, using STW and SOG, HDG and COG.

I didn't open the spreadsheet or read thru all that. It just sounds complicated and I don't see someone keeping a spreadsheet running the whole time they are traveling to keep making calculations when the electronics do it for you.

We went into there once, it was blowing like snot. We just kept checking against the chart plotter that it had us going down the channel and then checked against reference points (bouys, daymarks, heading to landmarks ashore, etc...) Sometimes the bow was pointed at an angle to the GPS course but if it we kept going down the channel, it works out. Yes, we were aware of the currents and had a rough idea of thier direction and strength. In such narrow channels, we wouldn't use the autopilot but that's a different subject.

[valhalla360]

Originally Posted by chala

I know for a fact that that if I drive up or down a hill with my car the car GPS speed will be inaccurate. Also when the boat is ashore, due to the relative accuracy of the GPS, the COG will be everywhere and if Tracks is on, the track distance after 24 hours recording can be as high than 10 nautical miles with the boat not moving at all.

If you are regularly getting hits 10 miles from your true location, you need to replace your defective electronics. If it was a stray glitch you saw one time or you dock next to high voltage lines or some other oddity, maybe....

Yes, it moves around a bit when in a slip but the vast majority of the time it's dead on within a few feet and rarely more than a couple slips away.

[Don C L]

This thread should be published and made into an online navigation course.

[The Smokester]

Originally Posted by Don C L

This thread should be published and made into an online navigation course.

I agree as long as Dockhead curates it.

[Dockhead]

Originally Posted by valhalla360

I didn't open the spreadsheet or read thru all that. It just sounds complicated and I don't see someone keeping a spreadsheet running the whole time they are traveling to keep making calculations when the electronics do it for you.

I agree -- unlike a multi-hour CTS calculation to get efficiently through changing currents, which your GPS can't calculate, this problem is just to get the heading you need for a given COG.

The GPS does this calculation for you. You just steer until the COG read out from your GPS matches what you want. Or better yet, just lay down waypoints and use your pilot's track mode. Your pilot will keep you on the course line. You can set it to steer more or less actively -- that is, to be more or less tolerant of deviations.

However, I think it is extremely admirable to have such skills. Even if you never need them, because you have multiple backup GPSs on board as everyone these days does -- understanding this will really help you understand other nav problems. As we have seen, it is very easy to completely lose touch with the principles of navigation, when you let the electronics do everything for you. It even becomes incomprehensible to some that there are some navigation problems that our electronics won't do for us -- but there are some.

[noelex 77]

Originally Posted by Dockhead

But I don't think it's a very good idea to use Ground Wind for sailing. It's not what you sail in. It will lead to big mistakes in areas with strong tidal streams. You can be 5 knots off, in a place where the tide regularly reaches 5 knots, like in the Channel

The only thing your sails respond to is apparent wind. This is the simplest display shown by all wind instruments and subject to the least error.

As a second display if the instruments offer the option (as yours do, although will they display ground wind relative to the bow?) we can choose ground wind or true wind. To use your own (I hope hypothetical example) of 1020 knot ground wind and 1000 knot current, apparent wind and true wind would give no indication of the extreme conditions. Only ground wind would do this.

I think I would want to know about 1000 knots of wind.

However, the big advantage of ground wind is that is always accurate, always calibrated. There is no fouling or miscalibration of the log to give deceptive readings.

If you have a racing boat it is worth spending the time to ensure the log reading is spot on: identical on different tacks, calibrated regularly and regularly cleaned, but few cruising boats go to this much trouble. GPS units have improved leaps and bounds in responsiveness and accuracy over the last decade but many instruments only offer wind calculated from the frequently inaccurate log.

Ground wind is the best option if you want to answer questions like: Has the wind become weaker, stronger or changed direction? Does the wind match the forecast? Or what will be the wind strength and direction when I anchor or dock? True wind will not answer any of these questions correctly.

Information about both true wind and ground wind have their usefulness, but with the typical instruments fitted to cruising boats the option of accurate ground wind beats true wind that is vulnerable to bits of weed, or some hull fouling. At least companies like Raymarine could offer us the option.