GPS drifting ? At anchor

[Dave_S]

I spent most of last night trying to work out if my anchor was dragging.

I was anchored in 2-4 meters of water on clean sand in strong current with 15 knots of breeze with 20m of chain. I set the anchor alarm location by scope and direction after I set the anchor rather than when i dropped it.

The alarm went off and the trace looked like tide change, I checked my visual markers and it looked Ok. I thought I had screwed up my anchor location so I reset it to the middle of the trace.

The alarm kept going off and the trace kept getting bigger and bigger, I don't like to blame the equipment when things don't match up but it said I had drifted over 180m and my visual markers didn't look like I had moved much but it was a moonless night. So after a dodgy night's sleep I awoke to find I hadn't dragged but that the equipment was wrong.

Does GPS do this sometimes or is the tablet I'm using crap at reading the satellites or is "anchor light" software crap?

[Tetepare]

Your boat weighs 7000kg and has a fair amount of windage, while you were in 15 knots of air with current.

True, you had 5:1 scope, but it was only 20m; in fact, if one considers the hight of the gunwhale or anchor chain attachment point, your scope drops to approximately 4:1.

sand is not the best holding ground. You didn't say if you had any non-chain bridle or shock absorber.

Given all that, I wouldn't sleep well, as I'd anticipate a high likelihood that the boat would drag anchor.

[Cavalier]

OK.. this is a question that comes up often on this forum, and has a simply answer.

Remember that your "computed" GPS coordinate is the result of a calculation of at least 3 satellite signals. A very rough fix will be given unless all 3 satellites are located in the same region of the sky, in which case you'll end up with a very poor positional result/accuracy. Typically 4-5 satellites with a decent geometric spread are required for an ok fix, and with each additional satellite the signal-to-noise ratio of each signal can be averaged such that the SNR is reduced.

Even though a marine-grade differential GPS receiver may have the ability to 'see' 12 or 16 or 24, stations/satellites, according to their specifications, it doesn't mean than it can actually see that many.

At the moment, there are 31 GPS satellites covering the entire planet, they are not homogeneously distributed. At any one point, your unit may see only 5, and then 20minutes later later, be able to see 7, for example.

When a satellite 'sets' below the horizon or a new one 'rises', the computational algorithm in your unit can temporarily crap-out and cause spurious positions, especially if only a small number of satellites are visible to the unit. This is why people sometimes see that, when looking at their GPS SOG log, that their maximum speed during an offshore passage was '64knots' (wow, must have been surfing down a wave!). The usual SOG is actually calculated as a rolling average. A single spurious position calculation can lead to an instantaneous position being 100m away from where you are - averaged over a second, would result in your max SOG showing some silly high speed for a single reading. How could one hop 100m to the west and back in only 2 seconds? the only answer, according to the algorithm is that your velocity must be high..

Regardless of what the spec-sheet says on iPhones/iPads etc, they are not true marine GPS receivers.. their positional accuracy will vary and as their mission-profile is based upon land use (cars and feet). Remember that your device was probably located close to or under the waterline, so its view of the sky would be further occulted (compared to your boat's GPS antenna) - i'm guessing that if you placed the device on the ground overnight you'd find the same erratic wondering going on..

My guess is that your device's computed GPS position is wondering all over the place during the night, as satellites rise and set, and that your boat's position is not actually wondering - ultimately, you verified your position with Mk1 eyeball 'transits' which is the right thing to do. Sit there for log enough and continental drift may become a factor.. ;-)

[B25Matt]

Judging from your track plot I would say you were dragging. if it were due to GPS position errors, you would see significant jumps in your position and I don't see that.

[zeehag]

what did your eyes see when you took a bearing each time--did you use same distant item for each bearings check/? did you take bearings?? eyeballs are the ultimate betrayor. same gut feelings.
if you have that gnawing sensation. move boat. you will rarely be incorrect.
happy sailing.

[Stu Jackson]

Originally Posted by Cavalier

>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>>>>>> - i'm guessing that if you placed the device on the ground overnight you'd find the same erratic wondering going on..>>>>>>>>>>>>>>>>>>>>>>>
>>>>>>>>>>>>>>>>>>>

First thing I did with my "marine" LOL Garmin GPSMap76Cx when I bought it years ago was to turn it on and let it sit on a table. It moved around a few meters.

I just tried it again on my kitchen table, it still moves 50 feet when sitting still.

I set the anchor alarm location by scope and direction after I set the anchor

Perhaps you didn't set it "perfectly."

I am not one who uses "anchor alarms" but have read many threads about them. In San Francisco, before I moved, for 35 years I anchored out frequently.

Perhaps those who use anchor alarms could chime in.

[S/V Alchemy]

Another factor is if you have surrounding hills which can "mask" the satellites, the number of which may be visible to your receiver can vary enormously.

It's not so much a flaw of the GPS as an acknowlegement of its limitations. I'm sure I'm not the only person to have been cruising at five knots, only to have the GPS report a SOG of 55 knots for a few seconds and a sudden relocation of a quarter mile. As has been pointed out, GPS satellites clustered either close together or on the horizon (or blocked by hills or trees or buildings) can do this. GPS signals aren't strong.

So while you may, in fact, have been dragging, without taking bearings to land-based marks, or by having, say, a guard ring on a radar (I guess the anchorage would have to be otherwise empty) which told you when your "anchoring circle" touched the beach or the cliffs, that's hard to confirm.

[Mike OReilly]

It looks like a slow drag, but if your morning transit sights were identical to when you set the anchor, I dunno… If it were GPS error I’d expect a more random pattern. This one seems to have a clear path.

I don’t use anchor alarms, per se, but if I’m concerned about my anchor I sometimes mark a point on my Garmin 76Cx and sometimes my iPad running iNavx. I’m used to seeing statistical noise in the positioning plot, especially with the iPad. But if I ever saw a clear trend I’d be up on deck in a shot.

I think Tetepare’s analysis is sound, and Zee is right. Gotta use an actual bearing, preferably with a sighting compass. Eyeballs are easily deceived.

[B25Matt]

I agree with your conclusions, however, there are some factual errors. I suspect that you know this and were simply boiling down a complex problem into something understandable for the layman. The following is for the benefit of those who are interested in the inner workings of GPS.

To set the record straight a GPS receiver requires at least four satellites to come up with a position solution. You are solving for four unknowns x,y,z, and time. Time being the error in the local device.

A GPS satellite does not transmit its position of where it is right now but rather a time stamp of how long ago it passed overhead a ground based reference station. Hence the need to also broadcast an almanac of all of the satellite trajectories. It takes 12.5 minutes to transmit this to a GPS device. Hence, the incredibly long time for a device to cold boot where as a WiFi enabled devices can grab this from the internet and get a position fix almost instantaneously.

Getting back to time, if the local time of day is off by even a second, it would calculate a position that would be off by possibly a 100km given the speeds at which the satellites are orbiting. An accurate position fix is an artefact of your local device calculating the time of day correctly, down to the nano-second.

All this is to say that if you happen to have a synced atomic clock in your back pocket, then indeed, three satellites is all you require for a position fix and you are correct after all!!!

My experience is that most (not all for sure) mobile devices have pretty good GPS's and are far more accurate than my "marine" GPS. Almost all new GPS models support US GPS, Glonast, WAAS, etc. and this includes my $10 USB gps device. Limitations as per your discussion above. Clears sky antenna views are critical. GPS antennas have poor reception inside boat (and houses too) with multi-path giving bad position fixes. You will see lots of random big position jumps in such cases (+/- 100m) and I don't see this in the posted track plot. I say he was drifting.

Matt

[farmall.dude]

Unless you have an expensive Sub-Meter or Sub-Foot GPS, you're going to see that. It's only as accurate as it's designed. I've done the same test with my Handheld Garmin on the dash of my truck (parked). Over the course of 24 hours it "Moved" 10-20 meters in every direction. It looked like a star on the screen LOL. I'm guessing only the military (or very wealthy folks?) have GPS accuracy to the square inch. The rest of us have to deal with the drift.
I always thought the GPS satellites were in a geosynchronous orbit. Meaning their position in space is constant relative to their position on the earth. I was surprised to learn they are not. Thanks for the education!

[noelex 77]

Originally Posted by Dave_S

I spent most of last night trying to work out if my anchor was dragging.

Yes, you were dragging.

This type of slow drag is quite common and it is not unusual for owners to unaware of what is happening. At least you suspected something was amiss, which is good to see .

[kas_1611]

Most "smart" devices available over the last couple of years are able to not only receive the US GPS system but also the Russian GLONASS and the EU Galileo systems too. Some are even able to pick up the Chinese system. My 2017 smartphone can see anything upwards of 20 satellites at a time although frequently only "grabs" half a dozen.

The basic principle of the system is 3 fixes are used to give a position on a flat surface. A 4th fix will fine tune this fix and give altitude above a datum (WGSD84). Additional fixes continue to fine tune the position but will never give a truly precise fix even with military grade equipment (hence why the US military are still able to miss their target sometimes!). Thanks to the Clinton administration "selective availability" was switched off some years ago allowing us mere mortals access to <1m accuracy.

Due to many variables including the thickness of the atmosphere and the "excitability" of the ionosphere the signal from the satellites in Low Earth Orbit are constantly varying. This is what causes the error in the fix and creates the apparent motion of a vessel (or SatNav unit) at rest. Some apps and hardware have a filter option to smooth out the variations thereby reducing the apparent motion. Check the settings of whatever application or hardware you are using to see if this is available (Kalman filtering is a common one).

When I set my anchor alarm I always include an extra 10m on my boundary to account for any GPS drift but that said I have seen the error be as much as 40m from time to time. A good anchor alarm will allow you to "ignore" a number of erroneous points to reduce false alarms. Best thing to do is dig around in the settings for filters and sensitivity and play with these to reduce the chance of false alarms but to also trust your gut and the Mk1 eyeball. If you think you might be dragging don't take any risks, get up, get dressed and get watching.

[noelex 77]

Originally Posted by farmall.dude

Unless you have an expensive Sub-Meter or Sub-Foot GPS, you're going to see that. It's only as accurate as it's designed. I've done the same test with my Handheld Garmin on the dash of my truck (parked). Over the course of 24 hours it "Moved" 10-20 meters in every direction. It looked like a star on the screen LOL. I'm guessing only the military (or very wealthy folks?) have GPS accuracy to the square inch. The rest of us have to deal with the drift.

Modern GPS units with a clear view of the sky and away from multi path distortion will do better than this.

Many owners install gps units in positions where the results are similar to what you are experiencing with the unit on dash board of your truck. For an overall position this is fine, the accuracy will be better than the charts most of the time although SOG and tide calculations are likely to have significant errors.

A good installation can achieve better accuracy (although there will still be errors) which for some applications such as a sensitive anchor alarm without any false positives can make a significant difference. A 20m error is too much even without a satellite differential fix.

[SURV69]

I'm thinking you're pretty far south and that the "GPS", system works better much more north. You possibly have a good number of Sat's near the horizon, which means more atmosphere .... more error in time, and possibly less Sat's. In the U.S.A., we generally get a full, or nearly full compliment of Sat's above the horizon.

Is it possible to mask the lowest 20-30 degrees on your tablet? Also keep in mind that where your tablet is kept(while at anchor), might tax-out the usefulness of a patch antenna(somewhere on your tablet), and any metal on the boat that also interferes with Sat's. Maybe a Delorme(or other), Bluetooth antenna would give a more precise, and maybe a more accurate location.

Aside from that, in time, your position should begin to average in and centralize of a point(in the north, this should mean to average 10-20 feet in somewhat circular pattern. If I remember right, there tends to be more error in Lat(I think), or Lon ... offhand can't remember).

Unless you have a whole lot of anchor rode, your pattern should not only be tighter, but crossing and recrossing a roughly central location .... How long is your rode?

If these anchor programs could take into consideration, wind, or even forecasted wind, the patterns would be even tighter.

[Kelkara]

Originally Posted by B25Matt

To set the record straight a GPS receiver requires at least four satellites to come up with a position solution. You are solving for four unknowns x,y,z, and time. Time being the error in the local device.

This is true ... but since most of us don't actually care about the z-coordinate (altitude) most GPS devices will happily calculate a 2D fix from just three satellites.