Accuracy of Garmin 60cx

[+PFF]

I received my 60cx last week and decided to check its accuracy against a government benchmark. Unless the coordinates are listed as SCALED, benchmarks have coordinates which are accurate to a fraction of an inch. I selected EZ0645, where conditions are good (no tree cover).

 

For this test, I used the built-in antenna of the 60cx. The unit's averaging function was activated and a reading was taken five minutes later. This was approximately 400 samples.

 

Resulting readings, compared with the government benchmark:

 

LATITUDE:

Standard: N35 50 45.193

Garmin: N35 50 45.204

 

LONGITUDE:

Standard: W78 50 12.623

Garmin: W78 50 12.588

 

These readings were taken in the Decimal Degree mode and were manually converted to Degrees, Minutes, and Seconds. Hence, the extra digits in the Seconds position. The Decimal Degree mode does not increase the accuracy of the GPS receiver, but it allows the "rounding" process to be visible to the user.

 

Had I stayed in the DDMMss.s mode for the test, the unit would have rounded the figures automatically, and the display would have been precisely the same as the coordinates of the government benchmark. Needless to say, I am very pleased. For anyone who has access to a GPS with an averaging function, the location of the cache you are placing should be very accurate! If there is tree cover, I recommend letting the unit average over a longer period of time.

 

I have an eight-foot tripod which supports an amplified antenna. [see Below.] As time permits, I will run some tests on benchmarks located beneath various levels of canopy--since that is the most common situation when placing a cache. An added benefit of the yellow tripod is that nobody ever stops to ask what I'm doing. I suppose this is because they assume I am a surveyor.

 

-Paul-

 

 

[+EScout]

Nice tripod setup.

 

Some comments: You want to choose GPS sites, or Adjusted Benchmarks. This is a small percentage of the total. They are usually in very open spots. I do not think you will find one with any objects preventing a very open view of the sky.

 

Choose Super-Accurate benchmarks by USA State and County.

 

The benchmarks are listed to one-hundred thousandth of a second (5 decimal places after the seconds.) Most GPSrs only show to the tenth of a second. Also, Decimal degrees in one-hundred thousandth of a degree (5 places as in ddd.ddddd), decimal minutes in one-thousandth of a minute (3 places as in ddd mm.mmm), and Decimal seconds in one tenth of a second (one place as in ddd mm ss.s) are all of about the same precision, so you can convert one to another, but do not add any more decimal places.

 

Rounding the benchmark's coordinates to fit our usual ddd mm.mmm format puts the rounded coords within a maximum radius of 3 feet from the benchmark.

[+PDOP's]

You could also compare your waypoint to the benchmark using a program like GeoCalc which will give you the horizontal difference. In your test case the difference is 0.941 metres. IMO this is more informative than comparing how the coordinates differ.

[+TheRoundings]

You could also compare your waypoint to the benchmark using a program like GeoCalc which will give you the horizontal difference. In your test case the difference is 0.941 metres. IMO this is more informative than comparing how the coordinates differ.

 

I agree and great program BTW. (Hadn't come across it before)

[tosborn]

Paul:

 

So tell us more about your tripod. Does it come to a point at the bottom so that you can set it precisely over the punch mark on a disc? Does it have a level on it so you can get it perfectly plumb? How are you using it? Where did you get it? Have you got a picture showing the whole thing? The antenna looks like the Garmin GA 27C and I guess you used one of the Garmin mounting brackets for the GPSr, or is the GPSr just hanging? Very interesting.

Edited September 24, 2006 by tosborn

[+Red90]

Am I missing something?? You results look WAY off. IME with benchmark checks, the unit is alway less than the reported accuracy. In the open results of <1.5 meters are normal.

 

I believe the unit stores in DD.DDDDD format which is more precise than DD MM.MMM Normally in the open you should be within 0.001 minutes in both directions.

 

Are you on the same datum??

[+PFF]

Hi, Red90, and others:

 

This exercise was sparked by a post from the owner of several cache's. He said he had received complaints that his coordinates were not accurate. My theory was that an averaging GPS unit would help.

 

Let's take the case of benchmark EZ0645. On GEOCACHING.COM, benchmark coordinates are in the same format as a cache would have; i.e., Degrees with decimal minutes.

 

The benchmark is

N35 50.753

W78 50.210

 

The Garmin 60cs reading was

N35.84589

W78.83683

 

This converts to:

N35 50.7534

W78 50.2098

 

Which rounds to

N35 50.753

W78 50.210.......or precisely the coordinates given on the data sheet from GEOCACHING.COM.

 

Therefore, it appears that if the owner of a cache used a GPS receiver which averaged the coordinates over time, the cache position would be accurately represented--within the limitations of a consumer-grade GPS unit. Escout and PDOP both mentioned the 3-foot/1-meter resolution, and they are correct. As has been said many times, the GPS receiver will get you close to a cache or a benchmark. Then you must rely on your eyes and experience to cover those last few feet.

 

Having submitted nearly 1,000 recovery reports to the National Geodetic Survey, I am well aware of the difference between SCALED and ADJUSTED coordinates. Nevertheless, I am glad that a couple of responders mentioned it, since it is good to get this information into general circulation. Many participants enjoy blending in a few benchmarks during searches for caches. It is common to see a benchmark or two in close proximity to a cache, and "cross-over" activity can be fun.

 

One responder mentioned that it might be difficult to find a benchmark under a tree cover, for testing. In my area (North Carolina), it is very common to wade into the woods to recover a mark. I estimate that 40 percent of the benchmarks I locate are submitted as "not suitable for GPS observations" because of a poor view of the sky. In such situations, an amplified antenna, coupled with a long averaging time, may prove helpful.

 

The 3-foot limitation of our equipment is a good segue into tosborn's question about how the tripod is positioned exactly over the target. I wish I could describe an elaborate system of plumb lines and levels, so you'd be impressed. The truth is that this tripod is simply a stand from one of those 500W quartz shop lights that they sell at Home Depot, etc. The antenna platform is made from two flat "L" brackets.

 

I combined two tripods so that the height is about 8 feet when fully extended. It telescopes down to about 3.5 feet for transporting. I position the center over the target, as much as possible. [see photo, below.] On uneven ground, the pole might lean, slightly. I try to position it so the antenna is reasonably close to being directly above the benchmark. Given the 1-meter resolution of a consumer-grade GPS, eyeballing the alignment should be adequate.

 

Best regards,

-Paul-

 

[+Red90]

Sorry, brain fart. I see you had posted in DD MM SS.SSS, brain saw something else.

 

Just forget what I said

 

Anyway, I still "think" the units only record to DD.DDDDD so that is the limit of their precision.

[ArtMan]

How long to average?

 

I just got a etrex Venture cx, my first GPSr with the ability to average waypoints.

 

I've been playing with this feature a bit, and sometimes as I watch the observation count increment, the estimated accurace improves. But sometimes the radius of uncertainty ticks upwards, rather than down.

 

Is there a general sense of how long one should average readings to add to the report on a scaled benchmark? Is two minutes enough? Five? Fifteen? (My Garmin seems to tick off one measurement a second.) If a larger number of readings results in considerably greater accuracy, it must be weighed against the time burden of a long observation period.

 

Does anyone have some hard data to support what they're doing?

 

-ArtMan-

[+PFF]

Hi, ArtMan:

 

Congrat's on the new unit, and welcome to the group of "average" folks.

 

I've been letting my unit run for three to five minutes--generally while I am doing other things like dusting the benchmark with cornstarch, measuring distances, etc. This seems to be about 300 readings.

 

Like you, I've noticed the "uncertainty radius" get smaller, then get larger. Oddly, the elevation starts out pretty close to the published data, and then drops. (But altitude is not the strong suit of a consumer-grade GPS receiver.)

 

My theory is that if you start with a strong constellation, keep the time short. If it is poor, let it run longer.

 

And, if you left the vehicle's engine running......well, use your own descretion.

 

Perhaps some of the "pro's" will weigh in, although I don't think we'll ever leave the unit unattended and let it "cook" for as long as they do.

 

-Paul-

[ArtMan]

Thanks, Paul,

 

I have a program called SAWatch, which takes output from a GPSr and plots it so you can see how each individual observation is scattered in an easy-to-visualize graphical display.

Unfortunately, as far as I can tell, it only supports serial output, not USB; and it requires NMEA data sentences as its input, which the etrex Venture cx (and Vista and Legend cx models) don't offer.

 

Striving for average,

-ArtMan-

[andman]

The 60 series has both USB and serial connections.

 

Here is the adapter to use serial: http://shop.garmin.com/accessory.jsp?sku=010%2D10141%2D00

[+Red90]

 

Unfortunately, as far as I can tell, it only supports serial output, not USB; and it requires NMEA data sentences as its input, which the etrex Venture cx (and Vista and Legend cx models) don't offer.

 

Striving for average,

-ArtMan-

 

You can get NMEA through USB with Garmin's free "Spanner" program.

Edited October 16, 2006 by Red90

[ArtMan]

You can get NMEA through USB with Garmin's free "Spanner" program.

 

Red90,

The dowload page I found says "Spanner works only with GPS 18 software version 2.90 or later." Do you know that Spanner works for current etrex cx series units?

 

-ArtMan-

[+hogrod]

You can get NMEA through USB with Garmin's free "Spanner" program.

 

Red90,

The dowload page I found says "Spanner works only with GPS 18 software version 2.90 or later." Do you know that Spanner works for current etrex cx series units?

 

-ArtMan-

 

spanner works for all garmin USB GPSr as far as I know. I have used it with my legendC before.