I have made three specific offers to use some of my older Promark Xcm units as "bases" and have been flatly turned down 3 times. Twice wit the argument that they(whoever they are) would rather wait a year or 2 and do it with L1/L2 units, go Blue Book and the whole nine yards.

I've explained to them that the specs for a station aren't gonna change between now and then and that L1 would help a lot of surveyors until they decide to go L1/L2, but no one seems to want to hear it.

Right now I have my own BASE and I use it extensively with extrememly good results . . . even to 20 plus miles . . . consistently.

What's wrong with the idea of a local L1 base, especially when there's no available L1/L2 CORS station within 33 miles or more?

I mean . . . why in the hell is it better to wait 2 years(and maybe longer) to set up an L1/L2, than to go ahead dril a 10 feet deep hole, put a CORS specified monument in and use it for L1 until the L1/L2(or maybe L5) gets put in at a much later date?

I would take a guess that "they" don't want the liability and/or the responsibility of a L1 only CORS station.



I'm very interested in how you get extremely good results.... even to 20 plus miles.... consistently and what techniques you use to do that with a L1 unit.

John



Modified By John Roberson on 9/11/2004 at 11:42 PM

Wel, the "consistent"(actually about 7 times out of 8) accuracy of 0.05 or so feet is not an absolute FACT.

I have a county, concreted GPS point(not blue-booked or anything) about 1 mile from my site.

I ran multiple sessions to a 7' long 1" pipe(left out of the ground about 6") and averaged probably 5 different sessions(about 0.01-0.015 difference, so I just did a simpe average).

When I set up "Stosh"(my 16.23 feet higher GPS station) I imbedded a disk at the 4-1/2' level, directly below the "ridge" of my new GPS antenna and measured up from the disk.

I then re-ran a few sessions to the 1 mile GPS. I always run 1 second sessions, unless I expect a long work day, then I run 2 second data)

BEFORE I got a chance to check the "local" CORS(17 miles), it went down and has been down for the past 4 or more months. So, kinda on a lark, I decided to use the CORS which is aout 33 miles south of me. I have ran 9 sessions, of 7 of those sessions, I've gotten AS to report accuracies of about 0.12 or so, the accuracy(based on the local county GPS), was about 0.05+/_for northing and easting. Sometimes the vertical isactually closer, but usually off about 1-1/2 times more.

1 session gave me results of about 0.23 or so(A.S.), while my comparative results was about 0.09. 1 session was WAY OUT. A.S. reported an error of about 0.4', while the comparative error was about 0.12'(keep in mind this is 33 miles and on these sessions I was using 5 seconds from the CORS).

Now, when I use the next CORS(to the west and 45 or s miles away), the data is consistently about 0.25-0.37 or so in disagreement. This is what led to my question abut N-S and E-W statistics.

What's interesting, is that I would get multipath and slips(from a nearby tree AND a 10 degree mask . . . I don't wanna hear it guys) . . . anyway, lately the multipaths and slips seem to have pretty much disappeared.

Now, also keep in mind I have only once ran 4 hours(CORS), but usually run 2-3 hours(1 second), on the 5 seconds session I ran 4 hours)

As far as why the data "appears" to be good, I don't know. I don't even know if I can rey on the GPS point 1 mile away for comparative purposes, but all-in-all, the data does look good.

One of these days, I will actually run multiple sessions(after the GUSTavus) site gets back up and let A.S. to the stats.

You just answered your own question....

It appears that what you’ve reported is the relative error in your measurements. Even from your own assertions, you’re getting a wide spread of values, “…1 session was WAY OUT…” What is critical is repeatability of the coordinates. The relative error is a measure of how well the measurements for that session fit with themselves. If you run 7 sessions, what is the spread of the coordinate values for your data? What are the statistics for the multiple sessions?

One obvious answer to your question as to why some sessions are better than others is the changes in atmospheric conditions from place to place. There are times when the conditions will be uniform over a fairly wide area. In those times, using data from sites far away will work. Unfortunately, with L1 only data, you simply do not have the capability to model the atmospheric conditions. You will never know if the conditions over a CORS site 30 miles away are consistent with your local conditions. There is nothing you can do to make that determination because the L1 only data doesn’t give you the ability to model the conditions.

It helps to look at an easy to understand example. If you put a stick in a pool, the stick appears to bend at the waterline. This is due to the fact that the refractive properties of water and air are different. The same holds true for our atmosphere. As the radio signals from the GPS satellites enter our atmosphere, they are refracted. The amount of the refraction will change considerably with changing atmospheric conditions. The signals on L1 and L2 are in different frequency bands. The refraction of each is different. There are a number of solid mathematical models that can interpret the refractive properties of the atmosphere with each measurement as long as the two frequencies are available. That’s the simple definition of the atmospheric modeling available when you have L1 AND L2 signals in use. With L1 only signals, no modeling is possible.

In most cases, moving 10 miles away from your present location will get you to a place where the conditions are different enough to see the effects of the “refraction” of the radio signals from the satellites. When you’ve only got L1 data, you cannot account for the differences. When you have L1/L2 data, you can account for the differences at both ends of the vector.

For a simple experiment, take your L1 data and process it against the data from a CORS site a few hundred miles away. Try the same thing with data from a thousand miles away. Occasionally, you’ll get interesting result with resolutions in the centimeter range. However, that does not mean that you can consistently process data against the CORS site MAUI and useable results.

This is not an indictment of you or your methods. It is a simple statement of fact based on the physics and mechanics of the system and our natural world. L1 only data is not as uniformly usable as L1/L2 data. That’s why no one in your area is interested in and L1 only base station. The results are not reproducible. The data will not be consistent. The reproducibility of the coordinate values is simply not there.

The move from traditional optical survey measurement methods to GPS survey measurement methods is far greater than the move from steel tape survey measurements to EDM survey measurements. The EDM spit out numbers that could be reproduced with amazing accuracy. However, there were surveyors who did not understand that the measurements they were making were slope distances from the EDM to the prism. They would use vertical angles from the scope to the point on the ground. As long as they were measuring very long lines that weren’t too far apart vertically, they were OK. When they were measuring shorter lines to points that were very different in elevation, the results would go “haywire”. This was due to the lack of understanding of what and how they were measuring.

GPS is significan

Ian...... You Da Man......





-----------------------------------

John F,

Basically Ian is saying it doesn't work because it will "wobble"



Modified By John Roberson on 9/12/2004 at 3:19 PM

Well, yeah, to a degree.

What I’m kind of saying is that if you have ten people who say that 2+2=5, you have a high degree of correlation but your accuracy still sucks.

If you use L1 only data 50 miles apart, you may well get tight stats on each point, but the data for the vector between the points will be pretty poor.

Try a series of sessions, say 1 or two hours each. Process the data against a set of CORS sites 50 miles away to 1000 miles away. Look at the lengths of the vectors from your experimental site to the CORS site. They’re going to be all over the place.

Now, get a series of L1/L2 data at the same place. Process the data against the same CORS sites. You should see a much higher degree of correlation between the L1/L2 vectors.

That’s why no one is interested in an L1 only base station. The resulting VECTORS are not sufficient for survey grade measurements.

End of story.

Hi Ian,

I tried e-mailing you a week or so ago about the problems you are having getting FieldGenius to connect to your Hiper Lites. I don't know if you got the message because I did not hear back from you. Could you e-mail me at "jroberson@comcast.net" and I will give you the details as to what I did to resolve the same problem that I had with my Sokkia RTK gear and FieldGenius?

Thanks

John Roberson

Well . . . okay.

I ALMOST ALWAYS use 1 second epochs, rarely 2 and one time EVER, I used 5 second epochs. That one time was the time that I had the session that was "WAY OUT". But that was the item of another post I made questioning the advantage of 1 second epochs(to which everyone says is a "waste" of time).

SO BASICALLY . . . IGNORE THE ONE THAT WAS "WAY OUT", because that one was not done in compliance to my Modus Operandi.

That means that, in actuality, my dta was pretty dogone tight for a 33 mile baseline.

I think the gist of my posting and I also think the gist of your postings must be in agreement.

L1 suffers from atmospheric conditions, more and faster . . . okay. What are we really talking about here?

That at 33 miles, my L1 suffers from problems(at 1 second), that in about 5 weeks of sessions has resulted in a difference of as much as 0.12', although almost always within about 0.05'? That an L1/L2 system would've given me results(at 1, 2, 5, 10 or 15 seconds) that would've been "tighter" and therefore, more reliable? Do you think I don't already believe that?

ANYWAY . . . we're not talking 33 miles, we're generally talking about 10-15 miles(the distance from my "base" to my job sites). IN FACT . . . even for the L1/L2 users, I seriously doubt we're talking more than 30-50 miles IF there's a CORS station available.

Maybe what we "are" talking about is that L1 GPS has become a bastard-child of the surveying community. The kind of GPS that only the very cheapest and the least progressive of surveyors continue to use. How differently would this discussion progress if we were back in time, before L1/L2 and some enterprizing person wanted to set up an L1 base? I venture to say the discussions would be a million percent different . . . why?

MIND you . . . mind EVERYONE, I am in no way saying that L1 GPS compares to the utilitarian use of L1/L2 and that L1 units will give the same results in a wide-ranging set of circumstances.

I am simply saying that even the "cheapest" of us surveyors can make good use of a local private L1 GPS base and that a "local" L1 GPS base is much better, for us L1 users, than a more-distant L1/L2 CORS, of which us "CHEAP" L1 GPS users can't take advantage of ANYAY.

As a "CHEAP" L1 user, I wouldn't even have this discussion if there were more of the L1/L2 CORS stations. Fact is, for L1 users, there aren't enough CORS stations around and many of them that are around seem to constantly suffer from hiccups".

I do wonder if the bulk of the posters that tend to be of the "naysayer" type are L1/L2 and happen to live in "that" world with no consideration to the L1 users? Is this a form of "class" warfare?

I'l tell you this, an L1 user is going to feel just as bad and get just as mad, when returning to his/her base, the units has been lifted, even though they are cheaper. AND, I venture to say that the time LOST because of the theft is equally disgusting.

I would just like to know why some people are so absolutely ANTI- L1 basestations?

I would just like to know why some people are so absolutely ANTI- L1 basestations?

Because they do not supply enough data to make survey grade measurements possible. End of Statement!

No, John, we are NOT in agreement!

The L1 only data you collect are wonderful for your project. They are simply not up to collecting data to establish relationships to projects a long way away. PERIOD.

It's not an indictment of your methods or procedures. It’s a statement of fact.

No, I do not have it in for L1 users or equipment. I carry three L1 only ProMark2 units in my truck every day. They add enormous capability to my L1/L2 HiPer Lites. When the HiPer Lite base is set up and running in conjunction with my three PorMark2 units, I get FOUR vectors from known points to each of my rover static sessions. The difference is that I do not try to process the L1 only data against L1/L2 CORS site data. I use my L1/L2 base for that purpose and then process the L1/L2 rover data and the ProMark2 data against the adjusted base data.

Do yourself a big favor and take some fundamental classes in GPS theory and practice.

I would just like to know why some people are so absolutely ANTI- L1 basestations?

Because they do not supply enough data to make survey grade measurements possible. End of Statement!

No, John, we are NOT in agreement!

The L1 only data you collect are wonderful for your project. They are simply not up to collecting data to establish relationships to projects a long way away. PERIOD.

It's not an indictment of your methods or procedures. It’s a statement of fact.

No, I do not have it in for L1 users or equipment. I carry three L1 only ProMark2 units in my truck every day. They add enormous capability to my L1/L2 HiPer Lites. When the HiPer Lite base is set up and running in conjunction with my three PorMark2 units, I get FOUR vectors from known points to each of my rover static sessions. The difference is that I do not try to process the L1 only data against L1/L2 CORS site data. I use my L1/L2 base for that purpose and then process the L1/L2 rover data and the ProMark2 data against the adjusted base data.

Do yourself a big favor and take some fundamental classes in GPS theory and practice.

Modified By D. Ian Wilson on 9/12/2004 at 4:29 PM

Just read your 3:57 posting.

WHAT A CROCK.

You're talking 50 miles . . . I DON'T WANT TO TALK 50 MILES. The whole bulk of my posting is because I don't want to talk 50 miles

What the heck does that last sentence even mean? Are you saying L1 vectors ARE NEVER SUFFICIENT for survey grade measurements?

Well, it's a good thing you don't speak for Topcon, Leica, Leitz, Ashtech and whathaveyou, because they all beleive that L1 units ARE SUFFICIENT for survey grade vectors.

ARE THEY ALL WRONG?

Should I return my L1 GPS, or maybe I should just TRASH them if they're no good for surveying.

NOW . . . if you just saying they're no good for 50 mile and 1000 mile vectors, then clarify the statement and understand the gist of my postings.

I am not trying to REPLACE L1/L2 bases. I am not trying to say L1 is as good as L1/L2 over a greater distance.

I am ONLY saying that L1 bases would be worthwile for one whole hell of a lot of surveyors and I can't help but believe that there's a whole lot more L1 users than there are L1/L2 users.

I am only saying that MORE local L1 bases would be MORE useful for L1 GPS users that LESS and more distant L1/L2 CORS.

I am only ASKING why an L1 base is't better than having NO LOCAL GPS station.

Obviously, the only answer I can come up with, is that the "real" users of GPS(that is the L1/L2 & RTK), can't understand the utilitarian purpose of local L1 bases for L1 GPS users.

What I don't understand is what concern is it for L1/L2 users. You guys ALWAYS claim you gan get good data from 1000 miles. So . . . maybe we can shut down ALL of the L1/L2 CORS stations and only leave one up in each state(maybe only one per every 3 states), then, with all that money saved, we can place L1 "bases", every 10 miles or so all across America?

Nah . . . you L1/L2 users only need a few bases in all of America. We'll leave one in NewYork, one in Chicago, one in Las Vegas, one in San Francisco and one in Florida . . . yeh . . . that'll do ya. Then we can place L1 bases every 5 miles or so.

I won't reply out of fear of getting kicked off the board site.

No, John, I have not said that L1 only recievers are not capable of survey grade measurements. In fact, if you actually understand what i wrote, you'de see that i use the PM2 in conjunction with my HiPer Lites EVERYDAY!

John, I suspect that the reason none of those you've approached about establishing an L1 CORS has less to do with the limitations of L1-only than it does with the overall job of building a reliable CORS.

The cost of the receiver and antenna is only one part of the budget. You'd still have to provide a secure site, a power supply, data management software and an Internet connection. Even with a donated site, free electricity, publicly-available software and access to an Internet connection, someone still has to set it up, test everything, positioning the antenna, and keep it all running. Taken as a whole it's not a trivial project, so it's easy to understand why you're having trouble generating interest. I'd venture to say that even if you were to come up with a dual-frequency receiver and antenna, you'd be facing much the same resistance.

P.S. Ian gave a detailed, comprehensible and and pretty thorough explanation of the range limitations inherent in L1-only vector processing. I'd add only that L1-only processing software *does* model the ionosphere, but uses a canned model that assumes similar conditions at both ends of the baseline. As he noted, the problem arises when this assumption isn't correct.

READ your first sentence after your included question from me at 4:24 and 4:26.

Tell me you didn't say that L1(read ProMark2) GPS ". . .do not supply enough data to make survey grade measurements possible. End of Statement!". Can what you said be any plainer?

You stated it three times, all free-standing and twice as lead-in statements. If you don't mean it . . . don't state it.

John Francis' question: I would just like to know why some people are so absolutely ANTI- L1 basestations?

My response: Because they do not supply enough data to make survey grade measurements possible. End of Statement!

So where does that imply that L1 units do not provide survey grade information? Are you confusing a base station with a unit?

Again, check out page 15 of this month's POB. You'll find a picture of me setting up one of my ProMakr2s to collect survey grade information. BTW that's not a base, just a unit over a found monumnet.

For a "learned" person, your last post is out of line.

You might be right that your statement does not "imply" that L1 units are not capable of survey-grade data . . . IT FLATLY STATES IT, openly and specifically.

Any person with a remedial understanding of the King's English should be able to read that statement and understand what it's saying.

I think I understand what you mean, but that is NOT what you said and that is NOT what you continue to say.

As far as "what" a base is. I'm am totally unaware that a "base", by virtue of "definition" even exists with specifications that we're discussing.

I implore you to divy me up this "definition" of a base that emphatically states that a "base" could not be ANYTHING.

I understand the term "base" in this instance to mean a GPS unit that is solid, not moveable and including a way of gathering GPS data. Tell me why an L1 unit, attached to a permanent-type monument, topped of with a GPS antenna and gathering data, can't be considered a "base". Is the use of the word "base" in this respect illegal or something?

Maybe I got this screwed up with a "base" camp or the "base" of a flagpole.

This is like a "Vector Paternity Tests REVEALED" episode of The Jerry Springer Show! Don't they film that in Ohio these days?

JERRY! JERRY! JERRY!

I may be out of line, but this thread has evolved into the old saying that "I can teach a dumb person, but I can't do anything with a stupid person"



Modified By John Roberson on 9/13/2004 at 3:15 AM

I would say no one is out of line if someone is trying to gain some sort of knowledge by asking questions or replying to questions based on the actual question asked.

That makes your 11:37 post of yesterday the last intelligent thing you said regarding this thread and it makes your post of 3:13 post this morning an example of the pinnacle of ignorant babble.

This thread would not have delved into the realm of the ridiculous if some posters had the mental capacity of understanding and reading a question for what it is and addressing that question as it was presented.

But no, some posters had to take the thread off-tangent into issues that were not being presented or contemplated and some posters, for some reason, felt compelled to make profoundly STUPID statements, which in and of itself, were absolutely not true, supposedly leaving the reader to ASSUME some unassociated facts to give the statement veracity.

Had you . . . guys had taken the time to read AND comprehend my posts . . .

You would've known that my "base" coordinates were derived from a county GPS monument about 1 mile away.

You would've known that the use of the CORS(at 33 miles) was initially used, "on a kinda lark" and that the results to/from the CORS is a sideline issue of interest.

You would've known that my spreads from the CORS(at 33 miles) were relatively tight to data from the LOCAL county monument

You would've known that the one "bad" session occured when I changed my epochs to 5 seconds(in fact my "base" might have been at 2 seconds, making the common epochs at 10 seconds)

You would've known that I only intended to use the L1 base to a distance of up to 12 miles with consideration for slightly longer distances in those "special" situations.

You would've known that I . . . NEVER . . . implied or stated that an L1 "base" would surplant an L1/L2 "CORS" BASE. But, rather, a monument made similar to a CORS spec'd monument could be used for L1 UNTIL the hardware for an L1/L2 unit could be put in place AND THAT A BASE CAN BE L1 AS WELL AS L1/L2.


In short, since it seemed certain posters went off on their own little tangental direction, I'd have to say, regarding this thread, they didn't know JACK.
Modified By John Francis on 9/13/2004 at 9:00 AM

John F,

You need serious anger management.

Wow . . . and I thought I lived a sheltered life.