Antenna height problem?

Are you looking at only single occupations at each station? You need at least two if you're going for height, in my experience.

How long are your sessions? Height results often improve noticeably up to 4 hours, less so after that.

Are you using the same CORS in each case? If you use different CORS sets, you can expect different height results, at least around here.

I doubt the problem (if you call 1 cm in height a problem) is due to your measure-up.

3.2 hrs, 5.1 hrs, 5.1 hrs, and 4.5 hrs, with whatever CORS are selected by OPUS-S so not always the same ones. And of course I have no way to know what CORS the published data came from. The GPS On Benchmark program calls for a minimum of 4 hours but I've seen no expectation for multiple occupations.

They are all plenty close enough for comfort, except that all the height differences are in the same direction and that seems highly suspicious.

Surveyors that I have worked with point me to the antenna list that OPUS provides and verify that antenna matches with mine. Others found that name to be imprecise. That may explain a bias.

Sent from my iPhone using Tapatalk

Do you measure to the top of the ground plane or to the bottom? If your ellipsoid heights aren't matching published values I'd look very carefully at antenna heights and make sure something isn't amiss. There isn't a whole lot that could go wrong with OPUS, your rod height is really the only variable input. Though 1cm with GPS only receiver seems pretty good to me without a pair off observations without at least 2.5 hours of start time seperation.

Can you swap the 2 recievers?
It's what I think I'd do...
Another trick, is to set them up with a carefully leveled beam, about 2 ft apart, where you know they are at the same elev.
N

Oh, i see. It's one receiver off opus, to a point of known elev...
N

it can never hurt to double check the chart/table you made...murphy's law and all..... here's a quick spreadsheet clip for that ant. SL measured to bottom of groundplane notch.....

If my goal is a good elevation I use OPUS-RS.

Next, use the measuring rod to find slant height. Separately measure your height to the ARP, which is usually easier to do without the antenna mounted.

Paul in PA

GISJoel_GetItSurveyed, post: 412033, member: 11867 wrote: the antenna list that OPUS provides and verify that antenna matches with mine

I'm always specifying the exact model as read off the nameplate of the antenna.

Paul in PA, post: 412053, member: 236 wrote: If my goal is a good elevation I use OPUS-RS.

OPUS-RS is not an option for OPUS Shared solutions, as needed for GPS On Benchmarks program.

Rankin_File, post: 412041, member: 101 wrote: here's a quick spreadsheet clip for that ant. SL measured to bottom of groundplane notch

Very similar to my spreadsheet that I printed my table from

geopro_consultants, post: 412034, member: 9959 wrote: Do you measure to the top of the ground plane or to the bottom?

... except that I find it easier to measure to the top of the ground plane and use dimension B instead of C. The Trimble manual didn't seem to say which they considered standard practice so that's the way I got started doing it.

Paul in PA, post: 412053, member: 236 wrote: Next, use the measuring rod to find slant height. Separately measure your height to the ARP, which is usually easier to do without the antenna mounted.

As noted in the original post, I've checked my slant measurement and calculation against the height of the tribrach adapter the antenna sits on.

"OPUS-RS is not an option for OPUS Shared solutions, as needed for GPS On Benchmarks program."
Yes. But you are looking for a problem; not submitting results.
Take 1.5 hours of data from each observation and submit them to OPUS-RS with the same measured antenna heights. Then compare the results. If you get similar variations from published data; it probably is an antenna height issue.
Hopefully, the results will vary and you can be sure of your procedures. You may have simply reached the limits of your equipment or OPUS.

Its been awhile since I used the Trimble 4000, this was designed back when you really had to know what you were doing, if I remember correctly they had an option of 4 places where you could measure the height to the antenna. Also I thought you just measure to the antenna and you say if you had the ground plane on the antenna or not, and it computed the slope distance based off ground plane or no ground plane.

Today's GPS is so much easier, grab a fixed 2 meter pole enter 6.56 for rod height and start working.

I would double check you are measuring to the correct mark on the antenna, and try it without calculating the slope distance.

Scott Ellis, post: 412101, member: 7154 wrote: I thought you just measure to the antenna and you say if you had the ground plane on the antenna or not, and it computed the slope distance based off ground plane or no ground plane.

"It" being the Trimble software, I have read that is true. OPUS asks you for the height of the ARP over the mark so you have to calculate it yourself.

As I said, I have checked (more than once) that my calculation matches the top of the tribrach adapter the antenna sets on and the measuring rod matches another scale.

I have a theory: I'm using too long an antenna cable and the cable's differential delay between L1 and L2 is adding error.

I think it is usually assumed that the antenna cable has equal delay for all signals. There certainly is no place I've seen to specify any cable attributes. Gross delay makes the receiver track time a hair differently but has no effect on position calculations, so this seems like a reasonable assumption.

But cables do have increased delay as the signal frequency increases and approaches the range of greater attenuation. One of my friends is a radio circuit design engineer. Without having the specifics of my cable beyond length, he made a SWAG that the difference in delay for L1 and L2 would be on the rough order of 2 cm of free space travel time. That indicates to me that cable dispersion makes sense as a candidate mechanism to be considered.

Thus, if my cable has differential delay that doesn't match the cable NGS used to measure the L1 and L2 phase centers of this model of antenna, I should expect some additional error. The widespread use of GNSS receivers built into the antenna assembly has made this issue irrelevant for newer equipment.

I've been using the 10.6 meter (35 ft) antenna cable that came with the ST antenna. I need to switch to the 4.25 meter (14 ft) one that came with the 4000sse receiver. If cable delay is significant, that should reduce the effect.

you may be comparing apples to oranges. Your answer might be just fine for OPUS but not for a NAVD88 elevation, if that's what you are trying to do.

Also ellipsoid heights will vary depending on the EPOCH.

They may both be good but different.

Remember CORS is a moving target, each Epoch change and GEOID MODEL update changes the locations of any monument using CORS/OPUS. And it shows up in the ellipsoid height and GEOID height more than horizontal changes, at least in my area.

MightyMoe, post: 412114, member: 700 wrote: Your answer might be just fine for OPUS but not for a NAVD88 elevation

Since the OPUS report and the published data sheets both use GEOID12B to relate ellip and NAVD88, it doesn't matter which kind I compare; I'll get the same difference.

Bill93, post: 412019, member: 87 wrote: marks that have published GPS data

So I'm never comparing leveling data versus GPS.

MightyMoe, post: 412114, member: 700 wrote: Also ellipsoid heights will vary depending on the EPOCH.

I wasn't aware that the North American plate had a significant vertical motion (over a few years) relative to the ellipsoid. Can you direct me to more information? Might you be referring to holding leveling values and computing back to ellipsoidal height via the geoid model?

Bill93, post: 412137, member: 87 wrote: Since the OPUS report and the published data sheets both use GEOID12B to relate ellip and NAVD88, it doesn't matter which kind I compare; I'll get the same difference.

So I'm never comparing leveling data versus GPS.

I wasn't aware that the North American plate had a significant vertical motion (over a few years) relative to the ellipsoid. Can you direct me to more information? Might you be referring to holding leveling values and computing back to ellipsoidal height via the geoid model?

I'm looking at a HARN datasheet and there are four different ellipsoid heights ranging from 1173.304 in 1994 to 1173.185 for the NAD83(2011) number. The other ellipsoid heights shown are 1172.26 in 2001 and 1173.219 in 2007, these are in meters. This is a first order bench mark and a HARN point.

Interesting. PID?

PW0138, I suspect you aren't doing anything wrong.

If you look at a CORS site it should have superseded control also, assuming it's been in existence long enough, the ones I use in this area are new so.....