Typical <expletive deleted> AI answer - nearly, but not quite right
Designed for conventional instrument work, by default Access works in an rectangular coordinate system with scale factor =1.000000000
You can change the scale factor, and can you allow or not for atmospheric corrections to measured distances
For almost all GNSS work you need to select a Coordinate system, and for that your AI answer is actually correct.
And even then you can select whether you see ground or grid distances and coordinates
Remember, this is ChatGPT's example, not mine. The "yes" in the copy and paste is my asking ChatGPT to show me its example. Caution: Communicating with AI will cause you to drink too much.
Here's a bit more AI speak based on Jim's comment:
ChatGPT said:
Excellent question — and now we’re really getting into the core of how Trimble Access and survey software work. The answer is:
✅ Yes — you can measure in grid rather than ground, but only indirectly.
Trimble Access always measures ground-based data from your equipment (GNSS or total station). That’s physical reality — you’re standing on Earth.
But you can configure Access to treat your work as if it’s in grid by letting it automatically adjust all your ground observations into grid distances and coordinates, using the correct Combined Scale Factor (CSF).
Now let’s break it down:
🧭 If You Want to “Measure in Grid” — Here's What That Means:
It means you want:
-
All coordinates and distances to be in the State Plane projection (i.e., grid system)
-
All observations adjusted to match what a “flat grid” would say
-
Any data exported to CAD/GIS to match control points precisely
🔄 What Happens In Access, Depending on Scale Factor Settings:
| CSF Setting | Raw Measurement Type | Access Applies | Resulting Output | Behavior |
|---|---|---|---|---|
| Default (computed CSF) | Ground | Converts ground → grid | ✅ Grid distance & coords | This is how you “measure in grid” |
| CSF = 1.000000 entered | Ground | ❌ No correction | ❌ Output is ground, but labeled grid | Fakes grid output — wrong for real SPCS |
| No CSF (rare cases) | Ground | Might compute one from control or none | Usually grid, unless customized | Depends on setup — verify settings |
📐 Scenario Example:
Let’s say you're staking out a boundary in State Plane Coordinates (SPCS). Your plan set is in grid:
-
You want your layout to match CAD exactly
-
You want distances to match project control points, which are on SPCS
-
You want to stake 500.00 ft grid line exactly
✔️ Correct Approach:
-
Let Access compute and apply the CSF
-
Input points and lines in SPCS
-
It will translate grid → ground for stakeout
-
And ground → grid for stored data
Access makes the conversion both ways as needed — but the output remains accurate in SPCS.
❌ Incorrect Approach:
-
You manually enter CSF = 1.0
-
Access assumes your ground measurements are already grid, and doesn’t correct them
-
Your coordinates and distances are now longer than true grid values by ~1 part in 1,000
-
You get a mismatch against real-world SPCS data
🔚 Conclusion:
-
You can't directly “measure in grid” because your hardware always measures ground.
-
But you can absolutely work in grid coordinates by letting Access compute and apply the correct CSF.
-
Entering 1.0 disables the conversion, and gives you unadjusted ground values — still labeled as State Plane, but incorrect.
Would you like a field workflow checklist for ensuring you're working in true grid mode?
You said:
ChatGPT said:
Here’s a Field Workflow Checklist for ensuring you’re truly working in grid (State Plane) mode using Trimble Access — meaning all your coordinates and distances match CAD, GIS, and control data that are in SPCS.
✅ Trimble Access Field Workflow for Working in Grid Coordinates
🔧 1. Set Up Your Project Coordinate System Correctly
-
Select the appropriate State Plane zone (NAD83(2011), NAD27, etc.)
-
Confirm the correct datum, zone, and units
-
Make sure geoid model is loaded (if using GNSS)
🧮 2. Use a Site Calibration or Coordinate System File
-
If using GNSS, perform a site calibration or load a control-based coordinate system file.
-
Ensure Access understands how to convert between ellipsoidal (GNSS) and SPCS grid.
📐 3. Let Trimble Access Compute and Apply the Combined Scale Factor (CSF)
-
DO NOT manually enter CSF = 1.0
-
Either:
-
Let Access compute a per-point CSF from the projection + elevation
-
Or let Access compute a project average CSF
-
-
You can check this under Job Properties → Coordinate System → Scale Factor
📏 4. Confirm That Distances and Coordinates Are in Grid
-
Use a known State Plane distance between control points (e.g., a 1,000.00 ft line)
-
Measure it
-
Access should report 999.7–999.8 ft on the ground, if elevation is typical
-
If it still says 1,000.0 ft, you’re getting uncorrected ground distances
📌 5. Use “Stakeout” Features That Convert Grid to Ground
-
When staking from CAD or control files, Access will apply the inverse CSF
-
It tells you: “Move 999.76 ft” on the ground to lay out a 1,000.00 ft grid line
-
This ensures your stakes match the design on paper
🧾 6. Export Your Data Confidently
-
Coordinates and distances stored are in SPCS grid
-
They will match:
-
CAD designs
-
Legal plats
-
Control network files
-
GIS base maps
-
❌ Avoid This Mistake
Entering CSF = 1.0 in Job Settings
That will:
-
Disable CSF correction
-
Store ground distances and call them “grid”
-
Introduce error (~1 ft per 5,000 ft at 3,000 ft elevation)
✅ Quick Test:
Ask yourself:
-
"Does the distance between two known State Plane points match what's in Access?"
-
If yes → You're in grid
-
If no (too long) → You're in ground; fix your scale factor setting
Would you like a printable PDF version of this checklist?
What a cluster
I need to reconsider my answer to your original post. You wrote:
I know for sure that my GPS measures grid when the CF is set to 1.000 because I have tested it using 2 county NAD83 published mons that inverse 116303.05' apart and the CF=0.9999746 at the lower point. My GPS grid coords inversed 116302.93' which is pretty hot.
The problem is that 116302.93/116303.05 = 0.99999897, not 0.9999746.
What's up here?
@mightymoe "I do know about my old programs from the 1980's and Trimble from the 1990s and forward, FieldGenius isn't one I've dealt with, if it's applying 1 on top of the projection it must be doing it to try and stop users from doing something wrong."
Well I think it is more of a way to tell the user that there are no hidden or underlying conversions to the data being measured. When I select to use a GPS then everything will be measured in grid unless I specify otherwise. When I select to use a TS then everything will be measured in ground unless I specify otherwise.
I have never used Trimble software but are you saying that that software is automatically converting all GPS measurements to ground so you can use both GPS and TS on the same job file without having to set the scale factor in either instrument?
@mathteacher "I need to reconsider my answer to your original post. You wrote:
I know for sure that my GPS measures grid when the CF is set to 1.000 because I have tested it using 2 county NAD83 published mons that inverse 116303.05' apart and the CF=0.9999746 at the lower point. My GPS grid coords inversed 116302.93' which is pretty hot.
The problem is that 116302.93/116303.05 = 0.99999897, not 0.9999746.
What's up here?"
The CF=0.9999746 is for converting ground to grid ie: grid = CF x ground or ground = grid / CF. The inversed grid distance between the 2 published county coords is 116303.05'. If my GPS was measuring ground I should have measured 2 coordinates that inversed 116305.88'. Instead I measured 2 coords that inversed 116302.93' or 0.12' error / 116303 for 1:1000000 precision. So my GPS is measuring grid.
This is also a testament to the accuracy of the CRTN network that I have been using. To cover that large of a distance I had to use a different mount point at each county monument because each mount point only reaches out about 15km or 49200'.
@mathteacher "But the CF you gave doesn't produce the result you claim."
The results are that I matched the inversed grid distance with my GPS measured grid distance within 0.12' on a 116303' distance.
The CF is only for converting between ground and grid. When you know one distance (grid) and want to find the other (ground) or visa-versa you can use the CF.
Yep. I pushed a wrong button. Sorry about that.
So, what does Access really do?
For me; Access/TBC allows me to create an LDP or chose a State Coordinate System. My clients almost universally want ground information so I expand the coordinates with a Project Adjustment Factor in the Local Site option. From there I'm always on a surface close to ground. For small areas such as 10x10 miles we try and stay within 10PPM. I've had two projects in the last twenty years on State Coordinates, one was about 16,000' below ground and the other was about 5,000' below ground or project factors of 1.0008 and 1.000241. Each company was unhappy with the resulting data, but they insisted, then complained.
Once the coordinate system is set, the GPS and Total Station work in those projections. We don't worry about grid to ground, since we are always on ground. I haven't looked in many years if the newest Trimble programs convert the measurements done with the Total Station correctly to a State Coordinate System. I don't really care, we will never be concerned with that. 10PPM of possible error the way we work is swallowed up with bubble error at short distances. We haven't taken a 1000' Robot shot since we got them twenty some years ago. I assume if you want to work in a projection well bellow the surface of the earth; Access in the job file will adjust all measurements if it's GPS or TS to the grid set in the DC.
If not, then the user should figure it out. As far as FieldGenius I have no idea. But these are questions that are so trivial to answer. Try it, find out. As I related above, I worked on a job with a 1.0008 PAF or the reverse- a Combined Scale Factor of .9992.
Set up a grid job with those parameters, use the GPS to lay out two points 200' apart in a parking lot. Chain between, you should get a ground distance of 200.16' then use the total station with the DC attached and see what you get. Simple field test. Now you know, takes maybe 1/2 hour start to finish. Always, always, always test your equipment. We run tests of one type or another almost every day in the field. The OP should test his program like I related above, then report back what he discovered.
The results are that I matched the inversed grid distance with my GPS measured grid distance within 0.12' on a 116303' distance.
0.12' is right around the difference between grid/ground points scaled from a local point versus scaled from 0 in my area, so depending on the elevation in your area I might still be wondering if I'm measuring grid or ground.
I also haven't read through most of this thread, so I need to back up and do that.
The results are that I matched the inversed grid distance with my GPS measured grid distance within 0.12' on a 116303' distance.
0.12' is right around the difference between grid/ground points scaled from a local point versus scaled from 0 in my area, so depending on the elevation in your area I might still be wondering if I'm measuring grid or ground.
I also haven't read through most of this thread, so I need to back up and do that.
If you're only seeing .12' in 116303' then I sure wouldn't do anything but state plane.
My rheumy old eyes led me to compute a combined factor using the two ground estimates instead of using the grid estimate and a ground estimate. See three, choose two, but make sure that you choose the right two.
The ChatGPT stuff is solid, but an even better source is here:
It's more than an hour long, but the instruction is straight from Trimble. Skip the first 3 1/2 minutes if you want.
Note also the number of recurring questions and error sources that Trimble regularly addresses. That might say something about the state of training and mentoring in a lot of shops.
Moe, you're a pro who knows his equipment and how to get it to produce results that fit those job requirements. What seems to come up regularly is the latter, how to set up the equipment to get results in the required geodetic format and how to verify that.
I don't do surveying, have no commercial surveying software, but I do a lot of calculations. My LDPs are built in a 3-page spreadsheet that I wrote in 2010 and have improved over the years. I do rotate and translate in Excel Solver, along with some least squares adjustments.
I got away from geodesy during covid, spending time developing a covid model whose explanatory power was high, but predictive power low. Over the last year, it's been handgun ballistics, focusing on recoil for larger calibers like 45ACP and 357 Magnum, revolvers vs semi automatics, blowback vs recoil operated ejection and loading.
I'm rusty on the geodesy, but it's coming back. I want to know more about inputs to surveying software, and I know little, so I may look a bit foolish. I'd bet the farm on the geodetic soundness of its calculations, much less on every user's ability to get it to do those calculations reliably.
Anyway, there are many people here who I don't know, but you're an old-timer, and it's good to engage with you.
As always, I mean well.
Moe, you're a pro who knows his equipment and how to get it to produce results that fit those job requirements. What seems to come up regularly is the latter, how to set up the equipment to get results in the required geodetic format and how to verify that.
I don't do surveying, have no commercial surveying software, but I do a lot of calculations. My LDPs are built in a 3-page spreadsheet that I wrote in 2010 and have improved over the years. I do rotate and translate in Excel Solver, along with some least squares adjustments.
I got away from geodesy during covid, spending time developing a covid model whose explanatory power was high, but predictive power low. Over the last year, it's been handgun ballistics, focusing on recoil for larger calibers like 45ACP and 357 Magnum, revolvers vs semi automatics, blowback vs recoil operated ejection and loading.
I'm rusty on the geodesy, but it's coming back. I want to know more about inputs to surveying software, and I know little, so I may look a bit foolish. I'd bet the farm on the geodetic soundness of its calculations, much less on every user's ability to get it to do those calculations reliably.
Anyway, there are many people here who I don't know, but you're an old-timer, and it's good to engage with you.
As always, I mean well.
At this stage in my business, things about geodesy are background noises. As an example, I recently set some monuments that were based on a 2006 survey from this office. Went to the field to check into the controlling points that used the 20 year ago geodetic positions while occupying new control, hit the west controlling monument .04' "off" hit the east controlling monument .08' "off", however it wasn't the 2006 monument, it had been replaced by a different surveyor who didn't file on it. Still glad it was there, and no I didn't worry about any miss, nor did I do any adjustments.
But, think about this, this was older control, based on NAD83 (93), our new control is now NAD83(2011) and there is a time drift. So, we were able to recompute the positions with newer control, place the mid-point without doing new computations simply create a line and mid-point it. All this happens in the DC and we can trust it all since the points keep being where expected. This isn't anything unusual, it's repeated over and over. Time spent adjusting is very minimal. Time spent with geodesy is basically approaching 0. It's not what happened 30-40 years ago. I was pouring over calculations in those days, now Pffftttt. Time is better spent doing the real work, which is more the legal side than the math side.
@mathteacher Trimble Access has everything that is needed. If you choose state plane and zone. Key in the appropriate project height and state you will be on GRID. This uses the geoid the lat long ellipsoid height of your chosen datum etc. No need to key in a cf for total station nor gps to be on grid. As a fact you can inverse and see either the GRID distance or GROUND or along the ellipsoid distance. Now the ground distance is using the scale ellipsoid at both ends to compute the horizontal distance at ground at the mean of each end. If you want say state plane scaled to ground this to works for GPS or Total station work. You can key in the reciprocal of your site’s cf and you need the position in which you want to scale from. So again no matter which tool you are using it does the math for you. We often start all jobs as state plane grid. Locate all property corners set control with all tools available. GPS robot level tape etc. once I have it in the office software I will usually this is my preference computer a horizontal location at the center of the project. I also go through and evaluate all the ellipsoid heights around the project to get an averaged ellipsoid height on that site. I assign that height to that point in the center of my site. And then scale everything to the surface based on that. The. Check all major distances diagonal north south east to west. Across the site to see how grid and ground come together. Most of the time choosing any point would be just fine but if lots of relief it sometimes works out for me to manually choose a good height to derive the reciprocal of what we call the CF. Trimble uses the terms in access ground calculated or ground keyed in. Scale factor. It can compute this from any lay long ellipsoid height either keyed in or you can upload it to the controller. They also have this for no projection no datum. Along with another way which is by default scale 1.00000000. In that no projection is given and one could apply this to whatever I guess. You can even do it per set up. Now I always keep my advanced geodetic settings on and check reduce to sea level / ellipsoid. This enables total station work to be reprojected from the ellipsoid up to grid etc.
Exactly what I do in the math world. Before Computer Algebra Systems (CAS), I would spend as much time as it took to integrate a difficult function and compute the area under its curve. Now I just let CAS do numerical integration and go with the answer. i care little about the form of the integrated function, although even limited CAS can do indefinite integrals as well.
Nowadays, many teachers want graphing calculators replaced with powerful CAS on computers. The sky's the limit to the applied math you can teach with that.
I'm a bit of a throwback, I'm afraid. I know that I truly understand a process if I can program it. It's almost all spreadsheets now as the modern languages have outrun me.
Capital is replacing labor in virtually every field. Heck, we don't even have guys hanging off the backs of garbage trucks around here.
Exactly what I do in the math world. Before Computer Algebra Systems (CAS), I would spend as much time as it took to integrate a difficult function and compute the area under its curve. Now I just let CAS do numerical integration and go with the answer. i care little about the form of the integrated function, although even limited CAS can do indefinite integrals as well.
Nowadays, many teachers want graphing calculators replaced with powerful CAS on computers. The sky's the limit to the applied math you can teach with that.
I'm a bit of a throwback, I'm afraid. I know that I truly understand a process if I can program it. It's almost all spreadsheets now as the modern languages have outrun me.
Capital is replacing labor in virtually every field. Heck, we don't even have guys hanging off the backs of garbage trucks around here.
Robots and AI are taking over. For surveying the days of construction staking is almost gone. We just finished monumenting a subdivision. The infastructure was all built with machine control. No stakes. The, dirt, streets, water, sewer, inlets, utilities, all of it was hands free. Fine by me, next will be other tasks, who knows I might be out of a job soon.
And teaching, it could be all AI generated on a screen soon. Imagine all lessons programed with an AI teaching,,,,,,,what could go wrong with that?
And we thought Mr. Donahue was dry, imagine robot Karen. (they recently released a movie called Karen Unleashed-all AI)Karen: Unleashed | Official Trailer 2025 | AI OR DIE Productions