# Single-mm precision within a 100 m area? **URL:** \ **Category:** GPS/GNSS\ **Created:** [July 14, 2024, 12:20am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389 "2024-07-14T00:20:02Z")\ **Posts on this page:** 14\ **Page:** 1
**Author:** ![rmann](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/rmann/32/6136_2.png) [@rmann](https://community.sparkfun.com/u/rmann)\ **Post date:** [July 14, 2024, 12:20am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/1 "2024-07-14T00:20:02Z")
Is it possible to get single-digit mm precision relative to a base station 100-150 m away? I know I can do ~15mm precision with RTK for absolute global positioning, but can I get 1.5 mm precision between points in an area? Say, for example, I wanted to lay out a rectangle that’s 90 m X 50 m, but its overall position can be +/-50 cm ---
**Author:** ![YellowDog](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/yellowdog/32/1782_2.png) [@YellowDog](https://community.sparkfun.com/u/YellowDog)\ **Post date:** [July 15, 2024, 11:58am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/2 "2024-07-15T11:58:34Z")
I do not believe 1.5mm precision is achievable with current technology. ---
**Author:** ![rftop](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/rftop/32/6121_2.png) [@rftop](https://community.sparkfun.com/u/rftop)\ **Post date:** [July 15, 2024, 1:08pm UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/3 "2024-07-15T13:08:59Z")
Agreed, I can think of a couple ways to produce a 90.000X M x 50.000X M rectangle on the ground, but GNSS isn’t one of them. This proposed rectangle is well into First-Order Accuracy Standards for Surveying. I’d start with a 1/2 arc-second Total Station. The field work would be significant as 3 of the 4 points will be moved/adjusted on the ground several times (causing serious practical challenges). A “Stake-Out” at this accuracy is much more difficult than measuring existing points to the same accuracy. ---
**Author:** ![rmann](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/rmann/32/6136_2.png) [@rmann](https://community.sparkfun.com/u/rmann)\ **Post date:** [July 16, 2024, 2:13am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/4 "2024-07-16T02:13:23Z")
Ah, okay. I was hoping you could do this. I don’t quite understand @rftop’s references to surveying standards. I tried to google it, but it requires more education than I can invest right now. Thanks! ---
**Author:** ![Yolosaki](https://avatars.discourse-cdn.com/v4/letter/y/41988e/32.png) [@Yolosaki](https://community.sparkfun.com/u/Yolosaki)\ **Post date:** [July 16, 2024, 2:17am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/5 "2024-07-16T02:17:16Z")
Achieving 1.5 mm precision relative to a base station 100-150 m away is challenging with standard RTK (Real-Time Kinematic) GPS systems. Here’s why: 1. RTK Precision: RTK GPS can provide centimeter-level precision (around 1-2 cm) in absolute positioning when corrected against a base station. However, this precision is typically for absolute positions on the Earth’s surface. 2. Relative Precision: When measuring distances relative to a base station, the achievable precision is affected by several factors: 3. Practical Limits: While RTK systems can achieve centimeter-level accuracy for absolute positioning, achieving 1.5 mm precision over relative distances (like between points in a localized area) is highly dependent on the specific conditions: 4. Alternative Solutions: For precise relative measurements over short distances (like within a 90 m x 50 m area), surveying methods such as total stations or laser measurement tools are often used. These can provide higher precision in localized measurements compared to RTK GPS over short baselines. In conclusion, while RTK GPS is excellent for achieving centimeter-level accuracy in absolute positioning, achieving 1.5 mm precision over short relative distances would typically require specialized equipment and careful calibration beyond what standard RTK GPS systems can provide. ---
**Author:** ![rftop](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/rftop/32/6121_2.png) [@rftop](https://community.sparkfun.com/u/rftop)\ **Post date:** [July 16, 2024, 3:20am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/6 "2024-07-16T03:20:20Z")
Sorry, I didn’t intend to cause any confusion. > [@rmann](#): > > Is it possible to get single-digit mm precision relative to a base station 100-150 m away? If you need sub-cm precision relative to the base (which is single digit mm), then the answer is yes 🙂 A very short baseline (your base being 100m away) is a good scenario. I would certainly expect to be able to reproduce RTK points to Sub-cm with such a short baseline. > [@rmann](#): > > but can I get 1.5 mm precision between points in an area? It’s the 1.5mm statement that throws this into a whole new realm of “Controls” Surveying and increased complexity. Does that help ? ---
**Author:** ![rmann](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/rmann/32/6136_2.png) [@rmann](https://community.sparkfun.com/u/rmann)\ **Post date:** [July 16, 2024, 6:30am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/7 "2024-07-16T06:30:23Z")
> [@rftop](#): > > If you need sub-cm precision relative to the base (which is single digit mm), then the answer is yes 🙂 A very short baseline (your base being 100m away) is a good scenario. > I would certainly expect to be able to reproduce RTK points to Sub-cm with such a short baseline. Ah, this is what I wanted to hear. Perhaps I’ll buy the components to build a base station and rover and see what I can do. Thanks! ---
**Author:** ![David\_Nicol](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/david_nicol/32/6597_2.png) [@David\_Nicol](https://community.sparkfun.com/u/David_Nicol)\ **Post date:** [September 6, 2024, 9:06am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/8 "2024-09-06T09:06:39Z")
Hello, I have 2 Sparkfun GNSS RTK Facet antennas, I would like to base one and use the other to have accuracy in difficult area. How can I do? Thank you. ---
**Author:** ![Eric\_S](https://avatars.discourse-cdn.com/v4/letter/e/aca169/32.png) [@Eric\_S](https://community.sparkfun.com/u/Eric_S)\ **Post date:** [September 6, 2024, 1:30pm UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/9 "2024-09-06T13:30:05Z")
With the last firmware update (1.50) of the F9P, there is a new mode: “Conservative Ambiguity resolution mode for applications requiring reduced false RTK fixing while increasing time to ambiguity fix”. It may help in difficult environment. (What do you mean by difficult area?) ---
**Author:** ![Eric\_S](https://avatars.discourse-cdn.com/v4/letter/e/aca169/32.png) [@Eric\_S](https://community.sparkfun.com/u/Eric_S)\ **Post date:** [September 9, 2024, 7:47am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/10 "2024-09-09T07:47:35Z")
BTW, to answer the initial question i.e. how to get precise relative positioning (for building construction), the best practice is to visit all points quickly (\<1 hour), so that ionosphere and satellites configuration don’t change. I would add that starting and finishing with the same point, you may use the difference between the two measurements to correct intermediate points. ---
**Author:** ![David\_Nicol](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/david_nicol/32/6597_2.png) [@David\_Nicol](https://community.sparkfun.com/u/David_Nicol)\ **Post date:** [September 9, 2024, 11:59am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/11 "2024-09-09T11:59:12Z")
the places where the RTK is floating ---
**Author:** ![Eric\_S](https://avatars.discourse-cdn.com/v4/letter/e/aca169/32.png) [@Eric\_S](https://community.sparkfun.com/u/Eric_S)\ **Post date:** [September 10, 2024, 1:21pm UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/12 "2024-09-10T13:21:52Z")
Why is it floating? Urban canyon? Vegetation? Is the reference station in an open area? Overall, if dual-frequency doesn’t allow you to go to RTK fix, you will have to go to triple-frequency to have better chance to get the fix. ---
**Author:** ![David\_Nicol](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/david_nicol/32/6597_2.png) [@David\_Nicol](https://community.sparkfun.com/u/David_Nicol)\ **Post date:** [September 13, 2024, 7:08am UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/13 "2024-09-13T07:08:46Z")
Thank you for your answer, how can I do it? is possible on NTRIP Lefebure? ---
**Author:** ![sparky](https://yyz1.discourse-cdn.com/sparkfun/user_avatar/community.sparkfun.com/sparky/32/6021_2.png) [@sparky](https://community.sparkfun.com/u/sparky)\ **Post date:** [September 13, 2024, 10:37pm UTC](https://community.sparkfun.com/t/single-mm-precision-within-a-100-m-area/59389/14 "2024-09-13T22:37:48Z")
Hi David\_Nicol - I am confused by your original question: > I have 2 Sparkfun GNSS RTK Facet antennas, I would like to base one and use the other to have accuracy in difficult area. How can I do? What do you mean by ‘difficult area’? Do you have cellular reception? Then use Lefebure as the NTRIP Client on your phone, or use the [NTRIP Client](https://docs.sparkfun.com/SparkFun_RTK_Firmware/menu_gnss/#ntrip-client) built into the Facet. What Eric\_S is referring to is a new setting recently released by u-blox that will require multiple steps including upgrading the firmware on the ZED inside your Rover Facet to HPG v1.50, then use the latest u-center (v24.05) to modify the CFG-NAVHPG-DGNSSMODE to RTK\_CAR. ![image](https://canada1.discourse-cdn.com/sparkfun/original/2X/c/c96f09c429d055625cb8ce13aa3d840956b7f236.png) This will be a bit of work so I want to make sure we understand what your root problem is.