Trimble Connect User Forum

 View Only
Back to discussions
Expand all | Collapse all

DA2, Catalyst, RMS Vs Estimated Processed Accuracy?

  • 1.  DA2, Catalyst, RMS Vs Estimated Processed Accuracy?

    Posted 08-18-2025 14:09
    Hi

    I'm using a DA2 with Terraflex and Catalyst 60 or occasionally Cat1, to measure mountain summit elevations, in various countries.

    1. I notice that Terraflex displays Estimated Processed Accuracy figures, for horizontal and vertical, in addition to RMS accuracy values. I see this in eg the United Kingdom, but not in eg Turkey, whether using RTX Internet or RTX Satellite. Why is the Estimated Processed Accuracy value displayed in some locations but not others? In all cases I'm using a workspace with the local coordinate system for the locality in question.

    2. How is the RMS accuracy figure derived? It's presumably derived from the variation the unit is seeing in real time. A bit of detail on this would be interesting. 

    3. What are typical/expected actual accuracy values for Catalyst 1, using RTX? Are there published studies or data? For various locations? Catalyst 60 same question - I can do my own analysis of multiple values for a single location near home in UK over many days, as I have an annual C60 subscription, but C1 would be too expensive, and Trimble must have done their own testing.

    4. Is there any info on how Catalyst correction is "downgraded" from C1 to C60?

    Any info appreciated, thanks.
    Rob Woodall 


  • 2.  RE: DA2, Catalyst, RMS Vs Estimated Processed Accuracy?

    Posted 08-23-2025 22:08
    Hi

    No response to this, unless I missed it?

    Is there somewhere else more suitable for these questions?

    Or a contact at Trimble I could discuss with?

    Thanks 
    Rob



    Original Message


  • 3.  RE: DA2, Catalyst, RMS Vs Estimated Processed Accuracy?

    Posted 08-25-2025 20:04

    Hi Rob,

    The 'Estimated Processed Accuracy' values in TerraFlex indicate that you have Offline GNSS Corrections enabled for your project. This is an alternative GNSS correction source that can be used alongside real-time GNSS corrections, and works by logging raw GNSS measurements in the field and then processing those in the cloud after. This process relies on GNSS measurements from another source, e.g. a base station network, to be used as part of processing the measurements. You can see here which base stations are available for TerraFlex to use, in the UK the OSNet network is available while in Türkiye there are much fewer base stations available.

    The RMS error is based on a 68% confidence interval, or if you were to log 100 positions on a point then 68% of them should have actual error less than the error estimate. We talk about the different sources of error in our blog post here that can impact the quality and accuracy of a position the GNSS receiver generates, and it is these sources that directly impact the accuracy value you see in TerraFlex.

    The Trimble DA2 Spec Sheet outlines the expected performance when using RTX corrections. For Catalyst 1 the specification is 2cm RMS horizontal error, and 5cm RMS vertical error. There is an important footnote there noting that there are a number of factors that can impact achieving this value - for example you should not expect to achieve this level of performance operating under tree canopy where there are a number of obstructions that can contribute to increased error. I would guess that mountain summits generally have a clear view of the sky/satellites and so you are more likely to achieve this level of performance!
    For Catalyst 60 the performance you can expect is limited by your Catalyst license, in this case to 60cm RMS. This is regardless of the GNSS correction technology that you are utilizing, and has the same requirements mentioned above regarding GNSS conditions etc.

    If you have any further questions please add them here and I can try and answer

    Thanks



    ------------------------------
    Hamish MacMillan
    TerraFlex Product Manager
    ------------------------------

    Original Message


  • 4.  RE: DA2, Catalyst, RMS Vs Estimated Processed Accuracy?

    Posted 08-27-2025 10:43
    Thanks Hamish - much appreciated.

    Regarding 'Estimated processed accuracy', these two fields are only being displayed in workspaces using the British coordinate system i.e. OSTN15/OSGM15. It's useful to have an indication other than just the RMS values, but isn't appearing in other workspaces.  I create workspaces in Trimble Connect, which are then updated to TerraFlex (premium version) for field use. Coordinate system is as proposed by Connect, based on the specified location. None of the overseas workspaces include the estimated processed accuracy fields; nor does the Lat long / WGS84 / height above ellipsoid. I'm just curious why this would be. I've not been using post processing, just real time correction (Cat 60). Apart from Turkey, other places where I've used the DA2 are Chile (plenty of base stations on the map), and Ethiopia (from memory, 3 or 5 base stations in Ethiopia although none show on the map) - Catalyst 60 seemed to be working as expected in these locations. Another difference I've noticed, is that for OSTN15 the correction is showing as Internet or occasionally Satellite, whereas for the other workspaces it's generally RTX (Internet) or RTX (Satellite)

    RMS being 68% confidence - what would be the expected ranges for 95% confidence, as a better indication of "accuracy" (Cat 60 and Cat 1)?

    The Catalyst 1 specification RMS values of 2cm H and 5cm V - this sounds right for my unit. Mountain summits tend to have items such as tall cairns and triangulation pillars, but staying away from these and allowing an offset for summit height, gets good results, typically using 30 satellites while correcting.

    Offline correction  - I've been trying to make this work over the last day or so, but no luck as yet. I create a form, to collect values corresponding to 2, 10 or 20 minutes,  Offline GNSS Corrections is turned on in the Offline GNSS tab under Positioning, the data is post processed within 24 hours and PDF reports created in the events folder, sometimes corrections are identified, but it's always stated that no forms are updated. My Cat 60 licence is active during correction - I'm not aware of how to disable it, but 20 minutes of data should already be sufficient to improve on C60 corrected data, so something isn't working. Is more guidance available?

    For context - I bought the DA2 for locating historic survey stations, which in the UK are often buried, sometimes 1-2 feet down. This works pretty well, with an annual C60 subscription and some on-demand C1 sessions where needed. More recently I've been using C60 for measuring summit heights too. Observing the read-out, and comparing measurements at a single location over time, I suspect vertical accuracy is usually better than 1 metre, but unsure, hence my question above. Post-processing should I hope achieve close to 0.1m? I can use on-demand but it's more expensive. So I want to have the option of collecting say 20 to 60 minutes data and post processing.

    Thanks 
    Rob





    Original Message


  • 5.  RE: DA2, Catalyst, RMS Vs Estimated Processed Accuracy?

    Posted 09-01-2025 21:58

    Hi Rob,

    If you are willing to invite me to your Trimble Connect project then I can review why the offline GNSS correction estimates are not available for all of your workspaces.

    Post-processing will still produce a result aligned with your Catalyst license - if you have a Catalyst 60 license when you capture the data then the best post-processed result you can achieve will also be 60cm.

    Thanks
    Hamish



    ------------------------------
    Hamish MacMillan
    TerraFlex Product Manager
    ------------------------------

    Original Message