Optimizing Base RTK performance

[charlie (Charlie HUIZENGA)]

My application (topography for gravity flow drinking water systems in Nicaragua) requires a differential vertical accuracy of ~10cm. Based on my testing as well as what I read here, it looks like a FIX solution is required to achieve this. Also, the nearest NTRIP station is over 150km away, so I think I will be limited to using my own base with LORA. I am trying to better understand when the rover is able to get a FIX and have the following questions:

1. The Correction Input data for the rover lists the number of satellites in view. Is this the total number of satellites in the data stream or is it the number of common satellites between the rover and base? I typically see a number like 10-15 here, when the individual receivers are seeing 25-30 satellites.
2. Is there any reason not to use all available satellite types in the base or rover?
3. Does it help to use more selective elevation or SNR masks in the base (or rover)?
4. In cases where I can’t get a LORA signal from the base, I can use PPK. However, there is really no way to know while surveying if it’s possible to get a FIX result. How do people manage this?

Any other insight to improving the chance of getting a FIX solution (apart from the obvious strategy to place the base in an optimal location) would be appreciated!

Thanks!

[PotatoFarmer]

Hopefully there is a way to PPP in your area.

I do not know if Canadas NRCAN PPP would process your data, hopefully it will. Its really easy to do.
Record the latest Rinex for 3-24hrs, the longer the record the more accurate it is. Then submit the file to the service immediately for very good results, or after two weeks for the most accurate results. Works awesome with RS2 dual band receivers.
They email you back a nice report with the coordinates.

With PPP you can make your own monument within 2cm or less of absolute position.

[charlie (Charlie HUIZENGA)]

Thanks - I’ll look into the PPP option which could be useful for some key survey points.

But my main application involves surveying a proposed pipe line from a spring to a tank, and then all faucet locations, so we have many points (typically 30-150). So collecting 3 hours of data per point isn’t an option. Also, I don’t need absolute accuracy, just relative vertical accuracy.

[PotatoFarmer]

You only PPP the Base Position.

Doing a survey relative to itself is great until someone else has to find the points in the future. If they are synced up with an existing datum it makes life much easier.

[kseniia.suzdaltseva]

Hi Charlie,

Welcome to our community!

The Correction Input data for the rover lists the number of satellites in view. Is this the total number of satellites in the data stream or is it the number of common satellites between the rover and base?

The satellites with quite a low SNR may be excluded during correction transmission. That’s why the number of satellites shown for the base differs when you check it on the base and rover. Since the signals with high SNR are transmitted fine, it shouldn’t cause any difficulties with obtaining Fix.

Is there any reason not to use all available satellite types in the base or rover?

We recommend enabling all satellite systems for Reach RS2 base and rover. If you use Reach RS+, please note that GLONASS and BEIDOU can’t be enabled together due to hardware limitations. So, GPS + GLONASS + Galileo + SBAS + QZSS is our default recommendation for Reach RS+. GPS + BeiDou + Galileo + SBAS + QZSS is recommended for APAC, where QZSS and BeiDou are visible.

Does it help to use more selective elevation or SNR masks in the base (or rover)?

In most cases, the default elevation mask 15 and SNR mask 35 should work fine. However, it may be useful to change these values in some cases. For example, if there are not enough data to calculate Fix, you can set a lower elevation mask to track more satellites. If there are too many noisy signals, you may increase the SNR mask.

In cases where I can’t get a LORA signal from the base, I can use PPK. However, there is really no way to know while surveying if it’s possible to get a FIX result. How do people manage this?

There is no straightforward way to check if you will get Fix in PPK during the survey. However, if the base and rover track at least 4 satellites with high SNR and the baseline doesn’t exceed 100 km for Reach RS2 and 30 km for Reach RS+, you should get Fix.

Please note that for points collection in PPK, you will need to record Time Start and Time End for each point manually. These values should be entered in RTKLib to extract data related to each point from the continuous log.

If you need to collect many points, it may be easier to place the base on another point and continue collecting points in RTK. As an alternative, you may consider using a pair of external radios. They are usually more powerful than the Reach built-in LoRa radio.