The RTK Express is a nice handheld GPS unit.

Outdoors the RTK Express as a standalone device in Rover mode seems to easily get less than 1 meter horizontal accuracy. When combining it with NTRIP data via a cell phone's Bluetooth, the RTK Express is able to get around 1 cm horizontal accuracy. I am using the RTK Express to do some surveying where I need horizontal accuracy much less than 1 meter. The RTK Express is working nicely for my purposes.

As mentioned on the website, the firmware is open source. I have been able to update the firmware on the unit from v1.10 to v1.11 using the SparkFun provided binaries. Additionally, on my Linux system, I am able to compile the open source firmware and load it successfully on the RTK Express.

I am blown away by the accuracy of this product. Even using a Base station 50 miles away, I am able to get RTK fixes with 20 - 25mm accuracy. I mounted mine on a 2-meter SECO rod with bipod supports. The recommended brackets work perfectly with this monopole, and allows me to handle my tablet without moving the antenna. I would definitely buy this again.

We've purchased quite a number of these units to serve as stand-in units for new team members who do not have access to the $5k to $30k systems from professional-grade units that we use from Novatel, Hemisphere, etc. We've found that this SparkFun system handles basic DGPS functionality or GPS rover functions quite well. And with the instructions hosted by SparkFun, both modes were quite easy to set up. However, the professional units still have obvious benefits including faster lock, hot start, selectable constellation support, better in-house integration with IMUs, etc. But for the cost, this unit does surprisingly well, particularly for basic, hobby-type applications.

Why a 4 star of 5 rating? We've seen a few design issues. And we are sharing them here with hopes that will hopefully fade away in future revs. For example:

1. The directionality of the antenna is not specified, so it's anyone's guess how to orient the antenna to north per NOAA DGPS base-station standards. Yes, this matters - twist the antenna, and the position lock measurably changes by a few mm for a DGPS base station, and only one of those orientations is true. Any why waste mm's due to something so simple as orientation? After some digging, our guess is that the manufacturer spec is followed like with other antennas, that the connector is the "North" side of the antenna disk. But the documentation for this is not easily found for this unit, and so this is still our educated guess.

2. The connector at the bottom of the antenna is placed such that the connector's output interferes with standard adjustable mounts for DGPS antennas (see: for example, https://www.allenprecision.com/seco-adjustable-tilt-monument-adapter-w-mpt-5-8-x-11-tpi) . These are very standard mounts, and to make an antenna that is not physically compatible is frustrating. To fix this, we used a threaded stand-off between the antenna and calibration mount, which is annoying as it adds positioning errors unnecessarily. Again, a mm here, a mm there, and one quickly loses the cm-level accuracy that the system can and should produce.

3. The serial numbers for the antennas are not clearly visible externally, so it is impossible to create the documentation required for documenting DGPS antenna specs for use in base stations (again, see the NOAA standards, or those used for setting up a CORS station).

4. There's not an obviously way to "hot start" the unit at a known location, so depending on the signal environment, we've found it can take about a minute - and sometimes much longer - to get a good initial lock even with a roof-top full-sky visibility. However, we've tested these on very dynamic conditions such as an RC car racing around a course at roughly 50 mph / 80 kph, and appears to maintain very good dynamic lock after initial setup. We haven't tested intermittent dynamic lock, such as driving a vehicle under a bridge - this is challenging for most systems, particularly due to the multi-path issues under bridges.

5. The antenna connections into the antenna are somewhat fragile - we've broken several with just basic setup handling (which we do quite a bit). We suggest that users buy a few extra SMA antenna cables. And even if your hands are truly frozen like ours were when setting this up on a rooftop during sub-freezing weather, you should handle the cables with gloves off to avoid damage to the cable. This is another difference versus professional antennas: our pro units use much more robust, but more expensive, cable connectors that we routinely handle with leather gloves on, for 20 years, with no wear/tear.

6. We've tested the maximum distance that we can string the antennas using strings of 10-meter SMA cables (see SparkFun product: CAB-17495), and we found that signal loss starts to occur at about 4 cables, sometimes 3. This chaining gave us roughly 30 meters of cable length from antenna to base station, which is within the spec given by the manufacturer (100 feet). This is similar to what we've seen for run-of-mill professional units. We're sharing this distance test in hopes this info is useful for others who, like us, want to mount the antenna far outside and far above the receiver.

We're still shaking out the antenna, and may update this review as we get more use, particularly in its ability to survive harsh weather. But so far, we are quite happy overall, even with the minor issues above. The units are so cost effective that we readily buy more as needed, which is is a game-changer in DGPS systems. And many of these issues could be resolved with even a bit more documentation. Such relatively minor updates could could push this unit into becoming the default hobby-level DGPS solution.

Overall: from what we've seen with roughly a year of use, this is the first hobby-level GPS kit that, in turn-key form, gives professional GPS receiver kits some very real price versus performance competition.

I found the unit very easy to set up from the instructions on the website. It worked well and was accurate according to my testing. The forum was very helpful and responsive. However after six days the display stopped working and the unit froze. I was instructed to return the unit for repair/replacement. I have not received the repaired unit back yet. When I receive the returned product I will update my review.