Mounting a Surveying Antenna to a Bulldozer Blade

GPS/GNSS
21

We are using ScreenStream to monitor the dozer operators screen remotely.

The dozer operator is connected on Discord, and we can provide audible communication about elevations, plans, and ensuring we’re all on the same page. Plus I have cameras where the Dozer is operating to monitor remotely.

image
image604×444 54.5 KB

22

Here are the step by step instructions we use. Through this method, we can essentially level the LASER itself to the exact elevation we need.

SUPPLIES TO TAKE OUT WITH YOU

  • Yellow Tripod GPS (with phone)
  • Black Monopod GPS (with phone)
  • Dual-Slope Milwaukee Laser
  • Torpedo level
  • Chest Harness or Headset for hands-free communication
  • Gloves (as needed)

CALIBRATING YELLOW TRIPOD GPS

  1. Make sure you got the green dot in SW Maps and Wi-Fi is OFF.
  2. Verify Yellow Tripod GPS Antenna is fully contracted
  3. Take Yellow Tripod GPS to the blue, center line on the Edge of Pavement
  4. SW Maps: “Bluetooth GNSS” should have a height of 2.02 meters, and make sure to hit the small check mark and see the verification “Instrument Height Updated.”
  5. The Elevation should be “2683.821.” If the number is off, verify that the black, upper pole is firmly inserted in the Yellow Tripod. (“Within .083 is the error”)

SETUP LASER

  1. Calculate the 100 foot section that you will be focused on, and note the Station numbers
  2. Use the Yellow Tripod GPS to mark approximately 2’ feet past the 100 foot Station (each station is 50 feet) FACING the dozer blade
  3. Place the laser over the mark that is 2’ past the station line
  4. Ensure laser legs are fully extended and firmly pressed into the ground. Ensure that the telescoping head is in its lowest position
  5. TORPEDO LEVEL: Verify that the laser head is level
  6. Ensure the X arrow on the top of the Laser is pointing toward the front of the bulldozer blade either directly NORTH or directly SOUTH
  7. The X-Axis MAINFALL is dependent on the road plans. Review road plans.
  8. The Y-Axis is the CROSS-SLOPE.
  • X-AXIS POINTING NORTH (NORTH FACING): Set the Y-Axis to +2%
  • X-AXIS POINTING SOUTH (SOUTH FACING): Set the Y-Axis to -2%

ALIGN LASER

  1. Verify Black Monopod GPS with laser receiver has a green dot with RTK Fix. Wi-Fi should be OFF.
  2. Have another person at the laser to make adjustments & make sure Bulldozer is parked so it’s not in the way
  3. Use Black Monopod GPS with Laser Receiver to measure elevation of the laser at the closest station line (“BASELINE”) and blue centerline. Record this elevation. ________
  4. Now measure the laser 100 feet away (2 station lines)
  5. Err on the side of laser too high than too low
  • 1% Slope = The laser elevation should be 1 Foot different from BASELINE (within 0.10 error) _________
  • 2% Slope = The laser elevation should be 2 Feet different from BASELINE (within 0.10 error)
  1. Adjust head left or right to change the height of the laser at 100 feet as needed:
  • NORTH FACING LASER: To LOWER the laser, rotate RIGHT or CLOCKWISE. To RAISE the laser, rotate LEFT or COUNTER-CLOCKWISE.
  • SOUTH FACING LASER: To LOWER the laser, rotate LEFT or COUNTER-CLOCKWISE. To RAISE the laser, rotate RIGHT or CLOCKWISE.
  1. Once height is set, confirm the height at 1 station closer to the laser (50 feet) to double check. Adjust height as necessary. Go back to the 100 foot station to recalibrate if needed.
  • 1% Slope = The laser elevation should be 0.5 Foot different from BASELINE (within 0.10 error)
  • 2% Slope = The laser elevation should be 1 Foot different from BASELINE (within 0.10 error)
  1. Put away the black GPS. We are finished with it.

CALIBRATE DOZER GPS

  1. SW Maps: Ensure green dot with RTK Fix in Dozer and on Yellow Tripod GPS. Wi-fi should be off.
  2. Set Dozer green dot 2 stations (or 100 feet) away from laser on blue center line
  3. Dozer blade should be on the ground and level
  4. Ensure all GPS and laser receivers on the Dozer are tightened
  5. Make sure the Laser Receiver is on
  6. Use YELLOW TRIPOD GPS to measure elevation of bottom of bulldozer blade in the center
  7. Dozer SW Maps: “Bluetooth GNSS” instrument height of 3.170 meters, and make sure to hit the small check mark and see the verification “Instrument height updated.”
  8. Adjust the instrument height to ensure the elevation matches the YELLOW TRIPOD GPS
  9. Margin of difference should be within 0.10

RAISING LASER TO DOZER RECEIVER

  1. Set the Dozer blade elevation to 0.167 feet (2 inches) above BENCHMARK (within 0.10 margin) elevation as indicated for the specific station on the blue line
  2. Adjust the height of the laser VERY CAREFULLY until the laser is aligned with the Dozer LASER RECEIVER
  3. The bulldozer operator should be communicating the directional arrows. If the UP ARROW is shown on the RECEIVER, that means that the laser needs to be LOWERED. If the DOWN ARROW is shown on the RECEIVER, that means the laser needs to be RAISED.
  • WARNING: Sometimes the arrows are wrong because of reflections and you need to verify with receiver
  1. Go past above the upper range of the Green box of RECEIVER, and then adjust down to the upper limit

VERIFY LASER AT STATION WITH DOZER

  1. Have Dozer drive to the next station 50 feet closer to the laser with the green dot on the blue, center line
  2. Verify the elevation is 0.167 feet (2 inches) above benchmark (within 0.10 margin). Err on the side of too high

GET PICTURE OF DOZER AUTOMATICS TO ENSURE SET CORRECTLY

image
image757×1422 54.8 KB

image
image714×882 34 KB

3 Likes
23

I have never seen any YouTube videos calibrating the laser height in relation to RTK fix, so I had to develop the procedures through trial and error.

We can complete this process in 45 minutes if everything runs smoothly. Usually takes about an hour because some kind of hiccup occurs which I have to improvise a solution outside the procedures.

I am very happy that I ended up taking on this project myself. Considering the complexity of the setup, I simply don’t know how standard road construction crews can get any sort of accuracy except by literally hiring licensed Surveyors to actually guarantee the placement of the laser.

I imagine hiring a surveyor to setup the laser every morning likely can get expensive.

1 Like
24

This is awesome, thank you for sharing!

2 Likes
25

So one of the EXTREME time saving measures arises when we’re having to remove feets worth of soil. How do we know it’s feets?

Because in the morning, we just align one of the Rovers with the bottom of the Dozer Blade (which essentially acts as its own Rover with 3.07 meter height). It takes maybe 10 minutes. As long as the ellipsoidal heights match (close enough), the dozer is good to go.

We’re assuming it’s unlikely that both the dozer antenna and the Rover antenna altered in relative height exactly.

No need to setup lasers or even send another surveyor out with the operator since we’re so many feet away. Then periodically through the day, we can “recalibrate.”

We have the ScreenStream.io and have him lower his blade to the ground, and we know the exact height. (this is assuming the antenna didn’t move feet in the middle of the day which is always a possibility). We can now let the operator know they are now only X feet away from their target height.

Generally, when operating with a 2 foot margin, you’ll notice if the antenna moved vertically 2 feet. Likely, it would’ve fallen off the actual dozer mast.

When we get within 1 feet of the target height, we will setup the laser.

1 Like
26

Almost a month and a half later and we’re still trucking along.

We’ve found that the CR700 paired is the most cost effective solution for laser receivers, but you MUST have a tech on hand to troubleshoot all the problems that arise. The Johnson laser receivers are OK, if we can get the remote device working. Everything is cabled.

Milwaukee dual slope lasers have been problem headaches. Thank God for Home Depot’s generous return policies. We just return them. I’m not sure if all dual-slope lasers are headaches.

Everything is strung together with any component breaking at any moment.

Speaking of which, the SparFun RTK Expresses have been having issues pop up including Bluetooth connectivity issues, charging components, etc.

The disaster is the shortage of SparkFun RTK Expresses. We can’t get replacements fast enough, and you can’t be stopping a multi million dollar project because the RMA item you sent in does not have a replacement available to send out.

I would like to push this product into my region, but if I can’t be 100% sure I’ll have replacement units, this simply will not work.

The antennas have been rock solid. I just think the vibrations in the Dozer and heavy equipment is perhaps a little too much for the RTK Express units. And then also on the survey sticks.

Despite the headaches, it’s still cheaper than the $100k and hiring professional surveyors.

If you look at my previous posts, I’ve ran into random little hiccups, but luckily, I have a VERY small crew, so any technical delays that we have to figure out, don’t cause me a lot of money. Also, I’m my own client, so I don’t get angry at myself, especially considering the $400/hour surveyors I’m saving on!

4 Likes
27

Late October, Google Satellite view updated, and I can see my KML lines on Google Earth matching perfectly with some of the roads and pads we’ve lined up with the Sparfun RTK products & Reference Station.

I can’t tell you how satisfying it is to see how well the satellite view matches with the KML lines. These are the same KML files that get uploaded to SW Maps, and from SW Maps with the rovers & RTK Express.

Everything has been coming together despite the various challenges.

3 Likes
28

I would also be remiss to not include the numerous times that the surveying capabilities outside of the dozer have been extremely beneficial.

Manhole placement, irrigation pipe, conduit for electrical lines. Elevation matters.

Normally, this is a $1600/half-day minimum sort of job.

Also, checking a monument every morning is important as the Reference Station may end up getting reset on accident and I haven’t figure out how to save the settings. Without the offsets, we’re about 6" off both horizontally and elevation wise.

Even without the bulldozer component, the Sparkfun surveying equipment already made back its cost.

2 Likes
29

To explain just how far behind the construction industry is and generally tech retarded, watch how cutting edge the laser is in video.

Meanwhile, one can get even more precision using Sparkfun RTK.

This is a potential business opportunity if you can get the construction guys to stop throwing their iphones in fits of rage.

1 Like
31

Road Project Update

Progress So Far

  • Sub-Base Layer: We’ve compacted the 8-inch pit run sub-base and are about to lay down 4 inches of 3/4-inch road mix, followed by 2.5 inches of asphalt.
  • Manholes: Installations are nearly complete, tying everything into the road’s infrastructure.

Equipment & Techniques

  • Laser Guidance: The bulldozer’s automatic sensors (except cross-slope) have been a bust, so we’re relying on laser receivers and lasers to guide the blade height. A track loader Bobcat with dual laser receivers on its 9-foot blade handles the fine-tuning within 2 inches. Cheapest option for automatic laser adjustments takes care of grading. Likely cheapest due to weird software licensing. Bobcat somehow got the license in-house.
  • Survey-Grade GPS: With half-inch accuracy GPS on site, we’re using it in ways most contractors wouldn’t dream of. Our operators measure blade heights directly and stake out manhole and pipe locations without lasers—way faster and dead-on precise. We’re double-checking everything against the blueprints, no shortcuts.
  • Process: Bulldozer pushes dirt over 2 inches, track loader refines it, then we water and compact with a Sany soil compactor.

Challenges

  • Equipment Durability: Masts break. Antennas have been robust. RTK Expresses keep breaking. Have extras and don’t hesitate to switch out as needed.
  • Precision Work: Half-inch accuracy doesn’t come easy. It’s constant GPS checks, spreadsheet calculations, and geographic math—stuff most contractors wouldn’t bother with. But hey, when you’re your own engineer, contractor, and surveyor, you make it work.

Why This Works for Me

  • Cost Savings: No $400/hour surveyors here. Our operators handle the GPS, slashing costs. And since I’m the client, I’m not suing myself if something’s off—I just fix it.
  • Control: I’m wearing all the hats—engineer, contractor, surveyor—so it’s all on me. No middleman, no finger-pointing.
  • Skills: This project’s a crash course in road building, and I’m soaking it up for whatever’s next—roads, homes, or beyond.

Industry Reality Check

  • Tech Resistance: Most contractors won’t touch this level of tech—too complex, too much training. You’re lucky if they can use a smartphone without chucking it in frustration.
  • Long-Term Focus: Precision now means a road that lasts. Most contractors slap on asphalt with a one-year warranty and call it a day. I’m building for the long haul because I’ll be the one dealing with any screw-ups.
  • IQ Factor: Let’s be real—pulling this off takes a certain level of IQ and determination. It’s not just about following a YouTube tutorial. You need to juggle Excel, geographic math, and a Frankenstein mix of tools. Most contractors? They’d quit halfway if you held them to this standard.

The Bottom Line
This project’s a blend of tech, elbow grease, and a refusal to cut corners. We’re hitting engineered specs at a fraction of the cost, and I’m triple-checking every step with survey-grade gear. If I hired a contractor and nitpicked their work like this, they’d probably walk off the job. But here, I’m building a superior road that’ll outlast the typical slapdash work—and I’ve got the skills to prove it. March’s Google Satellite update further solidifies the accuracy.

Not exactly DIY, since I built up a full-on construction company along with support staff, but I think can be DIY assuming you have the time, determination, and IQ.

4 Likes
32

Take a proof at the double ant module like the um982 or lg680p…imagine if the sw can be customized not for simple yaw but for your specific requirement.

1 Like
33

Update, RTK2Go, seems to lag on correction data upwards of 2 seconds. Polaris you’re getting around 0.8s. And a local RTK connection using Reference station we’re getting around 0.5s.

This is using Cell Phone data on the mobile. We don’t have Wi-Fi setup. If we did, I’d imagine it’d be even lower for correction data.

RTK2Go is free, so likely their servers just get overloaded, but this lack of correction data could mean inches in elevation.

1 Like
34

In practice, we were seeing 3 inches or more using RTK2go just this morning. We just figured this out now. And now we can’t seem to figure out how to see the delay.

Quantitative Comparison

Assuming a high-quality RTK setup (multi-frequency receiver, <10 km baseline, good satellite geometry):

  • 0.5 seconds: Elevation accuracy ~1–1.5 cm (0.39–0.59 inches), reliably within 1.27 cm.
  • 0.8 seconds: Elevation accuracy ~1–1.5 cm (0.39–0.59 inches), nearly identical to 0.5 seconds, within 1.27 cm.
  • 2 seconds: Elevation accuracy ~1.5–3 cm (0.59–1.18 inches), may meet 1.27 cm in ideal conditions but risks exceeding it if Float solutions occur.

Conclusion

  • 0.5 seconds and 0.8 seconds: No significant difference in elevation accuracy; both should reliably meet the 1/2 inch (1.27 cm) requirement with a good RTK setup.
  • 2 seconds: Slightly higher risk of reduced elevation accuracy due to potential Float solutions, which may push errors beyond 1.27 cm in non-ideal conditions. To ensure 1/2 inch accuracy, aim for delays <1 second, use a multi-frequency receiver, and keep the baseline short.
2 Likes
35

I personally think this is the most interesting thread on our forums…it has an excellent mix of RTK info, use-case nuances and in-field data…and it’s just a fun topic :slight_smile:

Thanks for continuing to update this with useful info!

2 Likes
36

@jeffersonkim , If you use MSM7 messages, they should include Range Rate Corrections.

From Google:
Yes, it’s possible for GNSS base and rover observations to have a slight delay of a few seconds. This delay, or latency, is due to several factors, including the time it takes the base station to calculate corrections, transmit them to the rover, and for the rover to decode and process them.
To account for the slight delay, the base station often transmits a range rate correction, allowing the rover to back-date the correction to match the exact moment it made the observation.

Please note, I’ve never drilled down into this specific topic before.
I don’t know if your backhaul can handle the increased bandwidth of MSM7 or not.
But this is very interesting to me :slight_smile:
Please keep the updated coming !

1 Like
37

This morning, we were seeing fluctuations using the reference station. Meaning, we would measure once, and just sit there for a couple minutes and watch it jump as much as 2.5".

The vertical accuracy was jumping between 0.394" and 0.44" on SW Maps.

We will be testing with Polaris paid reference stations. Perhaps there could be some issue with our reference stations. Hard to say. We have two SparkFun reference stations.

For horizontal accuracy, our accuracy is good enough. It’s just the vertical that becomes an issue since our road requires as little as 4" of road mix.

If it ended up being the reference station’s fault, then I would retract my complaints with RTK2Go

38

Last night, there was ridiculously high Ionosphere Activity in Mississippi.
You might be experiencing environmental errors this morning?

1 Like
39
1 Like