Flat earther Jeran Campenella is most famous for confirming the earth curves on the Behind the Curve documentary when he created a test to try and “prove flat earth”. It didn’t go as he wanted and when the results matched the globe and contradicted the flat earth he said “Interesting”. This page contains the details of this test.

Experiment Setup

There is a camera on one side and an assistant, Enrique, holding a bright light 3 miles away. In the middle, there is a single board with a hole cut in it. The documentary included a diagram that was incorrect since there were not two boards. The details of the experiment as told by an eye witness are listed below.

Flat Earth Prediction

The camera, the hole in the board, and the light are all 17 feet above the water. Remember, in the actual experiment, there was just a single board in the middle. If the earth is flat, the light should be seen through the board. The light was never seen when it was held at 17 feet high.

Globe Earth Prediction

With the camera still at 17 feet high and the hole in the board at 17 feet high, Jeran used the 8 inches per mile squared formula to predict a 6 foot drop over 3 miles. Jeran predicted that the light needed to be 23 feet above the water. The problem is that the board in the middle had a hole at only 17 feet. The panel would block the light’s path for a light held 23 feet above the water.

Usually, flat earthers incorrectly use the 8 inches per mile squared. but in this case, it is the correct formula. He just applied it wrong. The hole in the panel is too low, or the light is too high.

Let’s calculate the correct height of the light. To do this, we don’t need the drop from the light to the camera, we need to trace the light’s path and calculate the drop to the lowest point of the light’s path along the entire 3 miles.

The lowest point is halfway between the camera and the board: 2.25 miles from the light.

Using 8 inches per mile squared the drop is 40.5 inches.

Then the light’s path rises back up to the height of the camera for the remaining 0.75 miles. The same formula is used to get the rise of 4.5 inches.

Take the drop of 40.5″ and subtract the rise of 4.5″ and you get 36 inches or 3 feet. Add that to the 17 feet height of the camera and the globe prediction is 20 feet. However, this excludes refraction. To be more accurate the effects of refraction should be included. Refraction effects vary based on the local conditions, see the measurements of the effects of refraction for more information:

At night over land, a conservative Coefficient of Refraction, often denoted as “k” in the literature and formulas, is +0.167. I will calculate based on this value. If the Coefficient of Refraction was higher, it will reduce the predicted height of the light.

If anyone has more details on the weather conditions of the day the experiment was performed, a more accurate Coefficient of Refraction can be determined. Jeran, feel free to give your accounting of the weather so we can get a more accurate refractive analysis:

The formula “8 inches per mile squared” is correct for drop when there is zero refraction. To calculate the drop using the same general formula “X inches per mile squared” the value of X needs to be determined.

The formula to calculate the “8” in 8 inches per mile squared has been derived in detail by Walter Bislin. Specifically, see formula 23 from Walter’s page:

Note that 5280*12 = 63,360, this is the number of inches in a mile.

To include the Coefficient of Refraction, “k”, divide the radius of the earth by (1-k). The radius of the earth is in the denominator in this formula so (1-k) is in the numerator.

The corrected formula for drop including k=+0.167 refraction is “6.67 inches per mile squared”. Note that there is nothing “Standard” about this number. It is an estimate based on common conditions. Jeran did not measure the vertical temperature lapse rate so an estimate based on common conditions must be used. For example, using k=+0.333 the formula is “5.33 inches per mile squared”. Feel free to calculate based on different k values in the Python program I link below.

This gives the drop from Enrique to the low point of 33.76 inches. The rise up to the camera is 3.75 inches. The drop from the light to the camera including refraction is 30.01 inches or 2.50 feet. The camera is at 17 feet so the correct globe prediction for the light height is 17 feet + 2.50 feet = 19.50 feet.

The correct globe prediction for the light including moderate refraction is 19.5 feet above the water level.

I have written a simple Python program to do all these calculations. Please review it and inform me of any errors. Jeran, I’m asking for your input.

Python program:

Output of Python program predicting the correct height of the light for the globe


During the test, the light was seen never seen when the light was held by Enrique at chest level, which was measured to be 17 feet above the water.

When Enrique was asked to hold the light above his head, the light was seen in Jeran’s camera, as we see in the screenshot below. On many occasions, Jeran stated Enrique was holding the light 19.5 feet high.


The light was never seen when held at the predicted flat earth height, falsifying the flat earth hypothesis. The light was seen then held at the predicted globe earth height, confirming the globe earth hypothesis. This is strong evidence that the earth is a globe.

Eyewitness Information

An eye witness to the test explained the details on a comment on one of Jeran’s videos. Unfortunately, Jeran has deleted the video, but there are screenshots and video clips of the comment. Rather than reproduce all the details here I will link to the sources where I found this information.



Source: Sly Sparkace video “Flat Earth: 2 Tests and a Documentary”

Here I reproduce the text of the eyewitness, brygenon:

I was the globe-head there. In the video at Globebusters, Jeran shows an illustration with four panels. That may have been the plan, but in the field there were three: one right in front of the laser (which I think is useless), one 1.5 miles down-range with a 6-inch hole 17 feet above the water-line, and one 3 miles down-range, with height-lines drawn on it (I assume; couldn’t see it).

By the time I arrived the sun had set and the panel in front of the laser was up. Enrique got a volunteer, Steven, and they left to set up the down-range panels. I stayed at the laser location, as did Jeran and Daniel the camera-man. (Apologies if I have names wrong or misspelled.)

The laser with beam-expander, and the telescope mount that carried it, did not work. Jeran first adjusted the beam-expander with the beam hitting the ground. When Enrique got the 1.5-mile panel up, Jeran aimed the beam using a remote for the telescope mount. Aiming was touchy: it would fail to move, then move way to much. When he got it aimed, Enrique reported by phone that the beam lit the entire panel. To further adjust the beam-expander, Jeran had to hand-manipulate it, throwing off the aim.

Jeran switched to the back-up plan. At the 1.5-mile panel, Steven held a reasonably powerful flashlight. At the 3-mile distance, Enrique held a monster flashlight. Jeran had a long-zoom camera, 17 feet above water, pointed down-range toward Steven and Enrique. The three set up a conference call, and with Jeran directing repeated the following procedure about ten times:

Steven shined his light toward the camera, through the hole in the 1.5 mile panel 17 feet above the waterline. Enrique held his light at 17 feet above water, which was about waist-high for him, and walked to his left so that its light would go around the middle panel. Jeran and I could see both lights in the camera’s flip-out viewfinder. As well as Jeran could focus, the lights were sizable balls not dots, and they appeared to be at about the same height. Enrique, at 3 miles, would walk back to his right until his light was occluded by the 1.5-mile panel, centered as well as Jeran could estimate. Then Steven would turn off his light and remove it from the hole. The test was then whether we could see, in the viewfinder, Enrique’s light through the hole in 1.5 mile panel, which would support a flat earth. If not, could we see it if he held it over his head, well over 17 feet above waterline, which would support a round earth.

In about 10 tries, with the light held waist-high at 17 feet above water we never clearly saw it through the center panel hole. On one of the attempts, maybe the fourth of ten, we clearly saw it when Enrique lifted it over his head. On that one, Jeran asked him to raise it and lower it a few times, and it would appear when Enrique raised it and vanish when he lowered it. That was the “gasp” moment. Jeran said, “that’s interesting.” I noted it was the prediction for a round earth. When they repeated the whole procedure, it did not happen again. I suggested having Enrique move from side-to-side a bit when occluded by the panel, in case it was lateral alignment that was off. Jeran accepted the suggestion, but it made no observable difference light.

Jeran Planning to Reproduce the Test

Below are screenshots from videos where Jeran is discussing plans to reproduce the test. After getting absolutely humiliated in the documentary anyone would want to be vindicated by showing that the documentary misrepresented him by showing that the test works as he claims it should. But he never did this.

Jeran talks about going back to the same location to repeat the test. He talks about how to improve it. He talks about doing the test at a higher elevation to avoid weeds or slightly shifting the location to avoid them.

But he never published any results.

Why is there no publication vindicating his test after more than 4 years?

Jeran, do you want people to stop saying “Interesting” to mock you? Publish the results.

The screenshot below is from when he was discussing doing the test again. The distance of the south side of Victoria canal is 3.88 miles confirmed on Google Maps. This video was deleted from the Globusters channel, but I have found an archived copy.

Screenshot from Globebusters “Flat Earth – NASA Bashing 2.0”
Diagram from Sly Sparkane showing the correct globe vs flat earth predictions for Jeran’s planned test at 3.88 miles distance.