Nine Seconds Of Silence On The Tower Frequency. It Got Louder 61 Metres Down.
The near-miss was blamed on a construction site. Three receivers hired by this newspaper took bearings for eleven nights; they cross 340 metres from the cranes. Lowered down a storm-drain shaft, the signal rose all the way to the bottom.
A months-long reporting project. Documents cited below are held in The Vault and available to readers.
WASHINGTON, United States — For nine seconds on the night of 11 August, an air traffic controller in Washington transmitted into a carrier and nothing came back.
The recording is public. There is speech, then a tone, then the tone stops and the speech resumes mid-word. In those nine seconds a helicopter on an executive transport movement and a departing airliner closed to a separation the agency has given as four hundred feet.
The interference was attributed, six days later, to construction plant on a nearby site. That attribution is not wrong. It is not sufficient either, and the difference is measurable for about the price of a used car.
What we did
We hired three receivers.
They are not exotic. They are the sort of calibrated field units a broadcaster uses to chase a stuck transmitter, and the hire cost less for eleven nights than this newspaper spends on a single administrative appeal.
We put them on three fixed roofs, ran them from midnight to six, and logged every burst of three seconds or longer. Each station took its own bearing. The bearings were crossed afterwards, by somebody who had not stood on any of the roofs.
Eighty-one bearings. Our engineer discarded four of her own as multipath — reflections off a steel frame, arriving from the wrong direction. The remaining seventy-seven cross in an ellipse ninety metres across.
The ellipse is 340 metres from the construction compound.
The three nights the site was dark
On 14 August the site stood down for weather. On 15 August it was a weekend and no plant ran. On the night of 17 August the compound's supply was isolated between one and five for a service tie-in.
The bursts did not move. Forty, forty-one, forty.
A crane that is not turning does not emit. That is the whole of the test, and it takes a week and a notebook.
The hole
A horizontal fix is a horizontal fix. It says where, not how deep, and this newspaper has been rightly beaten about the head before for treating those as the same sentence.
So we went down.
The district water authority runs a storm-water inspection shaft inside the ellipse. It is 61 metres deep, it is open to permitted access, and the permit cost $180 and took nine days.
We lowered a fourth receiver on a calibrated line and took field strength every three metres.
At street level, the burst is 21 decibels weaker than it is at the bottom. It rises the whole way. It does not peak at fifteen metres, or at forty, or at the invert. It was still rising when the rope ended.
What that does and does not establish
It establishes that the source is below the street and not on a roof, which is the thing a bearing alone could never say.
It does not establish a depth. The agency's spectrum office makes the point, in its reply printed below, that a vertical concrete shaft is itself a fair antenna, and she is right that we cannot fully separate the two. We put the objection to our own second processor before we put it to her. He would not sign off on the depth figure either. We have printed the number anyway, with his refusal attached to it.
The log we were already holding
Since March this newspaper has held an extract of the Federal Aviation Administration's primary-return anomaly log for Sector 41, released under the Freedom of Information Act after an appeal, and published in The Vault.
It records twenty-three uncorrelated primary returns in November. Twenty-three in December. Twenty-three in January. Twenty-three in February.
In the margin, in a hand we have never identified: twenty-three every month for four months. ask CN.
We published that document in March and did nothing with it until an airliner and a helicopter came within four hundred feet of each other in August. Our own public editor has taken the paper apart for that at the foot of this page and we have printed it without a word removed.
We are not joining the twenty-three to the emitter. Nothing in our evidence joins them. We are saying that both sets of numbers are constant, that constants have causes, and that the second one is now 340 metres from a construction compound and at least 61 metres under the road.
What this desk takes from it
That the official explanation was true and stopped one question early, and that the question it stopped at cost $3,120 to ask.
The raw audio, all eighty-one bearings including the four discarded, the shaft profile and both invoices are published in full. If a reader with better receivers wants to stand on our roofs and find that we are wrong, this newspaper will print that on the front page.
The six o'clock edition
Three stories a morning, in your inbox before the coffee is made.
Sources & Method
We hired three calibrated receivers and ran them from three fixed points for eleven consecutive nights, logging every burst of three seconds or longer between midnight and six. Bearings were taken independently at each station and crossed afterwards, not during. The depth measurement was made by lowering a fourth receiver down a municipal storm-water inspection shaft under a permit bought at the counter for $180.
Who we spoke to
- Marisol Egbuna-Traut, Radio engineer; ran our three receiver stations. Contracted by this newspaper at a published day rate; logs supplied raw 8–19 August 2026 Took the bearings. Discarded four of eighty-one as multipath and says so in her log.
- Dr. Anselm Ferreira-Ostrowski, Spectrum measurement specialist, second and independent processor. Given the raw recordings with the metadata stripped; did not know what the story was August 2026 Reproduces the bearing crossing to within 90 metres. Will not endorse the depth figure.
- Federal Aviation Administration, Office of Communications. Written questions 19 August; answer received 26 August August 2026 Printed in full under Right of Reply.
- A retired controller, Twenty-two years on the position; named in our notes, not in print at their request. Interviewed twice by telephone; the second call in the presence of a second reporter August 2026 Identified the tone by ear from our recording before being told what it was.
- The district water authority, Access permit 2026-K-4471, storm-water inspection shaft. Applied for at the counter, granted in nine days, fee $180 August 2026
Documents
- PX-1330 — Federal Aviation Administration primary-return anomaly log, Sector 41, extract accepted
What we could not confirm
- What the emitter is. We have a frequency, a duty cycle and a depth floor. We have no equipment, no owner, no cable and no plan showing anything at the fix.
- How deep it actually goes. The shaft is 61 metres and the field was still rising at the bottom of it. Sixty-one metres is where our rope ended, not where the source is.
- Whether it has anything to do with the twenty-three uncorrelated primary returns logged in Sector 41 every month since 2025. The numbers sit next to each other in our notebook. Nothing in our evidence joins them, and we are not going to join them for the reader.
How Others Covered This
The same events, as reported elsewhere on the same day. We list what each outlet had that we did not, as well as what we had that they did not — including where we come off worse. Why we print this.
-
The Continental WireConstruction Work Cited In Helicopter Radio Problems
Ran the agency's account of the near-miss and the construction finding, accurately, the day it was issued.
Had that we did not
The separation figure, the times, and the agency's own explanation in full.
Left out
That the explanation is testable, that testing it costs about nine hundred dollars, and that nobody in the press pack has tested it.
-
The Hollow PostTHEY ARE JAMMING THE CAPITAL AND THE FAA IS COVERING FOR THEM
Reported the interference as deliberate jamming of civil aviation by an agency it did not name.
Had that we did not
That the interference is real, continuous, and not from the cranes. On that, this newspaper agrees.
Left out
That an emitter is not a jammer; that the bursts sit on a harmonic and not on the frequency; and that nothing we have found shows intent, which is duller and harder to write.
-
They Buriedthis newspaperNine Seconds Of Silence On The Tower Frequency. It Got Louder 61 Metres Down.
Rented three receivers, took bearings for eleven nights, then put one down a hole.
Had that we did not
Every bearing, including the four we discarded, and the raw audio entire.
Left out
We have held an anomaly log since March that says twenty-three every month for four months, and we published it, and then we did nothing with it for five months until an airliner and a helicopter came within four hundred feet of each other. The log was not waiting for a near-miss. We were. — V. Ashcombe-Doyle, standards editor
Right of Reply
They Buried contacted Delphine Okoro-Rasmussen, Deputy Associate Administrator for Spectrum, Federal Aviation Administration sent the full draft, the receiver logs and the shaft measurements; given seven days. Asked that her answer run whole rather than in quotation. Printed unedited.
I am not going to tell you there is no interference, because there is, and your recordings are better than the ones we have.
I am going to tell you that you have measured a real thing and then written a sentence your measurement does not support.
Direction finding in a city is not direction finding. It is direction finding plus forty blocks of steel frame, six hundred rooftop antennas, two river crossings and a subway box that behaves like a waveguide below about 200 megahertz. Your engineer knows this — she threw away four of her own bearings for exactly that reason, which is to her credit and ought to have given your editors pause about the other seventy-seven.
On the shaft. A signal that gets stronger as you descend a vertical concrete tube tells you about the tube. It does not tell you what is under the tube. You went down a structure that is, electrically, a rather good antenna, and you found that the antenna worked. I would have found the same thing.
What I will not do is pretend the construction finding is the whole answer, because your chart is the chart, and I have looked at it for two days and I cannot make the count fall on the nights the site was dark. That is a fair hit and I am not going to dress it up. Our finding was that the works contributed. It was not that the works were the only contributor, and where our language let people read it that way, that is on us and I have asked for it to be corrected.
What I would ask, in return, and I am aware I am asking a newspaper for a favour: publish the four bearings you discarded next to the seventy-seven you kept. You have made a great deal, correctly, of what other people leave out.
Published unedited under our right-of-reply guarantee.
How was this story?
We publish the result, whatever it is. Reader verdicts appear on the front page and in our newsroom metrics.
Readers' Letters 0
Printed at once under the name you give and read by the desk afterwards; anything unfit is removed, with a note saying so, and nothing else is ever deleted — only corrected. Letters that changed something in the story carry a mark saying so, and there are 5 of those across the archive.