Separate Receive Antennas: When a Worse Antenna Hears Better
Why a receive-only loop out-hears your transmit antenna on a noisy lot: the near-field mechanism, loop-on-ground builds, T/R switching, and what to buy.
A receive antenna that is 25 dB less efficient than your transmit antenna will often hear better. Not marginally better. Enough that stations you could not copy at all become workable.
This article contains affiliate links. I may earn a small commission at no extra cost to you.
That sounds like it should not be true, and it is the single most useful thing I know about operating from a noisy lot. The RFI hunting article ends by telling you that once you have killed every noise source that is actually yours, the remainder needs a receive antenna rather than more ferrite. This is that article.
Efficiency stops being the metric
On transmit, efficiency is everything. Every dB you lose in a loading coil or a bad ground system is a dB that never leaves the property. That is why the indoor HF piece spends so much time on losses.
On receive, efficiency is nearly irrelevant below about 10 MHz. What matters is the ratio between the signal you want and the noise arriving with it.
The mechanism: on the low bands, external noise (atmospheric plus the aggregate hash of every building around you) sits 30 to 50 dB above your receiver's own noise floor. You are never receiver-limited on 160, 80, or 40 meters. You are noise-limited. So you can throw away 25 dB of antenna output, feed the remainder through a preamplifier, and hear exactly what you heard before, minus whatever local noise the new antenna failed to pick up.
That last clause is the whole trick. You are not trying to collect more signal. You are trying to collect less noise.
This inverts above about 14 MHz. Sky noise on 10 meters is low enough that receiver and preamplifier noise figure start to matter, and a lossy receive antenna genuinely costs you signal. Receive loops earn their keep on 160 through 40, are situational on 20, and are usually pointless on 10.
Why a loop hears less noise than a wire
Household noise sources are close to you. That is not a vague statement, it is a specific electromagnetic condition.
Within roughly one wavelength divided by 2 pi of a source, you are in its near field, and the electric and magnetic components are not locked in the usual 377-ohm relationship. At 7 MHz that boundary sits about 22 feet out. At 3.5 MHz it is about 45 feet. Every switching supply, LED driver, and powerline adapter in your building is inside that radius.
Most of those sources are high-impedance voltage transients on unshielded wiring, which means their near field is dominated by the electric component. A shielded magnetic loop responds to the magnetic component. So against a close, E-field-dominant noise source, the loop under-responds badly.
Meanwhile the signal you want arrived from 4,000 km away. It is pure far field, E and H locked together at 377 ohms, and the loop receives it at full sensitivity for its aperture.
Distance compounds this. Reactive near-field terms fall off as the square and cube of distance. The radiated far field falls off linearly. Moving a receive antenna 40 feet away from the house cuts local noise dramatically and cuts the DX signal by an amount too small to measure.
A small loop also has a figure-eight pattern with maximum response in the plane of the loop and a sharp null along the axis through its center. Point the loop's face at a single dominant noise source and that source drops 20 to 30 dB. The conditions: one source at a time, arriving on a stable bearing (local or ground wave, not skywave), and an electrically balanced loop. An unbalanced or unshielded loop picks up enough E-field to fill its own null, which is why the shield and the symmetry in a good loop are not cosmetic.
The loop on the ground
For anyone with even a small yard and an HOA, this is the answer, and it is the one that fits this site best: a receive antenna lying flat on the grass is not visible from any angle.
A loop on the ground is 50 to 60 feet of wire laid in a circle or square directly on the soil, fed through a step-up matching unit (9:1 is a reasonable starting ratio) into a low-noise preamplifier. It costs about $40 in parts.
It is spectacularly inefficient. Output runs 20 to 30 dB below a dipole because the lossy earth immediately underneath absorbs most of what it captures. That loss is the feature. The same earth attenuates the near-field noise radiating out from your house, and the antenna sits below the level of every piece of household wiring in the neighborhood rather than up in the middle of it.
It is broadband with no tuning, essentially omnidirectional, and it favors high-angle arrivals, which suits regional 80 and 40 meter work well. It does not null anything, so if you have one dominant noise source you want a rotatable loop instead. If you have diffuse neighborhood hash with no single culprit, the loop on the ground is the better tool.
Keep it clear of buried utilities and irrigation wiring, and expect to mow over it. Insulated wire survives that fine.
Active loops: what the amplifier buys and what it costs
An amplified loop puts a low-noise preamplifier right at the loop and powers it over the coax through a bias injector. That solves the low-output problem cleanly and it is why most commercial receive loops are active.
The cost is dynamic range. An amplifier sitting in a strong-signal environment (anywhere near an AM broadcast site) generates intermodulation products that show up as phantom signals or a raised noise floor, and cheap amplifiers do this readily. The specification that predicts it is output third-order intercept point. Higher is better and the good units publish it.
| Option | Approx. cost | Amplified | Where it wins | Where it fails |
|---|---|---|---|---|
| DIY loop on ground | $40 in parts | Needs external preamp | Invisible, broadband, diffuse neighborhood noise | No null, needs yard space |
| Airspy YouLoop (passive shielded) | ~$35 | No | Zero IMD, portable, indoor null steering | Low output, needs a sensitive receiver or preamp |
| MLA-30+ active loop | ~$40 | Yes | Cheapest working active loop | Modest noise figure, poor IMD near broadcast |
| W6LVP receive loop | ~$200 to $250 | Yes | High intercept, rotatable, sold direct from w6lvp.com | Cost, and it is still a small aperture |
| Wellbrook ALA1530 family | Used market | Yes | The long-standing reference design | Availability has been inconsistent since the original maker wound down |
Two market notes worth having before you shop. MFJ stopped manufacturing in 2024, so any MFJ receive loop or switch you find is a used-market item, priced accordingly. And the passive YouLoop is the cheapest way to test whether this whole approach helps at your location before spending real money. If a $35 passive loop hanging in a window improves your signal-to-noise on 40 meters, a permanent install will improve it a lot more.
Getting it into the radio without destroying it
This is where people cook a $200 preamplifier in the first week.
If your rig has an RX ANT IN/OUT pair on the back panel, use it. The radio breaks that jumper and mutes the receive path on transmit, and the whole problem is solved internally. The IC-7610, the TS-590SG, and the Elecraft K3 and K4 have one. The IC-7300, IC-705, and FT-891 do not.
If your rig does not have one, you need an external transmit/receive switch between the loop and the receiver. Two ways to trigger it:
PTT-line driven. The switch is keyed by the same line that keys the radio, so the receive path is disconnected before RF exists. This is the correct approach.
RF sensing. The switch detects your transmitted carrier and then opens. The mechanism guarantees a problem: RF has to be present for the relay to see it, so the leading edge of every transmission reaches the receive antenna before the contacts open. Milliseconds of full transmit power into a preamplifier input, on every single transmission. It works until it does not.
The DX Engineering RTR-1A receive antenna interface handles the switching and the bias injection together and is currently manufactured. Whatever you use, verify the switching actually happens before you key up at full power: key into a dummy load at low power first and confirm the receive path goes dead.
The failure mode that undoes all of it
Your receive feedline is also a receive antenna. If common-mode current picks up house noise on the shield and delivers it to the receiver, you have built an elaborate way to hear the same noise you started with.
Put a current choke at the loop feed point, and put a second one where the feedline enters the shack. Five to seven turns of the coax through an FT240-31 core at each end is enough for the low bands, and a clamp-on mix-31 bead is the fast field version. The full reasoning behind current chokes and why placement matters is in the RF grounding and counterpoise article.
Route the receive feedline away from your transmit feedline and away from house wiring. Not parallel, not bundled, not through the same wall penetration if you can avoid it. The routing techniques in the stealth coax article apply here, with one change: on a receive-only run, loss genuinely does not matter, so use whatever thin cable routes most invisibly and let the preamplifier make up the difference.
What I would build
Suburban lot, HOA, diffuse neighborhood noise with no single dominant source:
Loop on the ground, 55 feet of insulated wire in a square in the back corner of the yard, as far from the house as the lot allows. A 9:1 matching unit and a low-noise preamplifier at the feed point. RG-8X back to the house on its own path, choked at both ends with FT240-31 cores. Switched through a DX Engineering RTR-1A driven off the PTT line into whatever receive input the radio offers.
Apartment or condo, one dominant noise source you have already identified by bearing:
A YouLoop or a W6LVP loop on a small rotator base in the window furthest from the source, rotated for the null rather than for maximum signal. Choke the feedline. Accept that you are steering against one culprit and that a new neighbor with a new power supply resets the exercise.
Either way, measure before and after on the same signal, same band, same hour. Signal-to-noise, not S-meter reading. The S-meter will usually go down when you switch to the receive antenna, and the copy will get better. That is the entire point, and it is why this only makes sense once you understand that on receive you are not chasing signal at all.