Using the bladeRF with a directional coupler to analyze antenna return loss

I’ve written a basic script to analyze antenna VSWR based on measuring return loss using a directional coupler (ZEDC-15-2B) and plotting the results. The script is fairly obvious, and not ready for release, but attached are some plots of my HT antennas in the 70cm band.

Wiring diagram:

bladeRF TX port, 30 dB attenuator, Coupler OUT

bladeRF RX port, Coupler CPL

DUT, Coupler IN

The TX output seems to be around 15-20 dB more powerful than the RX port can handle undistorted, so with 15 dB coupling the RX port can still be overloaded slightly.

Using an attenuator on the TX port makes sense since the TX port will be isolated from the DUT, and it prevents open cables from overloading the receiver. No effective loss in sensitivity occurs for typical couplers since the directivity is usually around 30 dB or less, with -15 dB RX power in (assuming +15 dBFS from TX, 30 dB drop) and coupling of 15 dB the minimum signal level that will be received is around -60 dB, the noise floor is around -90, leaving plenty of dynamic range. For VSWR measurements anything beyond 20 dB return loss is just gravy anyway.

All the antennas were mounted directly onto the couplers SMA port output, as having pieces of coax between the measurement point and the DUT can cause various resonances due to the open or shorted coax acting like a stub filter.

We see that the Comet and Diamond stubby antennas are resonant right at the most used part of the european 70cm band, generally these antennas have performed well in my experience. 

The Icom stock antenna performs extremely poorly in this test, but I’ve found to work pretty well in general. It’s possible to design antennas that are not resonant but still work, especially when there’s no coax length between the PA and the antenna.

The Comet SMA-24 whip antenna performs similarly, this is another antenna that generally has worked very well for me. 

It should be noted that SWR is not a guarantee of performance, however antennas that have poor SWR and that are designed to work best when mounted on radios will most likely perform terribly when mounted on the end of a coax. Additionally I should consider adding a ground plane to the antennas being tested, as I’ve observed the Icom and the whip antenna perform differently when held, and the stubby antennas will shift their frequency up when I hold the coax shield during testing.

Tecsun PL-660 SSB BFO alignment

Alternate title: fixing the 1 kHz offset

Like a select few others I enjoy listening to shortwave radio at night when there’s a chance of picking up exotic stations, my only SW radio right now is the Tecsun PL-660, a very capable and affordable SW radio receiver.

Unfortunately in my copy, and I have heard reports from others about this as well: the SSB mode is offset by around 1 kHz. This is obvious when you switch from AM to SSB and immediately hear a loud 1 kHz carrier tone. Another way of telling is to tune 10 MHz in the Americas and 9996 kHz SSB in Europe. 9996 kHz is a Russian time-sync station which transmits in CW, meaning it outputs a carrier each second at exactly 9996 kHz. If you tune that frequency in USB or LSB mode, it should output a very low frequency beat or nothing at all. In the 660 it may output around 1 kHz beeps. This is wrong, for reasons that are beyond this post to explain. Properly tuned, you’d tune 9995 USB and hear a tone, or 9997 LSB and hear the exact same frequency tone. 

This is only a SSB problem, so I’m assuming the AM tuning is correct (mine seemed to be), AM tuning accuracy is less important anyway.

To tune this without using expensive and specialized equipment, you can use a CW time station like 9996 or 4996 kHz, the important thing is it needs to be a known frequency CW station. If you have a signal generator this can also be used. 

Basically, we’ll open the radio up and re-tune the SSB BFO frequency, then adjust the SSB centering, to check that we’re on frequency, you need to find a PC or mobile app with a spectrum analyzer that can tell you the frequency of the audio output from the radio. The reasons for this are clear when you know how SSB demodulation works.

First, open the radio, there are a few screws on the back and inside the battery compartment, the antenna can stay in place. The front cover with the wires for the speaker will come loose fairly easily. Just fold this to the side, you’ll need to hear the speaker output for tuning.

The adjustment points are under the LCD+button assembly, there are several screws holding this in place, and you’ll need to remove the tuning knob (just pull it off). Freeing the tuning encoder took some bending of the chassis for me, simply fold the whole assembly over. The adjustment points in the picture are now visible. Leave points 1, 5 and 6 alone unless you know what you’re doing. They don’t affect the SSB tuning.

I did this by tuning my reference CW station directly, switching to USB and adjusting point 2, center frequency adjust until the pitch of the tone got as close to 0 Hz as possible. That’s called zero beating, and is a way of getting very close to spot on frequency with basic equipment. After that, go 1 kHz down still in SSB, using the spectrum analyzer software check the frequency. If it’s not 1 kHz, adjust point 3 until it’s right on 1 kHz. To confirm tuning, go 2 kHz up (i.e. reference freq. +1) and switch to LSB. The audio frequency should be the same. You may need to go back and forth a few times to get this spot on. I didn’t adjust point 4 at all, but it seems to do something similar to point 3. 

As a final confirmation, tune an AM station in AM mode, make sure you’re on frequency (they’re almost always on multiples of 5 kHz), switching to SSB you shouldn’t hear any carrier whine. If you’re slightly off you’ll hear a strong humming in either LSB or USB mode, and you may want to consider redoing the adjustment. Obviously the BFO adjust wheel needs to be centered before doing any adjustment!!

To further check you may tune true SSB stations, I like to use 5505 kHz USB, which is an aeronautical information station. You should get natural pitch without adjusting the BFO wheel. Don’t use amateur stations to check unless you know the person transmitting has a calibrated rig because many stations are uncalibrated and may be several hundred Hz off.

Source for the adjustment image: http://www.radioscanner.ru/forum/topic42851-26.html 

Hints: I like to use SignalScope on iOS for this, it has a nice high res FFT and can put markers on signals of interest which is great for a proper tuning. VisualAnalyzer is a freeware Windows software that also works well.

Tooling is very basic, a small Philips screwdriver and a small flathead is all you need. If you have a piece of semi-professional test gear like a real frequency counter they often have a 10 MHz output  port on the back, if you connect a piece of wire to this it can be tuned on the Tecsun and you can use that as a reference which will give a nice strong reference signal. The FE-5680A rubidium reference is another great choice if you want true accuracy, but be warned: the Tecsun is unlikely to stay in perfect cal over temperature or time, it seems fairly stable to me, but if you start measuring this sort of thing you’ll need to know when to call it good enough.

Today we installed LD8NA, Narviks’ new D-Star repeater, the whole process took about 10 hours with three people involved. We installed power from the building nearby, mounted the housing inside the gondola receiving room, then installed the antenna and the electronics in the housing. 

We also found time for some coffee breaks.

The frequency is 434.775 DV -2 MHz split

Handheld antenna performance comparison

Following some discussions regarding the performance of the various antenna options available for VHF/UHF HTs I decided to use my spectrum analyzer to measure the performance of three types of antennas on TX and RX.

The test jig was as follows for RX testing:

  • Transmitter remotely keyed
  • Approximately 15 meters between TX and RX antenna
  • Fixed 1/4 wave on VHF, fixed 1/2 wave on UHF
  • 1W RF power (nominal)
  • Result read on spectrum analyzer
  • Antenna connected to spectrum analyzer, mounted vertically
  • Me standing in the same place about 1.5 meters away

And the following changes were made for TX testing:

  • Fixed mobile antenna for spectrum analyzer
  • Antenna under test replaced on the remote transmitter
  • Distance shortened to about 7.5 meters

Notes RX

RX performance on UHF was difficult to reliably measure as my position and posture could change the reading by up to 10 dB. 

Reference antenna was a Nagoya NA-771.

Some antennas also perform better when held by a person or provided with additional ground plane.

Notes TX

TX performance can be affected by the SWR of the antenna affecting the power limiter of the radio. This still indicated typical performance of the antenna on a radio.

Reference radio was a Puxing UV-973 in cross band mode.

Reference RX antenna was a Nagoya NL-R3 mobile antenna with a small ground plane.

Antennas tested (antenna type on VHF+UHF)

  • Nagoya NA-771 BNC 1/4+1/2 wave
  • Comet SMA-24 1/4+1/2 wave
  • Diamond CH-32 (~1” coil)
  • Diamond SRH-805S (~1” coil)
  • Icom FA-S270C (1/4+??, Icom standard HT antenna)
  • eBay brand spring VHF antenna
  • Celwave 2506 commercial helical VHF antenna (cuttable 1/4)

Of these antennas we would typically expect the NA-771 and the SMA-24 to perform very well, followed by the Icom and coil antennas and the stubby antennas performing significantly worse. The results were still a bit surprising, they are presented as dB relative to the first reference antenna.

VHF TX/RX performance

  • NA-771 +0/+0 dB
  • SMA-24 +0/+0 dB
  • CH-32 -20/-25 dB
  • SRH805S -12/-15 dB
  • FA-S270C -1/-3dB
  • Spring -1/-6 dB
  • 2506 helical 0/-3 dB

UHF TX/RX performance

  • NA-771 +0/+0 dB
  • SMA-24 +3/+1 dB
  • CH-32 -3/-3 dB
  • SRH805S -1/-3 dB
  • FA-S270C +3/-1dB

What we’re seeing is that the long whip antennas perform quite well on 2m TX/RX, the ultrashort stubby antennas might as well be left at home, and the Icom stock antenna + the helical commercial antenna both perform very well. Keep in mind the helical is a single band antenna as well.

On 70cm the SMA-24 performs very well but so does the Icom stock duck antenna, with only marginally worse performance I don’t see much of a reason to bring anything else with me if I’m going to be operating 70cm. Another surprise is how much better the stubby antennas are on 70cm, performance for the SRH805S is only 6dB down from the best antennas tested, one wonders why they claim this antenna works on 2m at all though.

As expected, in general the larger the antenna is the more signal it’s able to capture, and transmit performance is actually quite good even for quite short antennas like the stock ducky antenna from Icom. Personal experience and anecdotal evidence from testing suggests that larger antennas may be less sensitive to phasing issues, and may perform better while mobile because of this.


Time Teen


Time Teen

So I went for a drive today and found An Art in the middle of nowhere

Mar 8

This breaks form, but holy shit Friday night motherfucka by z0rg


This breaks form, but holy shit Friday night motherfucka by z0rg

My basic field kit for the multimeter
U1272A with two sets of crocodile clamps, insulated probes, SMT grabbers, low voltage crocodile clamps, thermocouple, fine pitch probes, RCA to BNC to multimeter adapter, and the LED lamp addon.

My basic field kit for the multimeter

U1272A with two sets of crocodile clamps, insulated probes, SMT grabbers, low voltage crocodile clamps, thermocouple, fine pitch probes, RCA to BNC to multimeter adapter, and the LED lamp addon.