Thursday, May 18, 2017

SDR Based Test Equipment

The "$20 SDR Dongle" has made Software Defined Radio approachable by the masses, and has enabled enthusiasts to experiment with equipment that not that long ago would have cost them more than the car they drive.
One big cause of this is mass production - the DVB-T (Digital Video Broadcasting) technology caused mass production of receivers that you can use to receive digital TV on your laptop or desktop PC. With mass production comes low cost.

Well, it didn't take long for some experimenters to realize that with some software changes on the host computer, these mass produced devices could be re-purposed to become general purpose receivers that work between 25 MHz and 1.8 GHz.

Further mass production led to these "$20" devices being available for less than $10 on ebay.

The Advent of the Raspberry Pi single board computer has also democratized the personal computer market. A computer that would have cost thousands of dollars a few decades ago is now available for $35. Oh, and that $35 gets you much more power!

Putting the $35 computer and the $10 SDR Dongle together can get you very useful and good performance for the price.

The following was shown at Maker Faire in San Mateo by the Amateur Radio club I associate with (Bay-Net).
Firstly, keeping thing really simple - an SDR receiver and general purpose software (GQRX)

GQRX will display a chunk of spectrum on the screen, amplitude vs frequency, so you can see where the signals are there, click on them and hear what information they are carrying. To be honest, a lot of people use a program like this to listen to a few stations and put the SDR Dongle back in the draw. One reason is that the computing needed to use this software was that of a laptop or desktop PC. 
With the launch of the third generation, the Raspberry Pi 3 is finally capable of running GQRX, so the $10 SDR Dongle and a $35 Raspberry Pi are capable of making a spectrum analyzer that works between 25 MHz and 1.8 GHz. Not bad!

Next step is to add a signal source - this enables you to perform measurements that would have cost thousands of dollars not that long ago. Want to check that FM band reject filter you just made? Use your $10 SDR Dongle and a $20 Noise Source as a signal generator. 

Performance of the $30 solution (left) vs a $550 solution (Mini-VNA, right):

The dynamic range (difference between strongest and weakest signals, or how far down you can see) is not as good on the home-made solution, but it is only 5% of the price!

Want to check that the 2m tape-measure Yagi you built is cut to length? Add a directional coupler.

This time I compare the $50 solution (we added a directional coupler) with a piece of test equipment that cost $14000 (yes, that's fourteen thousand dollars!) - the Anritsu S412E :

When it comes to measuring a 2m j-pole antenna, it is difficult to tell the difference here, and yes, the $14000 device can do much more, when it comes to trimming and antenna you can get great performance for very little out-lay.

The details of how to do this are outlined in the excellent site (more specifically here), but I have gone one step further and scripted the calibration stage. This will be the subject of a future blog post.

Saturday, October 15, 2016


This is the material I presented at Pacificon 2016:


More to follow later this week.

Friday, May 27, 2016

"Golf for Engineers"

A few years ago, before she got licensed, I was trying to describe the use of Amateur Radio to my wife. Aside from the obvious emergency communications aspect, and a technical hobby, I pointed out the professional networking potential of it.

As an engineer in Silicon Valley, you both can't move without tripping over other engineers but also cannot just go out and meet them in a non work environment (unless your kids play together).

The traditional way to network with people was to play golf. Now I am not very good at golf. I keep getting stuck at the windmill (OK, my father in law came up with that line), but I have found over the years that having an Amateur Radio call sign does tend to open up doors and provide a conversation starter.

For instance at my previous employer I had my business cards include my call sign, and on the occasion where I was along on a visit with the sales guy as the token engineer the card exchange often resulted in the customer/client saying "Oh, my call sign is....". We suddenly had something else to talk about other than the buyer/seller relationship.

Anyway, that's how I came up with the "Amateur Radio is Golf for Engineers".

Think it will stick?

Saturday, January 23, 2016

HT Harmonic Output Testing

Originally posted in 2015, but presented here also.

There has been a lot of discussion recently on how bad the "Chinese Radios" are about suppressing harmonic output.

I had access to a spectrum analyzer as well as a few different HTs, so I figured I'd test a few.

The units I tested were (prices approx at time of writing)

Icom ID-51A+         ($360)
Yaesu FT-1DR         ($270)
Kenwood TH-F6A    ($320)
Wouxun KG-UV8D   ($110)
Wouxun KG-UV3D   ($99)
Baofeng UV-5R       ($30)

The relevant section of the Part 97 rules states:

§97.307 Emission standards.

(e) The mean power of any spurious emission from a station transmitter or external RF power amplifier transmitting on a frequency between 30- 225 MHz must be at least 60 dB below the mean power of the fundamental. For a transmitter having a mean power of 25 W or less, the mean power of any spurious emission supplied to the antenna transmission line must not exceed 25 µW and must be at least 40 dB below the mean power of the fundamental emission, but need not be reduced below the power of 10 µW. A transmitter built before April 15, 1977, or first marketed before January 1, 1978, is exempt from this requirement. 

For a handheld transmitter with an output of 5W (+37 dBm), the spurious emission must be below 25 µW (-16 dBm)

tl:dr ("Too Long, Didn't Read") 
  • The Big Three manufacturers pass well.
  • Wouxun do an admirable job for the price of the hardware. 
  • The Baofeng UV-5R is in violation of Part 97 and should not be used.

An external 30 dB of attenuation was added to the input port of the spectrum analyzer to prevent damage. The 30.0 dB Ext Loss on the plots below show that this offset was added to the measurements. The green line on the plots is at -16 dBm, which as stated is the FCC limit for spurious emissions.

Icom ID-51A+
2 m and 70 cm both look great, nothing notable above -30 dBm (1 µW)
ID-52A 2m
ID-51A 70 cm

Yaesu FT-1DR
2 m and 70 cm both look great, nothing notable above -30 dBm (1 µW)

Kenwood TH-F6A
2 m and 70 cm both look great, nothing notable above -30 dBm (1 µW)

Wouxun KG-UV8D
2 m and 70 cm both look great, nothing notable above -30 dBm (1 µW)

Wouxun KG-UV3D
2 m looks great, nothing noteable above -30 dBm (1 µW)
70 cm first harmonic beginning to creep up, but still 10 dB of magin. I'd call this good

Baofeng UV-5R
The popular $30 HT that is recommended by many to new licencees.

2 m has the first 3 harmonics failing quite drastically. Also has some sub-harmonics that whilst not failing, are not pretty.
70 cm has the first 2 harmonics very close to (if not over) the spec line

The radios from "The Big Three" (YaeComWood) all performed admirably, as you might expect. After all these radios do all cost over $300.

The Wouxun radios tested both performed pretty well - the new KG-UV8D was no different from the YaeComWood units on the plots taken. It's older cousin, the KG-UV3D had a visible harmonic, but still had plenty of magin.

The Boafeng UV-5R, well let's just say if you key up on 2m you are violating the Part 97 rules you promise to upkeep. On 70cm, the data I took was close enough to the limit line that more data should be taken. However, my 3 UV-5R units are destined for the e-waste bin.