QCX Mini Measurements

 

This is a follow-up on my previous post on building the QCX Mini

Measurements

As always, click on the pictures to see them in a better resolution.

I don’t have a professional lab, so don’t expect a full report à la “Rob Sherwood” (see http://www.sherweng.com/table.html ) … maybe someone can send Rob a QCX Mini to have it tested.

But I do have a good Spectrum Analyzer with an interesting function to measure harmonics.

First I measured the total attenuation of my attenuators, a 30 dB 1kW Bird attenuator, followed by a 15 dB Microwave Modules attenuator.  Total attenuation on 14 and 18 MHz, including all cables, was 46.60 dB (I will neglect the 0.02 dB difference between 14 and 18 MHz). This was entered in the Spectrum analyzer as a reference offset.

With a maximum power of 6W (to be safe), expected input to the Spectrum Analyzer (SA) or Power Meter (PM) was -8.82  dBm or less. A safe level for both the SA and PM that I use.

 

Interlude 3: a cheap but accurate Power Meter

If you have nothing to measure power, you can of course use the method that Hans uses in one of his videos, using an oscilloscope to measure the voltage over a dummy load. Then get out your old TI-30 calculator and start typing … and hope that you don’t make any errors.

Or … you can look up this little beauty  https://nl.aliexpress.com/item/32990010852.html (if the link doesn’t work, search for “RF Power Meter -75~16 dBm 1-600MHz” on Ali, Banggood, eBay, etc.)

I have this cheap Chinese RF Wattmeter, but the name is misleading TWICE, since it only measures in dBm, not in Watt or milliWatt. I built mine into a mint tin, with two small Li-Ion cells (from cordless telephone origin) UNDER the piece of copper clad. As you can see, I put a FUSE in line with the two cells.

 

And the “cheap Chinese” might imply that it is a piece of junk, but wrong again!
Actually it is quite accurate!

The circuit uses the Analog Devices AD 8307 chip, and when I compare it to my HP Generator, the deviation is never more than 1 dB. Very good for that low price.

Maximum input is +16 dBm, so make sure to use sufficient attenuators. QCX output can easily be more than 5W, depending on band and supply voltage. 10W would be +40 dBm, so a minimum attenuation of 30 dB should be enough for QRP work.

The value of your total attenuation can be entered into the wattmeter, but only in 1 dB steps, so I didn’t use that here.

Now back to the measurements.

Vcc was 12V (minus the voltage drop over the idiot diode D33).

Test at room temperature of 20°C.

After tuning L4 and L3 for maximum power output (see build report), I obtained around 5.6 Watt on 17m, and 4.0 Watt  on 20m. I made a table of the power versus supply voltage (I will redo this from 7V up later if I find some time).

Vcc IN(V)	Voltage after D33 (V)  (QCX DVM)	Pout 18 MHz (W)	Pout 14 MHz (W)
11	10.41	4.5	3.1
12	11.39	5.6	4.0
13	12.36	7.0	4.7
14	13.34	8.0	5.5

 

PA Temperature

I’m not sure what happens  if I use the QCX Mini on 20m, while it was really built for 17m use.

I get less output power, but maybe more heat is dissipated in the three BS170’s ?

So I set up an experiment, measuring the temperature at the washer on top of the finals, with an infrared thermometer. I measured after 30 seconds and after 60 seconds of continuous key down.

Vcc = 12V.

18 MHz: Pout = 5.6 W, supply current = 810 mA
 

Time (s)	Temperature (°C)
0	22
30	26
60	29

 14 MHz: Pout = 4 W, supply current = 510 mA

Time (s)	Temperature (°C)
0	22
30	22.8
60	24.0

 So the good news is: the output power may be lower, but so is the current consumption and the heat production, so nothing to worry about.

Harmonics!

This is what you have been waiting for, right ?

Ok, let’s see. The 17m rig has a 17m Low Pass Filter, that should be OK.
Ftx = 18.080 MHz, Vcc = 12V.

First screenshot below is a regular spectrum, with two markers.
Marker 1 is set for the main carrier, and the unit was set in  Watt.
Marker 2 was set on the second harmonic, and reading the difference from Marker 1 in dB.

Below is a screenshot from a special mode in this SA, the “Harmonic Analysis”. It is normally a paid option, but for the first 128 hours, I can use all options to try them out … nice!

Just one push of a button, and you get an automatic report about all the harmonics.

Harmonic suppression is about 60 dB or more below the carrier, so very good!

Then the same was done for 20m, again a regular spectrum, and then the Harmonic Analysis screen.

Oh my … the harmonic suppression is only some 39 dB here ... so is it legal to use the rig on 20m then?

I checked the Belgian regulation and even asked a friend who works at the Belgian PTT (BIPT).
The result of all my investigations is that there really is NO standard for homebrew rigs, as long as you don’t build them to be sold on the market. I don’t know for other countries of course …

It comes to this: if you have a HAREC license (I have!), you can build anything you like, and tell the PTT that you “suspect” it conforms to the standards (for commercial rigs).
As long as there are no complaints about interference from your rig, you can use it.

Of course, harmonic suppression is one thing, but in combination with an antenna, the harmonics may further be reduced, especially when using a tuned antenna like a magnetic loop, or a monoband 20m groundplane, which will not radiate well on 10m (high impedance).

To remove any doubt, adding an external 3-element LPF section for 20m would probably reduce this 2nd harmonic enough to bring it within specs (but also reducing output  power even more). Something to try when I find some time ...

Now to be sure, I did a few more tests.

First I wanted to know if my Icom IC-7300 was meeting the specs, it is a commercial rig, sold “on the market”, after all.

Running at 100W on 14 MHz, I measured the second harmonic, and here the screenshots as before.

It is clear that the 2^nd harmonic is very low, but here the third harmonic is the strongest. It is within spec though.

For a real life test, I asked a friend who lives 1 km from my house, to listen for the harmonics of both the IC-7300 and the QCX Mini, and was reassured when he couldn’t hear any harmonic from either rig.

RX Sensitivity

 I don’t have the means to make exact sensitivity measurements versus SINAD audio quality, but I wanted to see how much effect the input BPF had on reception on the 20m band, compared to that on 17m.

I have heard China and Thailand on 17m, so I’m sure the receiver is pretty sensitive!

I used my RF generator to generate a signal on 17m, and put it at a level to get a scope reading of 4V peak-to-peak, at the best tuning point, in the middle of the CW filter.

I then switched to 20m, and increased the generator level for the same audio output on my scope. 

Surprise, surprise … I had to increase the level only a mere 5 dB, so you will lose almost one S-point in receive on 20m. I guess it won’t work for the weakest of signals, but there should be plenty of strong stations to have a QSO with! 

Have fun with your QCX Mini!

[EDIT : I made another post on how to use the QCX Mini on two bands! Read that story here]

73,

Luc - ON7DQ