The ARS QRP Lab Reviews the DSW-20 Transceiver
The ARS Sojourner
You can find basic information for the DSW-20 on the Small Wonders Lab web site, http://www.smallwonderlabs.com, and we won't repeat it here.
Our tests of the DSW-20 were performed without a case. The board was in a very quiet RF environment, and we have assumed that the lack of a case did not affect the test results.
Feeling a little tangled up with knotty technical questions? Many of our graphics have this portrait of a fellow electronics adventurer. Just click, and you're on your way to helpful background material.
Transmitter Tests
Power Output
Rated output for the DSW-20 is 2.5 watts. Out sample produced 2 watts with a 13.8 volt power supply and 1.3 watts with a 12 volt power supply.
Power Requirements on Transmit
With a 13.8 volt power supply, our DSW-20 drew 330 mA with 2.0 watts of RF output. With a 12 volt power supply, our DSW-20 drew 285 mA with 1.3 watts of RF output.
Spectral Purity
FCC regulations require the spurious emissions from a 2.5 watt radio to be at least 30 dB below the carrier. Our sample easily met this requirement.
We performed these tests with a Tektronix TDS 380 scope with Fast Fourier Transform capability. Follow this link for an explanation of the FFT approach, as well as some help on reading the annotations on the graphs.
Here is the link to the spectral purity graph for our sample.
Introduction to Receiver Tests
Many of our receiver test reports use bar charts that compare the unit being reviewed to a number of other HF transceivers. Please read this brief explanation of the purpose and layout of these charts.
Receiver Tests—When No External Signals are Present
Spurious signals
There were three birdies in our sample, at 14.002, 14.045 and 14.046 MHz. They were all near the minimum discernible signal and would probably be masked by typical 20 meter noise levels.
Power Requirements on Receive
Our sample drew 49 mA on receive with a 13.8 volt power supply, and the same amount with a 12 volt power supply.
Receiver Tests—When One External Signal is Present
Minimum Discernible Signal
On 14 MHz we measured an MDS of -134 dBm. Please follow this link for comparisons with the MDS of other HF radios.
Phase Noise
We were unable to make a definitive measurement of the DSW-20's phase noise, because gain compression occurred before we were able to detect phase noise. In general, DDS systems such as the one used by the DSW-20, are known for good phase noise performance.
If we had detected phase noise at the point at which gain compression occurred, the phase noise, 10 kHz from the carrier, would have been -131 dBc/Hz. So we are at least able to conclude that the DSW-20's phase noise 10 kHz from the carrier is better than -131 dBc/Hz.
Please follow this link for comparisons with the phase noise of other HF radios.
Spurs
This section of our report falls in the realm of "no free lunch." While the phase noise of a DDS system is usually good, the system has a major weakness. DDS generates both quantization noise and spurs. The spurs occur throughout the entire frequency range passed by the DDS's low pass filter.
The implication of having spurs in the receiver local oscillator can be profound. Even a spur that is 60 dB down from the carrier will mix with an incoming signal when the difference between the signal and the spur equals the IF. If the signal is very strong, the "extra" received signal may be above the noise, adding to the QRM.
We injected a -30 dBm signal in our DSW-20 at 14.030 MHz. We found spurs on both sides of the carrier. The carrier dominated the middle part of our test range, but we found about 10 spurs in the range of 14.000 to 14.020 MHz and another 10 spurs in the range of 14.040 to 14.060 MHz. Some of the spurs were close to the minimum discernible signal and probably would be masked by external noise. Others were quite strong.
IF Rejection
Our sample had an IF rejection of 79 dB. Please follow this link for comparisons with the IF rejection of other HF radios.
Image Rejection
Our sample had an image rejection of 58 dB. Please follow this link for comparisons with the image rejection of other HF radios.
Audio Output
Using the standard ARRL test tone of -70 dBm, we measured total harmonic distortion of 4.5 percent. The tone in our headphones was painfully loud.
Receiver Tests—When Multiple Signals are Present
Selectivity
We measured a 6 dB bandwidth of 670 Hz. You will find interesting information about the width and shape of the DSW-20's IF and RF response in this AF spectrum analyzer graph.
Blocking Dynamic Range
Our DSW-20 had a blocking dynamic range of 103 dB. Please follow this link for comparisons with the blocking dynamic range of other HF radios.
Third Order IMD Dynamic Range
In one of the more important tests of our review, we measured a worst case third order IMD dynamic range of 83 dB. Please follow this link for comparisons with the third order IMD dynamic range of other HF radios.
Third Order IMD Intercept Point
We calculated a worst case third order intercept point for our DSW-20 of -9.5. Please follow this link for comparisons with the third order intercept points of other HF radios.
Second Order IMD Dynamic Range
Although this test was ignored in earlier times, it is now receiving increasing attention. Our measurement of second order dynamic range for the DSW-20 could not be completed, because it was noise limited at 101 dB. Here is the link for comparisons with the second order IMD dynamic range of other HF radios.
Second Order IMD Intercept Point
Because our measurement of second order IMD dynamic range was noise limited, we did not calculate the second order intercept point.
In-band IMD
Another test that hasn't received much attention. But we have a hunch it will be important for the low power community, because it may shed light on the superhet/direct conversion trade offs, and because it may help us quantify the poor audio we see in some simple transceivers.
In the DSW-20, the test signals generated a rich array of strong IMD products. Here is the link to an AF spectrum analyzer graph. For purposes of comparison, here are links to the AF spectrum analyzer graphs for two much more expensive and complex radios, the Yaesu FT-1000 MP and the Elecraft K-2.
Conclusion
The DSW-20 is most notable for its clever use of firmware and DDS technology. It is also a fascinating case study in the tradeoffs that every radio designer must address.
This transceiver gets high marks for small size, feather weight, low power consumption, reasonable price, operating conveniences, and, in an uncomplicated operating environment, good performance. But like other simple radios, especially those using the NE602A (superseded by the SA612AN) as a first mixer, the DSW -20 is allergic to strong signals. Big signals can cause big problems, with IMD products, DDS spurs, desensitization, leakage into the IF, and interference from signals on the image frequencies. When the KWs appear, it might be a good opportunity to mow the lawn.
Overall, Dave Benson, the talented and amiable designer of the DSW-20, has dealt with design tradeoffs very impressively. It's hard to see how more performance could be squeezed out of a $95 radio.
Here are some important links:
The lab's goals and equipment.
