Digital display for QRP labs 20W dummy load – part 1

The QRP labs 50-ohm 20W QRP HF Dummy Load is an inexpensive kit for a low power dummy load.

The load comprises 20 x 1W resistors, time will tell what its continuous power rating is actually.

This article explores a possible design for a digital display of power using the provided pads for a half wave detector.

Dummy load

Above is a ReturnLoss plot from 1-100MHz, ReturnLoss is good below 60MHz, very good below 30MHz.

Above, the same thing presented as VSWR.

A 10W continuous wave was supplied to the dummy load to observe temperature rise.

Above, after just a few minutes the test was terminated as the PCB was reaching glass transition temperature (ie where the PCB material softens). Ambient temperature was 10°, so temperature rise was 115° and increasing.

The load was cooled and tested at 5W.

Above, again after just a few minutes at 5W temperature has risen over 50° to 63°, sufficient to burn skin.

The 20W power rating is pretty meaningless, there is no mention that it is not a continuous power rating in the instructions, continuous power rating must be less than 5W.

This prompts me to consider a temperature sensor and alarm in the digital display.

Paul Danzer’s review in QST Sep 2019 rated the load at 20W. It does make one wonder about ARRL product reviews!

Half wave detector

This build replaces the supplied 1N4004 with a 1N5711 or BAT46.

Above a plot from the kit documentation showing the expected relationship between power and DVM reading using the supplied 1N4004 diode, and the blue dots are measured response with a BAT46 diode.

1N5711

The DC voltage from the detector needs to be reduced to a range of 0-1V to suit the ADC input of the intended microcontroller.

Above, a candidate circuit was modelled in LTSPICE and the response plotted. The circuit provides a sufficiently low DC source impedance for the ADC process and to absorb expected ADC pin leakage. The PIV rating of the diode is 70V, equivalent to 49W in 50Ω.

Above is a plot of the modelled response from 2.5mW to 25W, and a third order polynomial curve fit to those points. The fit is quite good over a 30dB range.

BAT46

The DC voltage from the detector needs to be reduced to a range of 0-1V to suit the ADC input of the intended microcontroller.

Above, a candidate circuit was modelled in LTSPICE and the response plotted. The circuit provides a sufficiently low DC source impedance for the ADC process and to absorb expected ADC pin leakage. The PIV rating of the diode is 100V, equivalent to 100W in 50Ω.

Above is a plot of the modelled response from 2.5mW to 25W, and a second order polynomial curve fit to those points. The fit is quite good over a 30dB range.

Above is a second order polynomial fit with forced zero intercept compared to the previous case. The error is very small, and it solves the problem of displaying non-zero power at zero Volts from the detector.

The better choice is to use the full polynomial and force zero display when ADC input is below 2mV (ADC count < 2).

if(ADC<2)
  P=0
else
  P=0.0003191+0.3014*V+34.76*V**2

A SOD-323 BAT46 will probably be the final choice, it has better performance and can mount on the underside of the PCB where it can easily be more easily than a through hole diode as supplied.

Digital

The display will probably use either some variant of an Arduino Zero or Arduino Nano and possibly a small OLED display.

Above is an end to end test of a BAT46 prototype for function testing using an Arduino Nano and I2C LCD display. This prototype is well within 5% accurate based solely on the LTSPICE model, assuming no error in the voltage divider, tracking well from 1W to 20W. When calibrated for the voltage divider and ADC Vref error, power displayed was within 2% of a proven power meter at several spots from 0.8W to 25W, on a spot check it is within 10% (0.4dB) at 1mW.

Next

Next step is to build and measure a prototype.

The calibration coefficients will be stored in EEPROM so that multiple instances can use the same code, just the EEPROM image will be different for each (as needs be).

A work in progress… Digital display for QRP labs 20W dummy load – part 2