Thermistor RF Power Meter

Legendary  Thermal RF Power Meter

2019 10/31

A peak power meter, actually a peak voltage detector could be easily made. Diode power meter gain popularity over decades and everywhere.

 You  had to draw your own meter scales. The low power detection is harder for diode, at it's square root area. QRPer fancy the linear scales, lower power detected ability, and accuracy.  More than the linear, QRP and accuracy, the true RMS RF power meter,  which had magical attraction.

HP432A  is thermal sensor, which determine the RF power by measuring the thermal energy.  HP478A use  thermistor, and more specifically, use the NTC (negative temperature coefficient) resistor.  

Thermistor sensor have the best linearity, suitable for calibration lab, they are sensitive to exceeding power, so, they need well protected. 

Thermistor sensor bridge

While RF energy is applied to the thermistor,  its heated by RF power, it's resistance — changes. The feedback loop adjust the  DC power level delivered to the bridge,  make it balanced again.  The changed DC power is linear ratio to microwave power applied to the thermistor, therefore giving you an accurate power measurement.  This relationship is called DC substitution.

In most of today’s thermistor-based sensors, there is a second thermistor to detect and compensate for ambient temperature changes. 

Thermistor mounts are the only present-day sensors which allow power substitution measurement techniques, and thus retain importance for traceability and absolute reference to national standards and DC voltages.  

HP 432A use op amplifier do the math,  output  current  is proportional to RF power applied, this current then drive the needle meter.

PTC Thermistor sensor bridge

wa1mba  presented a extremely good article about the HP478A thermistor sensor.  Obviously,  home building such a sensor is challenge. 

Most activate  ham radio activity might be HF and low VHF range. If a thermistor sensor could built , satisfy QRPer,  it  would be a awesome device.

The HP478a use NTC  as bridge, the good mechanical manufacture make sure it's working well up to 10 Ghz, and the tiny teeny little thermistor is almost hardly to get,  it could be customized.   

But I'm look for a PTC sensor- a filament lamp devices sensor for a long time. There is few several article talked about it. A power meter presented on Radcom 1985/12 , attracts my attention,  the Accuracy HF Power Meter.  

This power meter achieved 1.8Mhz to 50Mhz, 5% accuracy on 50Mhz and  better than %3 at HF,  Power range 400mW.  This whole device is almost same complex as HP432.

Then i found another small piece of lamp power meter, which is much simple, and it is not calculating the power use ops. It utilized the  voltage which is linear proportional to DC power, it's fascinating.

I suspect this could not working up to 500 Mhz,  but even working in 50Mhz, it's still such a awesome device, capable response to 10uW way up to 300 mW.  I decided make one of this and  enjoy it's full capability

Filament Sensor Capabilities

I don't have exactly same bulb, obviously. And my filament bulb have almost 100 ohm resistance around 4V.  I use one Bulb and one 100 Ohm resistance to built that bridge. And not construct the range amplifier.

The most mysterious about this sensor is  the linearity.  The  original data in the schematic indicated that the voltage change for every 100mW is  114 mV no matter how much  the RF power is. From 10 uW to 10 mW, keep the same 'linear' factor.

My own testing result indicate the linearity  could be better!

Filament bulb Frequency Response