Revisit AD8307 Power Meter Calibration

This year at our club, annual pig picking BBQ and show-n-tell I had a chance to borrow the kind services of fellow Knightlites member OM John, KU4AF and his brand new and beautiful digital spectrum analyzer.

Connecting the Si570 based oscillator from my Sweeperino   into the Spectrum Analyzer and measuring the power level of the fundamental frequency we measured a -4dB difference in power when compared to measuring the squarewave on the AD8307 based wideband (~500MHz) Power Meter. Using a Low Pass Filter to filter out the odd numbered harmonics I resolved to measure the power in the fundamental on my work bench.

For the experiment I used the well known 10MHz calibration source as developed by K3NHI. This source can be used to establish the -10dBm point on the power meter through the measurement of a DC voltage level at the DC port. This reference is easy to build and is recommended for any homebrew starting out with a basic 50 Ohm bench.

Indeed my result showed a 4.1dB difference in power reading of the fundamental vs the wideband signal.

Herewith below are my Lab Notes of the procedure adopted.

RF Power Meter Calibration reworked

I reworked the calibration procedures again for the 1N34A based power meter and the 1N4148 based power meters. The full calibration results are included on this Blog page.

For the 1N34A I used a DC calibration procedure but this time I used my commercial attenuators to set the input DC Voltage levels. I then compared the measured voltages with the expected value. The correlation as can be seen from the table below is exactly the same. In other words the measured values = the expected values over the measurement range.

For the 1N34A meter I also injected RF power from my K2 rig into the RF power meter at FSD. I then inserted the attenuators using the same procedure as the DC readings. The Meter readings for the DC input and for the RF input at 7020KHz were exactly the same. This increases my level of confidence that the DC calibration method is valid, at least to within a single decimal place of a dBm measurement.

I also reworked the 1N4148 calibration. I reworked these due to the upgrade of the terminator to a newly built dummy load using 9X 2 Watt 470 Ohm resistors and 1X 1000 Ohm 2 Watt resistor (refer to the pics below). This allowed me to increase the calibration range up to 25 VDC. The voltage ranges used were set by varying the input DC Voltage from the variable voltage power supply unit.

Thus the RF Power measurement range now attainable from this power meter is summarized as follows:

1N34A
Vpk (V)  0.3 - 3
P (mW)   0.92 - 93
P (dBm) 0 - 19

1N4148
Vpk (V) 1.5 - 10
P (mW)  20 - 960
P (dBm) 13.5 - 29

Vpk (V) 1.8 - 15
P (mW)  30 - 2280
P (dBm)  15 - 33

Vpk (V) 2.91 - 25
P (mW) 80 - 6240
P (dBm) 19 - 38

Refer to the calibration curves below

1	Heading	1N34A RF Power Meter Calibration
2	Label	RF Power Meter Calibration
3	Date	12/23/2013.
4	Acknowledgements	1.      Author : ZS6RSH. 2.      Reference: EMRFD Section 7 paragraph 7.3.
5	Revision	Rev 2
6	Revision History	This is a repeat of the previous measurement but aimed at improved accuracy.
7	Scope	Calibrate the 1N34A RF Power meter using measured input DC voltages that are derived using calibrated commercial attenuators.
8	History	Reference 6 above
9	Configuration	Refer to Figure 2 below.
10	Test equipment specifications	1.      Keithley model voltmeter. 2.      Homebrew variable voltage current limited, power supply 2VDC-15VDC. 3.      3dB, 6dB, 10dB, 20dB attenuators. 4.      Connection leads.
11	DUT specifications	1. Homebrew RF Power meter using a 1N34A diode.
12	Workbench process	1.      Set the DC Voltage level and the variable pot to read 3VDC at FSD 2.      Insert the attenuators one at a time and record the DC input voltage and the Meter Reading.
13	Expected Results	1.    The recorded input voltages should be very close to the expected voltages when the attenuators are inserted.
14	Uncertainties	1. Keithly meter inaccuracies. No difference to 2 decimal points is seen between the Keithly and the Fluke voltmeters. 2. The RF meter may respond differently to RF input as opposed to DC input. This calibration method fundamentally assumes that at least at 7MHz, that there is no measurable difference between the DC and RF meter response. This was confirmed to 2 decimal places by measuring the same meter readings for RF power input against the expected attenuator values. 3. The commercial attenuators are assumed to be accurate. The theoretical voltage drops aligned with the measured voltage drops. Refer to the measurement table. 4. The termination load in the RF power meter is measured as 51.6 Ohms using the Keithly Ohmmeter
15	Preparation	Start with the DC voltage on the variable power supply set to minimum.
16	Perform validation measurements	Both DC and RF at 7020KHz were injected as separate measurements. Starting from a baseline of approximately 3VDC FSD and 3VAC FSD the meter readings were exactly the same between RF and DC for each inserted attenuator pad.
17	Perform the full measurement plan	Carried out as per 12 above down to approximately 0dBm
18	Observations	The results were consistent with the expectation.
19	Change Control	None
20	Computation	Calculate the RF Power for each meter reading as P = Vdc^2/2R. (where R = 50 Ohms)
21	Analysis	The plotted curve shows the transfer characteristic of the RF Power meter over the measurement range.
22	Conclusions	Using the calibration curve, the Power Meter will provide acceptable results. However this should ideally be confirmed against a calibrated RF signal generator in order to determine the accuracy with higher confidence. This procedure fundamentally assumes that a valid result can be obtained by using DC Voltage levels.
23	Documentation	Done as shown on the Blog.

1	Heading	1N4148 RF Power Meter Calibration
2	Label	RF Power Meter Calibration
3	Date	12/23/2013.
4	Acknowledgements	1.      Author : ZS6RSH. 2.      Reference: EMRFD Section 7 paragraph 7.3.
5	Revision	Rev 2
6	Revision History	This is a repeat of the previous measurement but aimed at improved accuracy. The 50 Ohm load termination was changed by using 2 Watt carbon resistors. This gives a 20 Watt power dissipation capability.
7	Scope	Calibrate the 1N4148 RF Power meter using measured input DC voltages. Calibrate the meter using 2 FSD values of 25VDC, 15VDC, 10VDC
8	History	Reference 6 above
9	Configuration	Refer to Figure 2 below.
10	Test equipment specifications	1.      Keithley model voltmeter. 2.      Homebrew variable voltage current limited, power supply 2VDC-15VDC. 3.      Connection leads.
11	DUT specifications	1. Homebrew RF Power meter using a 1N4148 diode and a 20 Watt 50 Ohm Terminator.
12	Workbench process	1.      Set the DC Voltage level and the variable pot to read 25VDC, 15VDC & 10VDC at FSD (Full Scale Deflection) 2.      Reduce the input DC voltage and take an input voltage reading for every 0.1 Ma meter reading as close as possible. Ie aim for 10 readings.
13	Expected Results	1.    The results should show a ‘fairly’ linear transfer characteristic.
14	Uncertainties	1. Keithly meter inaccuracies. No difference to 2 decimal points is seen between the Keithly and the Fluke voltmeters. 2. The RF meter may respond differently to RF input as opposed to DC input. This calibration method fundamentally assumes that at least at 7MHz, that there is no measurable difference between the DC and RF meter response. 3. Parallax error in reading the analog power meter 4. The termination load in the RF power meter is measured as 51.6 Ohms using the Keithly Ohmmeter 5. Termination load (dummy load) heating resulting in resistance variances.
15	Preparation	Start with the DC voltage on the variable power supply set to minimum.
16	Perform validation measurements	At the FSD levels the RF Power shown on the K2 Power Meter was the same as that derived from the RF Power Meter.
17	Perform the full measurement plan	Carried out as per 12 above between the levels shown below.
18	Observations	The results were consistent with the expectation.
19	Change Control	The 10VDC FSD was not possible to obtain with the variable pot and a 10K Ohm resistor in series. A ‘FSD’ level as shown in the table was used instead.
20	Computation	Calculate the RF Power for each meter reading as P = Vdc^2/2R. (where R = 50 Ohms)
21	Analysis	The plotted curve shows the transfer characteristic of the RF Power meter over the measurement range.
22	Conclusions	Using the calibration curve, the Power Meter will provide acceptable results. However this should ideally be confirmed against a calibrated RF signal generator in order to determine the accuracy with higher confidence. This procedure fundamentally assumes that a valid result can be obtained by using DC Voltage levels.
23	Documentation	Done as shown on the Blog.

Calibrating a 1N34A diode based Peak RF Power Meter

My homebrew RF power meter has 2 inputs. 1) using a 1N4148 peak detector and 2) using a 1N34A diode as the peak detector. This second port will read RF power from approximately 20dBm - 0dBm.

1	Heading	1N34A RF Power Meter Calibration
2	Label	RF Power Measurements
3	Date	11/27/2013.
4	Acknowledgements	1.      Author : ZS6RSH. 2.      Reference: EMRFD Section 7 paragraph 7.3.
5	Revision	Rev 1.
6	Revision History	RF Power calibration has been done for the 1N4148 version for power measurements up to +34dBm using a DC calibration method at 7020Khz only.
7	Scope	Calibrate a 1N34A peak power meter over the approximate range of +19dB maximum - 0dBm minimum. It is likely that the meter will not accurately perform below about 0 dBm. The calibration will be performed using calibrated attenuators and RF with the starting point being defined using a DC voltage near FSD.
8	History	This is a homebrew RF power meter built according to EMRFD Section 7 para 7.3
9	Configuration	Refer to the test schematic below.
10	Test equipment specifications	1.      Keithley model 8024B voltmeter. 20V scale. 2.      Homebrew variable voltage current limited, power supply. 1.5V – 15VDC. 3.      Connection leads. Regular leads that came with the voltmeter. 4.      Jumper leads used for the power connections. 5.      K2 QRP rig 6.      Coax connection from rig/attenuator output to power meter input 7.      Calibrated attenuators 3dB, 6dB, 10dB, 20dB. VSWR better than 1.5:1 from DC to 1GHz. Ref http://www.picotech.com/oscilloscope-accessories.html#TA050
11	DUT specifications	1.      Homebrew RF power meter using a 0.1mA FSD meter.
12	Workbench process	1.      Connect up the power supply and the Voltmeter set to read 3 volts 2.      Increase the voltage to 3volts. 3.      Quickly check that the meter is reading 3V at FSD 4.      If not then adjust the internal Pot until this reading is obtained. 5.      Set up the K2 connected into the Power meter. 6.      Adjust the power output of the K2 to read about 90mW 7.      Press the tune button and observe the Power meter reading. 8.      Make sure the reading is not greater than FSD. 9.      Record the reading on the ammeter. Being quick to do this reading. 10.   Insert the 3dB pad at the K2 RF output end of the coax connector (see diagram below) 11.   Quickly record the meter reading 12.   Repeat the above for the 6dB, 10dB and 20dB pads. 13.   Connect the 20dB pad in with the 3,6,10dB to try to get readings at 23,26,30dB levels. Not attainable.
13	Expected Results	Expect to get linear readings down to 0dBm.
14	Uncertainties	1.      Variation in K2 power output level from one reading to the next. May be necessary to repeat the readings three times to check this. 2.      Variation in actual performance of the Power meter at RF (test frequency= 7020KHz) vs at DC. 3.      The whole procedure assumes that the attenuation pads are correctly calibrated. The spec is as above in section 10.7. 4.      I used a DC voltage to obtain a reading at 0.9mA on the power meter. There could be an error between that reading and the actual RF power output from the K2.
15	Preparation	Completed
16	Perform validation measurements	Validation focused on setting the K2 to read as close to 90mW as possible (ie FSD on the meter). Once the 3VDC FSD calibration had been performed I then connected the K2. Since it is a digital power setting, the closest I was able to set the rig to FSD was a meter reading of 0.9mA. I then reconnected the DC power source and adjusted it to obtain 0.9mA meter reading. This setting was at 2.84VDC. Using the formula P =Vpk^2/2R I calculated the RF power output as 81mW. This was thus the starting point for the measurements.
17	Perform the full measurement plan	Performed as planned except for the validation/starting point measurements as above.  0dB, 3dB, 6dB, 10dB. 20dB attenuators were used.
18	Observations	The lowest practical reading was with the 20dB pad in circuit. This measurement just moved the meter to 0.04mA which is -0.92dBm (see table below)
19	Change Control	Refer above to the validation measurements change to allow for the fact that the K2 RF output could not be precisely set to 90mW FSD.
20	Computation	Refer to the table below. The Meter readings were recorded against each attenuation  pad in circuit. Thus it is assumed that the Pad is correctly calibrated.
21	Analysis	The graph shows a ‘knee’ around 0.25mA which would be as expected for this diode. More data points would be needed between the 10dB and 20dB marks in order to get a more accurate curve below 0.25mA
22	Conclusions	The calibration was completed according to the plan, however a better calibration would be possible with a calibrated RF signal generator.
23	Documentation	Completed.