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.

Checking the calibration of my squarewave & sinewave calibrators and AD8307 rf power meter

At our recent QRP club show-n-tell activity I was fortunate to be able to check the calibration of my home brew instruments. For this exercise an HP Spectrum Analyzer (model unknown) belonging to Gary, N3GO, was used to check my 10MHz, -10dBm squarewave calibrator and my  11MHz -10dBm, sinewave calibrator (see a previous post for details of these two units).

 I have also been lent (on a longterm loan by Chris KD4PBJ) an HP Signal generator HP8657B. Herewith are the results of the calibration check of my homebrew AD8307 rf power meter.

As can be seen below. The results are not consistent.

10MHz, -10dBm squarewave calibrator

Fundamental -12.56dBm  (54.9uW)
Third             -23.05dBm  (5uW)
Fifth              -27.76dBm  (1.68uW)
Second          -39.9dBm
Fourth           -40.13dBm

Total =  61.78uW  This is equivalent to -12.09dBm

Thus this shows a reading that is 2.09dB lower than the nominal 'calibrated' value of -10dBm. The -10dBm level was calibrated using a DC calibration technique.

Next we measured the sinewave generator

11MHz, -10dBm sinewave calibrator

Fundamental -8.77dBm
Second          -39.69dBm
Third             -48.53dBm
Fourth           -58.34dBm

Thus this shows a reading that is 1.23dB higher that the nominal 'calibrated' value of -10dBm.

The tables below show measurement comparisons between the power meter and the scope readings. This result shows good correlation. It is reasonable to assume that the HP8657B signal generator is accurate to tolerances significantly in excess of the power meter or the scope. This shows a maximum deviation of 0.53dB. This is much better than the comparisons above!

Next I will ask Gary if we can run the tests once again. Check the HP8657B against Gary's Spectrum Analyzer.

Power Meter calibration using the Weinschel step & fixed attenuators

Now that I had established a few calibration points I wanted to establish further reference points. These would assist in determining the variances in linearity over the range -10dBm through -70dBm.

On hand I had a step attenuator with the following markings:

MANUFACTURER: Weinschel Engineering, Gaithersburg, MD.
MODEL: 9332
SERIAL: 441 0955-0085.
STEPS: 0dB, 10dB, 20dB, 30dB, 40dB, 50dB

I compared the Weinschel with my fixed attenuators using the power meter as the comparator.

The following results were obtained @11MHz.

dB(marked)         Weinschel DVM (mV)         Fixed Attenuator (mV)        Difference (mV)
0                             1565                                      1566                                  1
10                           1361                                      1368                                  7
20                           1167                                      1166                                  1
30                           962                                         959                                   3

The above shows excellent fit between the fixed and step attenuators with a maximum difference between the 10dB pads of 0.5%.

I now derived a result for the 40dB & 50dB attenuators based on a nominal slope of 20mV/dB.

dB          Weinschel DVM (mV)     Slope     dB derived              
0                1565                                -             -
40              761                              20            40.2
50              565                              20            50

This gave me confidence that the Weinschel was indeed an accurate attenuator .

By combining my fixed attenuators with the Weinschel settings I was able to measure reference points from -10dBm through -70dBm in steps of 0dB, 3dB, 6dB, 10dB. Refer to the table as shown below.

Referring to the linear equations on W7ZOI's essential notes. http://w7zoi.net/Power%20meter%20updates.pdf 

For two reference points P1(dBm) & P2(dBm)  having respective DVM voltage measurements V1(mV) and V2(mV)

The linear equation has the following form  P(dBm) = A + B*V(DVM)

Where B = (P1-P2)/(V1-V2), A= P1-((P1-P2)/(V1-V2))*V1

Using an excel spreadsheet I developed an equation for each decade of the slope. This can be used to obtain the most accurate possible power meter reading should it be so desired. It is interesting to examine the table below. The dB/mV slopes are indeed 'slightly' different for each decade.

The equations below are tedious to plug into a calculator, however they do yield the most accurate result. Alternatively the last column slope numbers can be used.

   

Calibration of the AD8307 Power Meter using fixed attenuators

I was now in a position to calibrate some additional points on the Power meter having completed the calibration of the 11MHz sine wave generator to the -10dBm reference point.

On hand I have the following attenuators which were purchased new from a commercial source. These attenuators do not have a specified tolerance, however they are marked to be used upto 1GHz. For this purpose I assume they are accurate and will form the baseline for calibrating the system. The attenuator values are:
3dB, 6dB, 10dB, 20dB,

The attenuators now allowed me to derive the slope in mV/dB at each measurement point. This was an encouraging result shown in the following table.

Pad(dB)           dBm        DVM(mV)      mV/dB
0                      -10            1565               ----
3                      -13            1507              19.33
6                      -16            1449              19.33
10                    -20            1367              19.80
20                    -30            1166              19.95
30                    -40              958              20.23
39                    -49              775              20.26

This shows a slope variation over the measured range of 20.26-19.33 = 0.93mV  which equates to 0.93/20 = 0.05dB. Not sure this is a meaningful calculation?

Looking at it a different way.  Midpoint over the range -10dBm to -49dBm
= (20.26-19.33)/2 + 19.33 = 19.80

Thus at the bottom of the range @ -49dBm and using 19.8 we derive (1565-775)/19.8 = 39.9dB
ie. at the reference point of -49.9dBm the power meter shows an error of 0.9dB. Acceptable?

Herewith Lab notes showing the calculations.

AD8307 Power Meter. The Analog Meter Calibration

In this procedure the aim was to adjust the current to voltage converter that drives the analog meter so that the meter would read about 75% of full scale deflection (FSD) at the maximum rated input of the power meter which is +15dBm. 

Since the meter has a FSD of 1mA the 75% point would require a drive current of 750uA. I started out using 2 X 2.2K resistors which resulted in a reading of 517uA for +14dBm. 

Subsequent calculations and tests (shown in the lab notes below) resulted in the use of 2 X 1.5K resistors which gave a reading of 835uA at +14dBm. The input voltage to convert to 835uA with this resistor of 3K was 2.51volts. This was deemed to be satisfactory and would give enough safety to ensure that the meter would not be damaged if excess input is applied in error. 

For these measurements I used my K2 Elecraft as the RF source since that is what I had. I had previously calibrated my 1N4148 based RF power meter using DC voltages. Thus this was the starting point. Using this meter I set the K2 power output to 1 watt (+30dBm). I then inserted fixed attenuators (16dB) to bring the level down to +14dBm. This level was then used to set the meter deflection.

Calibrating the slope of the AD8307 Power meter

The first task after building the Power meter was to adjust the Calibration output (Port B) pot to as close as possible to the specified 20mV/dB.

The iterative procedure followed is shown in the lab notes below. At this stage in the calibration the slope was set to 19.95mv/dB.