In our experiments, it is often necessary to lock something on the amplitude of a carrier / sideband. Sometimes it is necessary to monitor the presence of a signal, like the 10 MHz Frequency Standard. In both cases, this - remote controlled - Powermeter can do the job.

And yes, we don't use 40 GHz, but Linear Technology was so kind to send free samples. Sharp tongues would say, that I used this "proof of principle" to build a Radarduino :-)

The date printed on the frontpanel was way too optimistic. This project was released to public on the first HAM RADIO (Friedrichshafen) after Covid-19, 25.06.2022.

To make things easy, we split the design into a Display-Unit and several Detector Units. By that, the measurement plane can always be brought close to the source to be measured. And we have a 'Specialist' for every Measurement-Job. Below is the Block Diagram for the Basestation.



The Basestation mainly consists of the Arduino Nano Every. A DC/DC converter generates the +12V and the -12V needed for the Opamp on the Detectorboard. This is done by an R1D-1212, capable of delivering ± 42 mA

In Version 2 we added a Voltage Regulator to stabilise this +12 V down to +9 V, as the Voltage Regulator in the Detector (which is used as a Reference for the A/D Converter) seemed to have an insufficient Line-Regulation.

As there was some space left, we also put the venerable MAX232 with a DSUB connector on the Board. Even so, the preferred methode is a usb connection.

As the supply voltage is critical (due to the DC/DC converter), the Arduino monitors it.

On the Detectorboard, an Opamp amplifies the raw RSSI signal (which is produced by the often installed logarithmic amplifier) and makes it available at a BNC connector on the Basestation. It is also digitized by a LTC2485 (24-Bit ∆Σ ADC). The Pullup - resistors for the I2C lines are placed on the Detector side. By that, the Base has several ways to see, if a detctor is connected. First it can measure the raw RSSI level, second it can measure the voltage level on the SCL and SDA lines.







Really not much inside. The magic happens is in the Sensor. And in the Software.

We use a KFV 60 (DIN Audio / Video Connector, 6 Contacts, Jack, Panel Mount, Solder, Silver Plated Contacts) from Lumberg (Farnell Order Code: 1193069) and a SV60 (DIN Audio / Video Connector, IP40, 6 Contacts, Plug, Cable Mount, Solder, Silver Plated Contacts) from Lumberg (Farnell Order Code: 1321478).

PIN	SIGNAL	REMARK
1	SCL	Pull-up on Detector side
2	SDA	Pull-up on Detector side
3	RSSI	Detector ouput, amplified, Vu = 3
4	- 12 V	unstabilized, max. 40 mA
5	+ 9 V	stabilized, max. 40 mA
6	GND

Every Sensor has its own calibration data stored on an onboard Eeprom. This data contains :

ADDRESS	BYTES	CONTENT
0x00	2	Type of Sensor, e.g. 8307, unsigned int
0x02	2	Serialnumber, e.g. 0099, unsigned int
0x04	4	Reference Voltage on Sensor, e.g. 5.024 V, float
0x08	4	Minimum Frequency in MHz, e.g. 1.0, float
0x0C	4	Maximum Frequency in MHz, e.g. 500.0, float
0x10	4	Minimum Level in dBm, e.g. -70.0, float
0x14	4	Maximum Level in dBm, e.g. +10.0, float
0x18	4	Slope @ Frequency[0], float
0x1C	4	Intercept @ Frequency[0], float
0x20	4	Slope @ Frequency[1], float
0x24	4	Intercept @ Frequency[1], float
...	...	...
0x78	4	Slope @ Frequency[12], float
0x7C	4	Intercept @ Frequency[12], float

The intermediate frequencies are calculated by a fixed scheme. (Other schemes mayst be used as well). From the minimum frequency (0x08) and the maximum frequency (0x0C) the difference is calculated. The delta-frequency is a fraction of the difference frequency. We use the following steps :

10%    10%    10%    10%    10%    10%    10%    10%    8%    6%    4%    2%

With Fmin = 1.0 MHz, Fmax = 500.0 MHz, we get a Difference of 499.0 MHz. The calculated Slope and Intercept values therefore refer to the following frequencies :

F[0]	F[1]	F[2]	F[3]	F[4]	F[5]	F[6]	F[7]	F[8]	F[9]	F[10]	F[11]	F[12]
1	51	101	151	201	251	300	350	400	440	470	490	500

Calibration is done with a trustworthy Synthesiser and a Worksheet (Excel) or another Calculation Suite of your choice. See the examples in the Detector Section. For every Frequency, a pair of slope and intersection values is calculated. These values can then be beamed up to the Eeprom via a serial link.





The Calibration of new Sensors is like a 'Kindergeburtstag', using the menu offered :-)

The Frequency Coverage of 'some' Sensors, beeing in the budgetary Range

DETECTOR	FREQUENCY RANGE	LEVEL RANGE
AD 8307	9 kHz ... 500 MHz	-60 dBm ... +10 dBm

This is the workhorse of a lot of RF-Powermeters out there. So we cannot write much new things and therefore let a picture speak. The case is a Sucobox.

DETECTOR	FREQUENCY RANGE	LEVEL RANGE
LT 5537	10 MHz ... 1 GHz	-70 dBm ... 0 dBm

This Detector is almost identical. Just in green. We also replaced the AD8307 by an LT 5537. The case, again, is a Sucobox. Around the logarithmic Detector are some 0402.

DETECTOR	FREQUENCY RANGE	LEVEL RANGE
AD 8318	10 MHz ... 8.0 GHz	-60 dBm ... 0 dBm

High accuracy: ±1.0 dB over 55 dB range (f < 5.8 GHz)
Stability over temperature: ±0.5 dB

As the code is long - and the copy & paste procedure causes additional challenges, the file is provided as a download. Exactly here.

And that's the view, if a sensor is connected ...




Thanks to Andi and his Team, the view from behind is also aesthetically pleasing.

The webmaster does not read these comments regularely. Urgent questions should be send via email. Ads or links to completely uncorrelated things will be removed.