Ik vraag me af wat die brom is in deze waterval plot. Je hoort wel vaker wat voedingsgeluiden, maar wat je hieronder ziet is iets anders. De slingering is een langzame oscillatie met een frequentie van minder dan 1/2 Hz.



Update 4-maart-2016

Gisteren met Henk pa0hpv over deze plot gepraat op de clubavond van Veron afdeling 37. Er zijn verschillende mogelijkheden, of de SDR bemonstert op een dusdanige manier dat er aliasing onstaat, of het is een regelcircuit (ergens) wat een lage oscillatie frequentie heeft, PIDs of PLL’s kunnen dit gedrag vertonen. De eerste mogelijkheid kan ik uitsluiten, de tweede mogelijkheid wordt gokken.

Ondertussen bij andere repeaters gekeken, op PI2NOS zie ik het niet, en op PI2RTD zag ik het vanochtend ook niet overtuigend. Wel zag ik dat bepaalde transceivers een kleine beetje frequentie instabiliteit hebben. Mijn SDR heeft een TCXO en zou het probleem niet moeten hebben, als ik op DCF77 afstem en een minuut of wat wacht dan zie je de zender netjes stabiel op een referentie frequetie staan, vervolgens kan ik in de software een PPM correctie doen. Frequentiestabiliteit is aan mijn kant het probleem waarschijnlijk niet.

Alles wat asymmetrisch is in deze waterval plot is al verdacht, FM modulatie hoort een overwegend symmetrische beeld op te leveren, en deze slingering is duidelijk asymmetrisch, wat zou kunnen is dat de oscillator in de transceiver heen en weer aan het slingeren is. Je ziet het bij sommige transceivers als ze beginnen te zenden.

Hieronder is een voorbeeld van een oscillator in een transceiver die een paar seconde tijd nodig heeft om te stabiliseren, op ca 1/3 van de onderkant zie je alle spoortjes naar links bewegen, dit is het centrale gedeelte van het PI2RTD spectrum, de vertikale lijntjes komen door de CTCSS modulatie.




Seven element 70 cm Yagi antenna

I found the specifications of the Yagi on the internet (here) and decided to run a NEC-2 analysis on it, here are the results.

Screen Shot 2016-02-23 at 11.24.19
The voltages across the elements in the Yagi, the driven element has the circle in the middle, behind it is the reflector, in front are the directors
Screen Shot 2016-02-23 at 11.22.22
This is what the NEC-2 program needs, element 1 has the excitation turned on, the ground is turned off so all calculations assume free space


Screen Shot 2016-02-23 at 11.23.55
For the center of the 70 cm band the impedance is 45 Ohm, it is mostly real and the theoretical standing wave ratio is 1.2 for a 50 Ohm transmission cable.
Screen Shot 2016-02-23 at 11.22.52
This is the same plot but not in three dimensions, note that the Yagi has a dominant front lobe but also a number of side lobes, the front to rear and front to side lobe ratios are pretty impressive. The directionality seems to be better than +/-30 degrees.

The driven element is symmetric and needs to be balanced with a balun transformer when you use a 50 Ohm coaxial cable, there are several solutions for this and a possible design is on this webpage.

Last update: 23-feb-2016

Environmental data collected this week

Within our house I made a radio network of several sensors, they send their data to a gateway that logs the telemetry in a 4Mbit flash winbond IC. When I dump that log there is a matlab script converting it to plots.



Telemetry logged by the kitty program, it talks via the USB com port to a moteino that is always running, that is, the usb hub is always powered, so it runs 24/7.

This is what the temperatures did in and around the house. All infrared temperatures are in red, and blue temperatures are for the contact temperatures collected by NTCs. In the lower right you can seen how infrared and contact temperatures can differ, the reason is that a clear sky leads results in a few degree temperature drop on the IR sensor because it is looking at a cold stratosphere. Something similar is happening in the kitchen, upper right, there red may drop under blue because a kitchen window is open, the NTC will then see something warmer because it is inside a box.

Temperatures in and around the house

At some nodes I also measure the remote battery voltage, air pressure and light conditions. Battery voltages are not very interesting to watch, all nodes usually run for several weeks up to a half year depending on how the low power consumption of the sensor and the arduino interact. An example of a composist plot is this one:

Composite plot, right column has light condition and pressure

Last but not least, I also collect a groundwater level in the basement, because in Rotterdam you need a pump to keep the groundwater at a low level. I don’t read out the pump actions, but there is a contact point device with several rods and their resistances to ground give an impression of the water table. Reason for having it there is that I want to know whether the pump still works.

Pump actions plot, once every so many hours a pump action is fine, when there is rain it becomes more active, in the summer it typically goes to level 0 for several weeks to months, during the rest of the year you get to see this. You guessed it right that the second rod is not recorded, it is probably a loose wire from the sensor to one of the rods.


Although most people probably call it the D-word hobbyists don’t, because they don’t want to be associated with the D-word. I made one last summer, it takes a couple of days and gradually you get something you like, and what you’re able to control safely.


This is the DJI F450



PMU of the quadcopter



RC controller set-up


You do need some 5 or more amps to charge the batteries, so there is a dedicated charger

And finally, add the gimbal, what is a gimbal? It keeps the camera straight, stable, whatever. And you add of course controls on the transmitter to adjust the gimbal remotely. Add a video transmitter, and you’re done.

Next, wait for a nice day without too much wind (4 m/s is really a lot already, maybe 8 m/s is the limit) go in the field and enjoy, just browse the internet to find out what’s possible, and also, where it is allowed. Here are some of the videos that I made: