RF efficiency: an interesting observation - Printable Version

+- WeatherDuino Forum (https://www.meteocercal.info/forum)
+-- Forum: Weather Stations - WeatherDuino Pro2 Line (https://www.meteocercal.info/forum/forum-4.html)
+--- Forum: General Talking (https://www.meteocercal.info/forum/forum-23.html)
+--- Thread: RF efficiency: an interesting observation (/thread-1151.html)

---

RF efficiency: an interesting observation - danner - 22-06-2017

Let me preface this by stating my other hobby is radio and designing antennas for that hobby is a sub-hobby within that hobby that I also pursue. 

Anyway, I've been monitoring 50MHz since I installed my WeatherDuino for any interference and fortunately enough I have not experienced any nor seen anything on my SDR. That's awesome. But, I started monitoring 144MHz and unfortunately I have located some interference that I have traced to my weather station and its transmissions. Not so awesome.

At this point I do not know precisely what is causing it. Maybe the 433MHz TX module produces some RF at 144MHz, I don't know yet. I have a 433MHz band pass filter with 50dB of attenuation at 144MHz on order coming from china. I will install that to the SMA connector of the TX board and see what that gets me.

In the meantime I tried some other basic things. Having 'some' experience with antennas a very common problem causing RFI, receiving funny noises, etc.,  is common mode current. I thought it might be possible the wiring coming off the TX board was acting as part of the antenna "system" and allowing me to receive something I shouldn't be receiving. I grabbed some UHF ferrite snap-on cores I had and put one on the UTP wire going to the temp sensor, another on the UTP going to the battery and a single larger one over both wires exiting the enclosure.

After installing the ferrite cores I did notice a slight decrease of the out of band noise I was picking up so I am on the right track. But there is usually a reason for common mode current. I am speculating now so take this with a grain of salt. It is quite possible the foil on the board is not a sufficient counterpoise to the rubber ducky antenna, thus causing the unbalanced RF to flow on the wires connected to the TX board.

The reason I say that is this. I've been semi-randomly taking readings of the battery voltage every few hours over the last week or so and recording that info on a spreadsheet, because it is an old battery and I'd like to know whether or not the thing needs to be replaced before this winter. Anyway, since the RF efficiency is available I've been recording that along with the temp and voltage. Since I placed the UHF ferrites over the wires exiting the TX board I have seen an interesting change in the RF efficiency readings. This is just a preliminary observation that I am sharing, I am not concluding anything. When I first started recording the battery voltage/temp/rf data I noticed the RF efficiency stabilized right around 74-ish%. And since last night my RF efficiency numbers are 95%+ and it has stayed around there for the last few hours that I have been monitoring it.

What does this mean and what do I do with this info? Well, it means there was common mode current flowing on the wires leaving the TX board, but it doesn't necessarily mean the 144MHz interference is a direct consequence of the common mode current. I will be experimenting with adding a tuned counterpoise wire(s)/radials/groundplane to the shield side of the rubber ducky. That should give the RF somewhere to go instead of on to the wiring leaving the TX board.

I am just sharing my observations and not drawing any firm conclusions yet, nor is this a criticism of anything or anyone. Nothing else has changed in the path between the RX and the TX. It is still a ~150ft path through one outside wall.

This is the typical RF efficiency I have been observing over the past week since I placed my WeatherDuino into service.



And this is what the RF efficiency looks like over the last 18 hours or so since installing the UHF ferrite snap-ons onto the wiring exiting my TX board enclosure.

---

RE: RF efficiency: an interesting observation - danner - 22-06-2017

Another screen grab. Interesting,no?

---

RE: RF efficiency: an interesting observation - danner - 22-06-2017

Initially I thought the TX board was rebooting itself and causing the inflated numbers, but there are no breaks in the recorded data to indicate that nor have I seen the RX complain it isn't receiving data.

---

RE: RF efficiency: an interesting observation - werk_ag - 23-06-2017

Very interesting indeed. Thank you for your studies in this matter.
This is a topic that I will be following with interest, and certainly it will catch the attention of many of us.

---

RE: RF efficiency: an interesting observation - danner - 25-06-2017

Don't have any updates to add. Still waiting on the band pass filter from china. I also ordered more UHF ferrite snap-on's.

RF efficiency has dropped slightly from ~95% to ~92%. This got me curious so I hooked up my RTL dongle and took a look around. I did not check the frequency when I installed this but it is currently transmitting on 433.765MHz. I don't know if it drifted down but I am hearing something else on the same frequency so I'll keep watching it to see how far it drifts. And it seems that despite me living in a semi-rural area on the edge of town there are still quite a few 433.92MHz devices around.

I played around with EZNEC a little today and modeled a vertical dipole using 24ga solid copper wire with insulation on, just messing around, simulating some Cat3 telephone wire I have. In case anyone is curious the impedance and reactance doesn't change much between 433.92 and 433.765 and if you were to measure the SWR you likely wouldn't be able to tell a difference.

Elevation plot of the vertical dipole 4ft above average ground (about the height of my enclosure). No sense in posting an azimuth plot since it will be a featureless circle.



For those curious about the specifics of the model, I measured the 24ga solid wire at .5mm in diameter and the insulation was .25mm thick. Using that and HD polyethylene as the dielectric (Diel C of 2.3) I got resonance with 155.5mm for each leg of the dipole at 433.92MHz. The resonant length will obviously vary with insulation thickness and type and conductor thickness.

Before I forget again. Here's what I did with the ferrite snap-ons.



And for what it's worth, it appears the rubber ducky has a small length of coax going up inside of it. That means the antenna does not begin at the connector and that adding anything at the connector would not be at the feed point of the antenna. It would amount to connecting something to the shield of the coax in the middle of the coax, and who knows what effect, if any, that would have. I did order some chassis mount connectors from ebay and when I have time I'll build a little ground plane antenna just to mess around a little.

---

RE: RF efficiency: an interesting observation - danner - 25-06-2017

I calibrated my RTL dongle to get a little more precise fix on its frequency and that was 433.788MHz. The signal coming from TX0 appears to be pretty strong so that's a good thing. You can see all the other services around 433.9-ish MHz and that was surprising considering I am not in town.

https://youtu.be/ITRzYgMpOS8

---

RE: RF efficiency: an interesting observation - danner - 25-06-2017

Back to the interfering signal on 144MHz coming from TX0 it looks like this.

---

RE: RF efficiency: an interesting observation - Palmyweather - 26-06-2017

Just as a side note: I see, Danner, with the photo you supplied showing how the ferrites are attached to the cables, there seems to be water inside the enclosure and reasonably sized holes top and bottom of the enclosure where the cable ties go around the outside support. 

Do you have any concerns that the moisture build up could cause problems for the circuit board in future (eg corrosion or a short circuit)?

---

RE: RF efficiency: an interesting observation - danner - 26-06-2017

(26-06-2017, 00:08)Palmyweather Wrote:  Just as a side note: I see, Danner, with the photo you supplied showing how the ferrites are attached to the cables, there seems to be water inside the enclosure and reasonably sized holes top and bottom of the enclosure where the cable ties go around the outside support. 

Do you have any concerns that the moisture build up could cause problems for the circuit board in future (eg corrosion or a short circuit)?

The water in the picture fell off the top of the enclosure when I opened it to take the picture. I do plan on doing something more permanent and sealing those holes. I've checked it a few times after rain and surprisingly no rain gets in through the holes.

As for corrosion I used a conformal coating to hopefully prevent or minimize any corrosion on the leads or anywhere else. I also used silicone dielectric grease on the pins for the connectors.

---

RE: RF efficiency: an interesting observation - werk_ag - 26-06-2017

(25-06-2017, 23:33)danner Wrote:  Back to the interfering signal on 144MHz coming from TX0 it looks like this.

Is this signal present only during the time the TX unit is sending data?

---