Trying to understand message rates

Too good to hide it in the attic

It has been a while since I started this thread.
In the mean time I purchased the FA SAW filter, as suggested here in the thread

Updated the Airspy with the latest @prog and @wiedehopf

When I put my new dark blue FA SAW filter before my Uptronics filter the message rates drop around 200/sec.

The drops are visible in the graphs.

I do live near all kinds of cell towers and as suggested here that might have a huge influence on my message rates and positions. Perhaps I am doing something wrong, but if someone could explain or help me, it is highly appreciated.




at least it is available, Have seen many times where they are out of stock. A great filter, I might add

It is STILL in stock :wink:
Guess so but € 196… a little bit too much for me

You didn’t leave it long enough to see in the graphs properly.

Anyhow if that degrades your performance it means your coax is likely attenuating the signal a lot already or your antenna isn’t very good.
Because with a good antenna / good coax the FA filter (not a SAW filter, LC filter) should never affect reception negatively that much.

Oh and if you tighten SMA by hand you need to do i pretty forcefully.
With a wrench on the other hand it’s easy to overtighten.

Water ingress into the coax is also always a good candidate for issues such as this.
(gradual diminished performance)

^^^
There is an answer explaining the drop you might have missed.
Find feeders nearby and look/compare their performance.

Anyhow, what does your setup build like?
Antenna:
Coax:
Filter: FA filter
LNA: Uputronics
Coax:
Dongle: Airspy mini
RPi: 4B?

If your signal is a bit intermittent, check/replace the coax cable from the antenna.
As @wiedehopf already mentioned: have a little more patience when comparing message rate figures. AND water ingress in coax or LNA is quite often causing these kind of issues.

As I live in the Netherlands, too (Ridderkerk, near Rotterdam) it might be relevant to compare your performance fluctuations with mine.



On october 10, I replaced my antenna, filter and LNA.

Send me a PM and I’ll give you a live link to my feeder. If you live in the vicinity of Rotterdam you can compare our feeders live (i.e. with MM2 and Tar1090) by opening two browsers beside each other.

I can confirm that (been there last sunday :grimacing:)

Antenna and cable: https://shop.jetvision.de/epages/64807909.sf/en_GB/?ObjectPath=/Shops/64807909/Products/67121
Uptronics directly connected to Airspy
Pi 4, not overlocked

I have mobile towers within abour 250m away and and alarm tower within 150m.

I live in Zwolle, and there are some buildings near by, so comparing would be difficult.

My question was why ,without any changes in the setup, my message rate dropped when I put the FA Filter in front of the Uptronics but increased when I took it away.

In my humble opinion a message drop of around 200/sec is rather much. Around 1750/1800 ->1500 mgs/sec → 1750 mgs/sec.

Will test again for a longer period, have done that before and saw the same results. Message rates dropped while using the filter.

I do not want to be stubborn, I simply do not understand.

Guess did not use the correct words. Bandpass SMA filter Bought this filter

Did as you suggested

1 Like

Hey,

Thanks for the offer.

My setups are
All Pro stick Plus dongles.
FA 1090 filters on all sites.
3 x LMR 200 cables. 2 indoor antennas FA antenna. 1 Vinnant antenna 1090 S model
1 x LMR 400 cable Outdoor antenna, Vinnant 1090 S model

1x Pi 2
1 x Pi 3b
2x Orange pi 2GB ram, and 16 GB EMMC storage on board.

All setups use adaptive gain and are running version 6.1 of piaware packages.

Beside that I have a test site with an RTL-SDR generic dongle.

I don’t have LNA’ S installed since I use the Pro Stick Plus.

I have one better performer nearby since 6 weeks, I think hé is using an airspy. Other users are doing less, I’m living in the southeast of Amsterdam, some high rise buildings nearby and some cell towers within a kilometer range.

Weather has a distinct influence since GA from Hilversum and Lelystad drops a lot when the weather is bad

1 Like

The filter has a certain attenuation, so it makes the signal weaker.
If the signal is already pretty weak (what seems to be the case), then adding the filter will reduce reception.

For the overall issue, my best guess is that water got into the coax at the connection to the antenna.
That coax doesn’t even specifications listed on the website which is never a good sign.
It’s probably acceptable coax though (not great either or i’m sure they would be advertising the loss / attenuation at 1090 MHz).

You don’t seem to have graphs that cover the installation of the airspy and the start of the message rate drop.
That would make guessing much easier.
Especially seeing if you also lost range as well as message rate.

Not sure what you want to see

I do understand this, but I did not change any of this. I will check a better coax, but that does not change anything of what happened with the filter.

I think you need to test over a longer period of time. Message rate is too volatile to be a reliable indicator over a short period. Here’s an example of why from earlier today:

image

What’s going on here? A sudden spike of messages, followed by a dip and then a return to approximately the normal level.

The number of aircraft remained fairly constant:

image

It’s only when looking at the message types received that you get an idea of what’s happening:

image

A sudden sharp increase in the number of DF11 all-call replies implies that there was some change in ground interrogation that caused a lot more aircraft to start sending that message. Perhaps a previously offline radar came online, or a military SSR interrogator was activated briefly - here’s an article from Eurocontrol talking about this issue. Perhaps there was some brief atmospheric window that allowed aircraft to see radars from further away. Who knows? In any case, nothing changed with the receiver and the effect is entirely separate from receiver performance.

Note that DF17 and other messages remain pretty consistent throughout that period.

My point is that total message rate is not a reliable indicator unless taken in aggregate over a longer period.

It’s possible that the extra filter is introducing loss or noise sufficient to cause some weaker signals to be lost. I notice that your gain increased to 21 during that period - in my gain testing the other week, having the gain at 21 had a disproportionately large increase in noise floor compared to increases at lower gains. It’s possible that the extra attenuation from adding the filter is sufficient to lower the signal strength enough for the gain to increase, but the resulting increased noise actually results in worse performance.

Thank you for your answer…
Will use the filter again tomorrow and let it run for a couple of days.

My DF counts dropped as well

Would it be an idea to set the gain to 20 while using the filter in stead of gain= auto?

You can definitely try that - if you make the gain a constant, then you should see a difference in the noise and RSSI graphs if there is significant attenuation. Putting a filter inline shouldn’t make a huge difference, but if there’s a less than perfect connection somewhere that could increase the loss more than it should be.

That’s a pretty good suggestion. If i do not tighten the SMA connector properly to my Airsquitter connect, the GPS signal (which comes through the same connector) is getting weak over time even if it does not impact ADS-B reception

I think that factor is underestimated in many cases. Loose cabling is never a good idea.
That reminds me to the setup of a friend, he had ups and downs every few hours. We identified it goes better if it’s warm.
What was the result? A not properly mounted connect.

Looking at your current graph i’m not really sure what your issue is in the first place.
Goes up to 1800 :wink:
Do you need all of the 10 m?
A shorter, higher grade coax is likely in order if just adding the blue FA filter reduces the message rate that much.
I’m not sure how to better explain this.
If you have a good antenna / good coax you can usually add a bit of attenuation before the LNA and still maintain your reception.
If on the other hand a small amount (1.5 dB for the filter) of attenuation significantly changes your reception it usually means that this part of your system is suboptimal to start with.

This whole thread is based on the premise that you had better reception at some point.
I’m doubting that and asking for graphs showing the drop.
But that might not be possible if you don’t have graphs for that long.

If you want to upgrade your coax cable (which I’d recommend if you want better performance), H.O.Veenstra does make custom cables for a fair price.

Just add cable, connectors (be aware to order the SMA connector, not SMA-RP) and the service (twice) to the basket.

Highflexx 7 (7.3 mm), Belden H2007 (7.3mm), Aircell 7 (7.3 mm), Hyperflex 10 (10mm) and RF 400/UF (10mm) are great cables with low attenuation. 10mm cables have lower losses, but are more stuburn.

It is best practice to connect the LNA as close to the antenna as possible to reduce the negative effect of the attenuation of the cable. So, if it is possible to mount the LNA closer to the antenna, you could order a short cable (N-SMA) and use the 10m white coax between LNA and AirSpy. You’d also need a N (Female) to SMA (male) adapter/pigtail.

By the way: Why did you order a SMA/MCX pigtail? The Uputronics and Airspy are both equipped with SMA connectors, don’t they?

Yes. As a matter of facts, they are equipped with SMA female connectors.

From what I can see of the stenciling on the coax it appears to be this: https://files.elsist.biz/ftp/Various/Bieffe/CO100AFeng.pdf. It seems to have rather high attenuation at 1GHz, close to that of RG6.

One source of bad performance can be a “wrong” SMA connector as well. There are some connectors missing the middle pin. They are working as a connector but the reception would go to almost zero.

The Uputronics/Airspy can be connected directly with a connector in between. Doesn’t need to be a pigtail, but it makes the mount of the devices a bit easier.
I have both connected directly but using a short USB cable to connect it to the Raspberry. Otherwise the combination would be too long to put it in my box.

A torque wrench specifically for SMA connectors would solve that problem.

S.

1 Like