C-band 40.0E Belgium

[deeptho]

In case it is helpful to others:

Last week I receive a cheap C-band LNB from China. The price was just below the threshold for paying import
tax and VAT. As I want to use it on an offset dish (laminas 1200). I bought a seperate conical feedhorn + mounting bracket as well.




The mounting bracket did not fit, but after drilling a hole, I managed to fit it next to the KUband lnb. I would prefer the label to be on the bottom (rain shielding), but currently the bolt I used to attach the lnb is a bit long and prevents me from doing that.



In my first experiment, the LNB was actually in a different location (farther away from the centre). Not sure what it was pointing to, but the universal LNB was pointing
to 53E and so the c-band signal must have come from somewhere between 45 and 53 E: I received a signal almost immediately, as confirmed by my blindscan software (carrier lock, good snr ...) but it had no useful DVB and I received nothing else. I also did not make much of an effort.

On Saturday I decided to take it to the next level: I installed the LNB in its current position, and moved the dish until I got a signal on a satellite meter.
Pretty quickly I received some signals from 40.0E but not many and unfortunately, almost all were T2MI, so no picture yet.

Another problems was that the LNB came without a dielectric plate. I could not immediately find any useful material to make one, so I
went for an alternative solution, first proposed by doktorofsat (see elsewhere on this site). Specifically I used the modified design from
pride 21 as described here Dielectric plate ?

However, I used some unconventional materials and so I am proud to announce the first fully biodegradable c-band depolariser:


Translation: this depolariser will destroy itself in the coming months.
Basically, I printed "pride 21"'s pattern on a piece of paper, glued some aluminium foil on it and then wrapped the paper around a piece of card board. After inserting in the LNB, the SNR on one of the polarisations went up by 3dB and the signal went away on the other one. I believe this is a pretty good result and so the design is a good one. I will pobably have it made on a PCB later. I used to have some FeCl and some left over cupper plate, but I cannot find it, so I will probably order it somewhere.

With this setup, I could receive most of the T2Mi transponders on 40.E (but a few are missing), and even one regular DVB-S transponder, featuring RGVK Dagestan. SNR levels were pretty good. For example, here are a few results - note that the frequencies are wrong because my code thinks the LNB is for Ku-band.

1. freq=10908.108V Symrate=5119 Stream=0 pls_mode= 0: 0 SIG=-55.90dB SIG= 45% CNR=10.90dB CNR= 54% DVB-S2 8PSK FEC_3_4 INV_OFF PIL_OFF ROLL_25
2. freq=10914.772V Symrate=5129 Stream=0 pls_mode= 0: 0 SIG=-53.71dB SIG= 47% CNR=11.00dB CNR= 55% DVB-S2 8PSK FEC_3_4 INV_OFF PIL_OFF ROLL_25
3. freq=10921.441V Symrate=5119 Stream=0 pls_mode= 0: 0 SIG=-54.08dB SIG= 46% CNR=11.40dB CNR= 57% DVB-S2 8PSK FEC_3_4 INV_OFF PIL_OFF ROLL_25
4. freq=10928.116V Symrate=5129 Stream=0 pls_mode= 0: 0 SIG=-57.20dB SIG= 43% CNR=9.00dB CNR= 45% DVB-S2 8PSK FEC_3_4 INV_OFF PIL_OFF ROLL_25
5. freq=10958.111V Symrate=5129 Stream=0 pls_mode= 0: 0 SIG=-53.98dB SIG= 47% CNR=11.80dB CNR= 59% DVB-S2 8PSK FEC_3_4 INV_OFF PIL_OFF ROLL_25
6. freq=10971.443V Symrate=5119 Stream=0 pls_mode= 0: 0 SIG=-56.20dB SIG= 44% CNR=10.80dB CNR= 54% DVB-S2 8PSK FEC_3_4 INV_OFF PIL_OFF ROLL_25

The following plot hows part of the spectrum (obtained with my blindscan software. Not sure why some transponders are missing, but
I tried to receive them with a regular scan - no success either. I did not try very hard yet).


Still, I was a bit annoyed that this satellite is full of t2mi streams and I had not seen any of them. This was simply not acceptable. So just to get something, I tuned the transponder using a tuning program, recorded a transport stream using dvbsnoop and then used t2mi and vlc to view the result (note that the frequency in the example commandmay be for a different transponder than the screenshots):

./neumo-tune -c tune -a 2 -f 10914000 -pH -U 1 -S 5120 -mPSK_8 --delsys DVBS2
dvbsnoop -adapter 2 -s ts -tsraw -b > /tmp/test.ts
cat /tmp/test.ts |./t2-mi 4096 1 > /tmp/x.ts
The result is 4 programs: one which is breaking up all the time (corrupt stream, maybe a bug in t2-mi?), but some others are working:



So what is next? Well obviously I need to scan the arc sometime. Also I need to modify my bindscan code to actually use a C-LNB. Right now, it still thinks my lnb is a ku-band lnb. In fact this modification is done, but I need to test it.

Behind the scenes I have been working very hard to extend the blindscan code. By summer I hope to release it (but in my experience it always takes a lot more time than planned):

• save all scan results to a database
• show epg as a list and a grid
• allow to record programs in the epg
• allow to live view these programs
• control a positioner; detect which sat the dish is pointing you and "speak" the result
• capture spectrum plots
• deal with channel lists longer than 10 000 channels
• deal with complicated LNB/dish setups
• decrypt streams using oscams (assuming you have a valid local viewing card; only local oscam supported)

Of course, tomorrow is April 1st, so it could all be an elaborate joke. On the other hand, today is March 31, and why would I ruin a joke by announcing it? Perhaps I am not good at joking? Or perhaps it is fake news? Or perhaps not? Feel free to guess.

 

[ozumo]

The result is 4 programs: one which is breaking up all the time (corrupt stream, maybe a bug in t2-mi?), but some others are working:

The corrupt one will be the channel encrypted in Gostcrypt. Some receivers like the Edision Primo show the same 'corrupt' video.

 

[satesco]

The result is 4 programs: one which is breaking up all the time (corrupt stream, maybe a bug in t2-mi?), but some others are working:

It's great that you're trying to get a C-band,especially when it's the first time you do it.But,in my opinion,it is better try to adjust your antenna on Windows.With dvb applications for this OS(Crazyscan,EBS,etc)and tbs5927 USB you have every chance to set the antenna much more precisely and I think more quickly.For example,Crazyscan shows you in real time and exactly the signal,in dB,on each transponder.With a laptop on Windows and 5927 usb you have the advantage that you work near the antenna and you can check what you do without further confirmation from elsewhere,even from this forum.
Unfortunately Linux does not offer such flexible dvb applications as they are on Windows.For example,even the application(above) used by you does not indicate the snr of each transponders found.
But about receiving the signal with Linux and Windows would be worth a separate discussion,especially since you are trying to develop such a software for the first system.
I know my picture doesn't help you with the Express-AM7 reception, especially since I'm using another antenna size,but the antenna adjustment I did with tbs5927.I used this device for tuning many antennas.

 

[deeptho]

The corrupt one will be the channel encrypted in Gostcrypt. Some receivers like the Edision Primo show the same 'corrupt' video.

Thanks. That is interesting info.

 

[deeptho]

It's great that you're trying to get a C-band,especially when it's the first time you do it.But,in my opinion,it is better try to adjust your antenna on Windows.With dvb applications for this OS(Crazyscan,EBS,etc)and tbs5927 USB you have every chance to set the antenna much more precisely and I think more quickly.For example,Crazyscan shows you in real time and exactly the signal,in dB,on each transponder.With a laptop on Windows and 5927 usb you have the advantage that you work near the antenna and you can check what you do without further confirmation from elsewhere,even from this forum.
Unfortunately Linux does not offer such flexible dvb applications as they are on Windows.For example,even the application(above) used by you does not indicate the snr of each transponders found.
But about receiving the signal with Linux and Windows would be worth a separate discussion,especially since you are trying to develop such a software for the first system.
I know my picture doesn't help you with the Express-AM7 reception, especially since I'm using another antenna size,but the antenna adjustment I did with tbs5927.I used this device for tuning many antennas.

I have used crazyscan and ebspro. Nice programs, but neither supports my main dvbs card very well.

The graph is just a sreenshot of a a python plot. My code outputs the SNR, so it could be added easily. The main problem is to find a non-confusing layout for the text overlays (text tends to overlap with other text or with the graph).

The positioning was no problem as the dish was already aligned on Ku Band, so that means that only the orientation needs to be controlled.
To do that, I used the buttons on the dish. I have a good sat meter, which makes this easy. For fine-tuning I often use tvheadend, along
with a program that "speaks" the current SNR. It works well, but only if the service can remain tuned more or less (otherwise tvheadend sometimes gives up). My software will be able to do the same.

So this is why I am developing some software myself. It has a positioner console, which is still a bit buggy, mainly because I keep changing other code. This is work in progress, and the layout is not yet optimal (in part due to bugs in the widget toolkit), but this is what it looks like when it works. The SNR colours also indicate the margin for the specific modulation (red is the threshold for reception, yellow is a good margin and green anything on top).

The console allows blind-scanning and detects modulation and such. The code also monitors lnb frequency offsets. It is worrying how bad some lnbs are.

I am currently reworking on the si parsing, which is why network and ts id are currently listed at 0. It has worked..I have freesat, sky and other epg also working, and of course regular viewing, but there is much to do and there are still many bugs.

 

[deeptho]

I did some dish maintenance and discovered my C-band LNB was full of water. After "emptying" it, it no longer worked.
So I opened it up and it was damp inside. Also, some white fluid and powder covered the back plate and
some of the electronics. The same white deposit was also in the wave guide.
I managed to clean it up and let it dry. It seems to work again.

What could this deposit be?

How would water have entered the LNB? Through the antennas in the waveguide? The entry point looks a tiny bit rusty.

What is the best way to clean this? I used a tiny bit of aceton

I now made a few changes:

1. I drilled a hole at the bottom. Next time it rains, at least the water will have some way to exit the wave guide.
2. I now rotated the lnb so that the electronics and cables are above the wave guide (used to be at the bottom as I expected this to shield it a bit more from the rain)

 

[ozumo]

The same white deposit was also in the wave guide.
I managed to clean it up and let it dry. It seems to work again.

What could this deposit be?

Aluminium oxide? I had a car which would blow white dust from the vents when the air conditioning was switched on, turned out the aluminium fins of the evaporator were corroding, the corrosion looked similar to the white deposits in your photos.

 

[deeptho]

Aluminium oxide? I had a car which would blow white dust from the vents when the air conditioning was switched on, turned out the aluminium fins of the evaporator were corroding, the corrosion looked similar to the white deposits in your photos.

It is possible, but there was also a white deposite in the wave pipe. Not sure if there is any aluminium in there.