GOES 16 GRB downlink vs GVAR

[KWX]

So I assume you are getting constant stream a images there KWX?
As again some of the dropout issues are not yours as I saw a couple of bulletins of the GRB stream having some tech issue and some channel striping issues for a short time.
One of the benefits of having a direct readout system is any issues that effect the satellite effect everyone getting the data, as a result the fix is usually quick as the data loss effects the models and forecasts.
Also with this being the hi res data, downtime is not an option unlike LRIT/HRIT and EMWIN for the above reason.

Correct, I'm getting a constant stream. So far the key has been making sure I upgrade all the substandard components while I was testing the proof of concept. I was able to get signal up to 8.5dB on my 2.4m, by replacing some cables from the probes to the 90deg combiner with 6GHz rated RG402 semi-rigid cables, so will see how stable the signal gets specially during the middle of the day where fluctuations in levels are higher (to be expected given wx conditions, sun position, etc...).

Currently, I'm running both DD and TBS in tandem to see how each behave, so hopefully I can get them both to run without loss of lock for over 48hrs... assuming no tech issues occur that would require a manual reset (given the "cheap" setup lol).

I'm being also using the DD Frequency Offset report along with the Ayecka link info to get the optimal offset and symbol rates in an attempt to further stabilize the signal. So far a Freq. of 1686.594 - 1686.595 has worked good for me and a symbol rate of 8664. Not sure if my Ayecka is behaving differently than weather01089's, but I get a lock immediately with no loss of locks that I've noticed when I hook it up to the antenna, now, I do have to say that I've not upgraded it to support GSE fully so that could possibly be why, but so tempted to have it be upgraded. I guess, I have much higher inclination on waiting on the Novra 400, so likely I'll wait.

 

[KWX]

Do you guys foresee an issue if I were to try to build or get a septum feed for a 2.4m dish? lol I would be more inclined to build one than to buy one assuming there's a calculator out there for it.

 

[GOES 12 GVAR user]

CCLine card receiver? Do you mean the Digital Devices receiver?

Yep, that's what I mean. Sorry about that.

In the the case of the DD card why would they do that since that's not the standard for DVB Generic transport?
A guess if fine for an answer.

So.. does GRBStreamer strip off the BB Header as part of sending the Data field blocks to the end point?
My guess is yes.
Similar to the satellite TV method is what I'm thinking.

When you describe satellite TV, that fits with this stream as (Data field) is inside of the BB Frame.
As DVB standard calls for.
It not MPEG video though it's data, hence the term Generic stream.
Interesting, so assuming that most receivers that output BB Frames should do it as the DVB standard calls for.
The reason I'm asking is I have found other receivers out there that output BB Frames (some I have posted here on the forum) I want to be sure that this isn't mfgr preference methods. Though I have seen in some of the specs that the receivers send jumbo frames.
Plus I can visualize what's happening to the stream as it goes through the decoder.
Great! just another reason to lose a little more of my hair that's already thin in spots.

What's the thickness of the copper there Brett?

 

[GOES 12 GVAR user]

First, I'll answer your question KWX

Do you guys foresee an issue if I were to try to build or get a septum feed for a 2.4m dish? lol I would be more inclined to build one than to buy one assuming there's a calculator out there for it.

A 2.4 M dish is small for this method in my opinion. Due to the smaller size of the dish there is feed shadowing issues. This in return causes loss and other problems that will adversely affect the septum feed's operation. Plus the overall lack of gain due to the reflector size.
Though I am impressed with how well it's working with just the single circular polarization.
Yes there is a calculator, Brett posted one when we were discussing feeds about half way back from here.
The calculator calculates a square septum feed though as Brett has built.
Even though Brett is having issues getting GRB there in Cal. his septum feed seemed to work pretty good when he had it up.
I made one too but it was round and since I haven't got any farther with my setup, it's not on the front burner to look into why it's not working properly.

 

[GOES 12 GVAR user]

Great on the constant stream. As I said it makes a difference on the components used and performance achieved.
Good choice on the RG402 cable, that cable is designed for that use as is my LMR100A. Fact it's even better then the LMR100A
Again I wouldn't complain as getting a 8.5 dB on a 8' dish is doing fine. That's strong as signal I get here with a 10' dish.
Those receivers should be able to lock on the carrier automatically as they have a range that the carrier can be within.
Symbol rate is exactly 8665 as stated in the PUG, so that's what it should expect though as Brett's GRBStreamer 8666 works as well.
As weather01089 stated that his Ayecka has experimental firmware in it (if I remember correctly).
I assume that you get no GRB data from the Ayecka receiver, just signal info?
As I stated on here before that I am quite unhappy with Ayecka.
There is no guarantee that Novra will come out with a receiver that's going to work with this. All I can say that the guy I talked with seemed interested in the idea.

 

[N6BY]

...
In the the case of the DD card why would they do that since that's not the standard for DVB Generic transport?
A guess if fine for an answer.

Maybe they are somehow restricted by the hardware to sending 188 bytes at a time. Or maybe that's what they thought would be most useful? Both are guesses.

...
So.. does GRBStreamer strip off the BB Header as part of sending the Data field blocks to the end point?

No. The GRB software expects the BBHeaders as part of the stream it gets.

...
What's the thickness of the copper there Brett?

24 gauge. Hopefully I guessed correctly on the thickness.

 

[GOES 12 GVAR user]

Maybe they are somehow restricted by the hardware to sending 188 bytes at a time. Or maybe that's what they thought would be most useful?

That's possible

No. The GRB software expects the BBHeaders as part of the stream it gets.

My correction, I thought that the header would have been stripped.

24 gauge. Hopefully I guessed correctly on the thickness.

Mine is .015" in that area, Yours should be fine. You can cut it with a pair of tin shears. The type that look like overgrown scissors. Make sure you clean the cut edges so that they are not sharp and have no burrs.

 

[N6BY]

...
Mine is .015" in that area, Yours should be fine. You can cut it with a pair of tin shears. The type that look like overgrown scissors. Make sure you clean the cut edges so that they are not sharp and have no burrs.

The 24 gauge copper sheet arrived today. Its too thick for me to form into a cylinder, so I will have to return it and get 32 gauge.

 

[GOES 12 GVAR user]

Hold it, hold it.

You should be able to form that Brett.
It's .020" right?

If it is, you can form it. But you need a mandrel and some large hose clamps/worm clamps.
Stand by for more info to come...

 

[N6BY]

Yes, its .020". But its not flexible enough to form a smooth curve when bent.

It would work fine if I wanted to make a square septum feed. I could bend it along a table edge.

I already ordered the 32 gauge sheet.

 

[GOES 12 GVAR user]

If it's a soft copper you can form it. But you won't do it by hand, you will require a mandrel.
Here's what I used to make my feeds you see in the pictures.
20" x 10' Copper Roll Valley
This is .015" maybe a little thicker, this is flashing copper nothing too special about it. It forms well and solders fine.

A mandrel can be made several ways. If you are soldering the copper then you must use a mandrel that can take the heat so plastic is out of the question.
My mandrel is of solid oak, and ya it was expensive.
The wood shop turned it to the exact size I needed.

One of the problems with this mandrel is that it can be a bear to remove after the can has been made. Hence why I suggest the eye bolt and 2 people to remove.
Otherwise it works fine and makes a good working waveguide. Some final shaping may be required after mandrel removal which can be easily done by hand.

There is a cheaper option to make a mandrel that I found, it uses plywood disks and a threaded rod.

Most of us have one of those Rotozip or similar tools now and they can make quick wooden disks for this job.
The key part is making sure that when you drill the centers that they remain centered. With the above tool the hole is already there for the center so all one should need to do is drill it out to the final size of the thread rod.
Each disk should be spaced about 4" or so apart to keep a solid dia. of the waveguide. Put the clamps on or very near the disks so they can be tightened tight and won't effect the final dia of the waveguide.
Use nuts/washers on both sides of the wood disks so that they are rigid. Use a large enough threaded rod that it don't bend (or very little) when the mandrel is assembled.
One of the advantages of this mandrel is the wood disks can be removed much more easily after the can has been made. I recommend to drill about 3 holes in each disk so that when removing a vacuum don't form and make mandrel removal tough. You can even put some kind of hook or device to grab a hold of the wood disk to aid removal. I also recommend to remove each disk at a time.
The end is easily made by cutting out of some sheet a copper a disk that will fit inside the wavegide. I find that soldering this while the mandrel is in place makes it much easier. I also recommend that when putting this end on that, if you install it inside (as I do) the wavegide that you may make some "small fingers" of the flat to aid the soldering of the end cap.
The hose clamps I use can be found at home depot. They are large enough to go around the waveguide. They are stainless so no rust of anything like that is a problem.
Once the seam of the waveguide had been soldered move the clamps so that the unsoldered areas can be finished.
In this process you may need a flat bladed screwdriver to hold the seam and press it together for soldering.

In the case of the septum feed a part of the mandrel will need to be removed so the septum can be installed. Also septum will need to be added before the end cap is installed and this will be a process. It is something I have never tried before so some careful thought should be used in the process of construction.

 

[KWX]

Great on the constant stream. As I said it makes a difference on the components used and performance achieved.
Good choice on the RG402 cable, that cable is designed for that use as is my LMR100A. Fact it's even better then the LMR100A
Again I wouldn't complain as getting a 8.5 dB on a 8' dish is doing fine. That's strong as signal I get here with a 10' dish.
Those receivers should be able to lock on the carrier automatically as they have a range that the carrier can be within.
Symbol rate is exactly 8665 as stated in the PUG, so that's what it should expect though as Brett's GRBStreamer 8666 works as well.
As weather01089 stated that his Ayecka has experimental firmware in it (if I remember correctly).
I assume that you get no GRB data from the Ayecka receiver, just signal info?
As I stated on here before that I am quite unhappy with Ayecka.
There is no guarantee that Novra will come out with a receiver that's going to work with this. All I can say that the guy I talked with seemed interested in the idea.

Correct, I don't see any data flowing through the Ayecka, which further validates is not GSE capable without that custom firmware.

Hope Novra can get something cooking.

I was able to get my signal now up to ~9.8dB, so hopefully after a few more tweaks, I should be able to squeeze a 10dB out of the 8 footer.

 

[KWX]

In regards the DD and TBS, I found that if there's a signal degradation enough to cause BB/CADU sync errors, that once such degradation abates, the TBS is able to resume an error free streaming, but the DD also continues to run but with CADU sync errors only. For some reason, there are no BB errors, so not sure if it's the drivers or GRBStreamer is having a little issue recovering with the DD cards as compared to the TBS, but hey it could just be me . Of course resetting the coax "resets" the condition causing the CADU sync errors.

Definitely eager to see GOES 17 up and running so N6BY can see how his setup fairs out with his DD.

 

[N6BY]

In regards the DD and TBS, I found that if there's a signal degradation enough to cause BB/CADU sync errors, that once such degradation abates, the TBS is able to resume an error free streaming, but the DD also continues to run but with CADU sync errors only. For some reason, there are no BB errors, so not sure if it's the drivers or GRBStreamer is having a little issue recovering with the DD cards as compared to the TBS, but hey it could just be me . Of course resetting the coax "resets" the condition causing the CADU sync errors.

Definitely eager to see GOES 17 up and running so N6BY can see how his setup fairs out with his DD.

That's interesting that the DD card is able to resume with good BBFrames, but CADU sync errors. I had to write the code for the DD Card 'in the blind' because I didn't have a DD card back then. The DD code has to do additional processing to strip off the transport stream bytes that they add.

Anyway, right now I have no signal to test with (except for rare occasions in the middle of the night), so any attempted fixes would be guesswork. Am waiting for thinner copper sheeting to build a new septum feed.

It sounds like the TBS5925 is the better performer than the DD card, no?

 

[GOES 12 GVAR user]

Correct, I don't see any data flowing through the Ayecka

You do with the Ayecka as I do with my Novra 300D receiver, use it for signal info.
I'm hoping so for Novra as well, they make good hardware.

I was able to get my signal now up to ~9.8dB, so hopefully after a few more tweaks, I should be able to squeeze a 10dB out of the 8 footer

I am impressed that you are getting that much gain on an 8' dish. I am also thankful you are sharing that info.
With that info I now know how weather01089 is getting both streams on a 10' dish.
Due to the gain location differences his 10' dish equates to a 12' dish here and with that much signal you can get both streams.
Also with that info I can safely say that by division of the country that:
*The strongest signal is over the Eastern part of the US for GOES R (Goes East), as to be expected.
*It's 3 dB down here in the Central US
*It's likely 6 dB down over the West coast.
Meaning that, an 8' dish on the East coast = a 10" dish here for the same signal.
A 10' dish here = a 12' dish on the West coast for the same amount of signal.
That would certainly explain the issues that Brett is facing there with GOES East.
When GOES West comes online it will be backwards with it's antenna in terms of gain.
I again didn't think that the satellite's antenna was that directional, but it is.

 

[KWX]

That's interesting that the DD card is able to resume with good BBFrames, but CADU sync errors. I had to write the code for the DD Card 'in the blind' because I didn't have a DD card back then. The DD code has to do additional processing to strip off the transport stream bytes that they add.

Anyway, right now I have no signal to test with (except for rare occasions in the middle of the night), so any attempted fixes would be guesswork. Am waiting for thinner copper sheeting to build a new septum feed.

It sounds like the TBS5925 is the better performer than the DD card, no?

You sure did an excellent job writing that code blind for the DD.

Hope the new septum feed build won't represent much of a challenge to you.

In regards the TBS vs. DD current state, correct, the TBS5925 is offering better recovery for the time being, but do hope that it improves once you start testing it out.

 

[KWX]

You do with the Ayecka as I do with my Novra 300D receiver, use it for signal info.
I'm hoping so for Novra as well, they make good hardware.


I am impressed that you are getting that much gain on an 8' dish. I am also thankful you are sharing that info.
With that info I now know how weather01089 is getting both streams on a 10' dish.
Due to the gain location differences his 10' dish equates to a 12' dish here and with that much signal you can get both streams.
Also with that info I can safely say that by division of the country that:
*The strongest signal is over the Eastern part of the US for GOES R (Goes East), as to be expected.
*It's 3 dB down here in the Central US
*It's likely 6 dB down over the West coast.
Meaning that, an 8' dish on the East coast = a 10" dish here for the same signal.
A 10' dish here = a 12' dish on the West coast for the same amount of signal.
That would certainly explain the issues that Brett is facing there with GOES East.
When GOES West comes online it will be backwards with it's antenna in terms of gain.
I again didn't think that the satellite's antenna was that directional, but it is.

Just glad to provide any info I can, but interesting observations for sure.

 

[KWX]

Well, this is interesting. Looks like some of the issues I had with my DD randomly loosing signal at different times during a given 24hr period (day or night) could had been attributed to a DC block I've been using with it. Interesting enough the TBS doesn't complain about using a DC block with it but the DD appears to.

So far both TBS and DD running as expected, by the way, I'm splitting that single feed into 3, one for the TBS, one for the DD, and one for one of my SDRs for HRIT. Thankfully still getting 9.7 to 9.8dB on both TBS/DD with the 8.5 footer. I shall see in another 24hr to 48hrs things are still stable.

 

[N6BY]

Well, this is interesting. Looks like some of the issues I had with my DD randomly loosing signal at different times during a given 24hr period (day or night) could had been attributed to a DC block I've been using with it. Interesting enough the TBS doesn't complain about using a DC block with it but the DD appears to.

So far both TBS and DD running as expected, by the way, I'm splitting that single feed into 3, one for the TBS, one for the DD, and one for one of my SDRs for HRIT. Thankfully still getting 9.7 to 9.8dB on both TBS/DD with the 8.5 footer. I shall see in another 24hr to 48hrs things are still stable.

That's great that you diagnosed the DD card problem with the DC blocker. Did you try swapping it with another DC blocker or something else?

There is a line in the streamreader.cfg file to turn off LNB power. But I have not verified that it actually works.

I wish these DVB devices had a hardware switch to turn off DC power. It might be possible to cut the trace on the circuit board where the power is inserted into the line, and avoid the need for a DC blocker altogether.

In other news, I have been monitoring Real time satellite tracking for: GOES 17 every day, which shows the current location of GOES-17. At first its position was all over the globe. But the past few days it has been oscillating within a decreasingly narrow range of longitudes and is nearing its test position at 89.5 West.

 

[KWX]

That's great that you diagnosed the DD card problem with the DC blocker. Did you try swapping it with another DC blocker or something else?

There is a line in the streamreader.cfg file to turn off LNB power. But I have not verified that it actually works.

I wish these DVB devices had a hardware switch to turn off DC power. It might be possible to cut the trace on the circuit board where the power is inserted into the line, and avoid the need for a DC blocker altogether.

In other news, I have been monitoring Real time satellite tracking for: GOES 17 every day, which shows the current location of GOES-17. At first its position was all over the globe. But the past few days it has been oscillating within a decreasingly narrow range of longitudes and is nearing its test position at 89.5 West.

Had the same thought, but thought to first give it a while to see how it works out without it and then try another DC block to see if its defective somehow.

You know, I don't think the setting in the CFG file doesn't work. It won't turn off the power on the TBS receiver.

In regards GOES 17, is almost at 89.5. It should be there later tonight or tomorrow AM. Hopefully it'll be close enough to you by the time they start activating sensors, so you can start tweaking your setup up.