LNB Offset Testing (sorry, no more free LNBs)

tylerhoot · 2019-05-25T20:36:41.445Z

othernet%20new%20lnb%205-25-19
does any one know if this is doing ok but this is with a cone on it and wondering why I am randomly dropping lock for like a sec at a time randomly don’t know what would be causing that?

wbrown · 2019-05-25T20:37:11.369Z

Ah! Read the specs. How novel (sarcastic jab at myself)

“Gain Variation 2dB” sounds suspiciously like what I’m after

Thanks Jim

tylerhoot · 2019-05-25T20:43:26.083Z

you know that the dreamcather 3 can support a micro usb Ethernet adapter if the WiFi cards are causing to much interferance but i just have to go digging thru the forums to find out what chipsets in the USB dongles are supported.

ac8dg · 2019-05-25T20:44:08.379Z

tylerhoot:

dropping lock for like a sec at a time randomly

watch if it occurs as the top line ‘Stream files’ changes to ‘Stream audio’ about once per second ?

ac8dg · 2019-05-25T20:47:16.995Z

edup wifi plug in is Chipset Ralink5370

tylerhoot · 2019-05-25T20:47:44.788Z

yep its dropping lock in the switch between files and audio but not always its more like every minute or few minutes but for the Ethernet one https://forums.othernet.is/t/dreamcatcher-ethernet/3660

kenbarbi · 2019-05-26T22:08:06.994Z

Now that you @Syed have some statistics about freq stability, is this new LNB something needed for Europe operations, or will a different satellite be able to work with the old Maverick? We have found the new Universal works fine in North America, all be it at a lower SNR than the Maverick.
Ken

ac8dg · 2019-05-26T22:59:27.963Z

kenbarbi:

old Maverick

This option to use a ‘universal’ is a great thing… My research shows the ‘old maverick’ is no longer in production.

kenbarbi · 2019-05-26T23:25:47.642Z

Yes. but I thought the reason to change LNBs to this new device was to provide freq stability. I wonder what the new European birds can handle/provide with respect to freq stability? Especially in Northern Africa, the Middle East, and Russia. Ken

ac8dg · 2019-05-27T00:10:50.625Z

I think you are right… initially thought the reports were that only PLL type would be stable enough for the narrow signals. My understanding is the ‘new lnbf’ are crystal based (non-pll) oscillators which in the ‘old days’ tended to drift and not be stable. The new one I am testing is quite stable… +/- 10kHz.

And I think price point is also important. I see universal lnbf on various sites for about $2.

wbrown · 2019-05-27T01:39:44.631Z

FWIW I did a baseline Maverick vs Maverick 1 hr startup from room temp (off 1 hr prior to test)

This plot shows the very initial startup and the large initial frequency offset

MM-early.png910×661 9.78 KB

And this is zoomed in for the close-in data for the rest of the hour:

mm-full.png910×661 35.5 KB

thomslik · 2019-05-27T14:04:26.435Z

I managed to swap LNB’s and get some numbers yesterday and today. As others have stated, I noticed the SNR was about 2dB lower than my old original Maverick with the Rssi about the same. I had to use my mesh cone/horn to receive a constant usable signal.

5/26/2019 about 1030 am EDT.
Ambient temp at start 88 F.
LNB case temp estimated 100+ F in direct sunlight.(Misplaced my infrared thermometer.)
First lock was < 1 minute from power up.
Initial Offset flashed high numbers 58702 a couple of flashes and settled down to around 1513 up and down.
After 5 minutes Offset was 218 - 1612.
After 1 hour ambient temp 90 F, case temp still 100+ felt very warm in full direct sunlight.
Offset 1562 -2108.
After 2 hours, ambient temp 91 F, Offset jumping around 3249, 670, 1215.
After 6 hours, ambient temp 86 F, Offset 769-1215.

5/27/2019 around 900 am EDT.
Again this morning, ambient temp 80 F, LNB case temp estimated around the same in shade.
Quick first lock again.
Initial Offset showed 59148 for 1 or two flashes, quickly dropped to 1016, 767, 471.
After five minutes, Offset 620-1265.
After 1 hour, ambient temp 84 F, LNB still mostly in shade, Offset 372 -719

Hope this info is of some use.

Tom
.

Syed · 2019-05-27T14:43:55.341Z

@kenbarbi The purpose of the LNB testing is two-fold.

Better frequency stability. This will be necessary in Europe as well as other new beams we start.
Replace the Maverick, as it is no longer in production.
Review any other quirks of the part.

The difference in SNR that you are seeing is still confusing. SNR is more important than frequency stability. I’ve tested about 20 of these, but have not seen the drop in SNR you and some others have noted.

wbrown · 2019-05-27T14:48:11.248Z

I did an initial test last night of the Universal, bare LNB, indoors. It was bad weather so lock was marginal for both Maverick and Universal. I thought a fringe-reception case may be useful to compare performance, so I let it run overnight.

To the data of specific interest, here is the Freq Offset graph for the first hour on the Universal LNB:

UNI-Startup-20190526.png1022×438 15.8 KB

And here is a document with more details, including all 12 hrs of data from last night, and comparison of lock rates, SNR, RSSI, bitrates. It’s in this PDF vs clogging up the thread with these graph images (and they are not the primary point of the thread)

Universal_Test_20190526.pdf (361.0 KB)

I haven’t measured temperatures yet. I am working on automating that, then I can start repeating fully automated tests with little effort, and collect the full suite of data requested.

Syed · 2019-05-27T14:50:04.410Z

This is definitely a PLL IC, but it does need some type of reference, which is the XTAL. The $2 LNBs that I have tested have a stability in the many hundreds of kHz, as opposed to the ten or twenty kHz of this one.

Syed · 2019-05-27T14:57:09.279Z

@wbrown Just looked at the PDF. Very nice work–thanks! I’m surprised to see that the Maverick had a stability over time that is much tighter than the Universal.

Based on everyone’s input, what I am seeing is normal variability of a consumer-grade product. We are safe as long as stability is better than 50 kHz.

ac8dg · 2019-05-27T15:50:31.556Z

Here is my observations on using the test lnbf without any cone/dish extensions.

When the SNR drops below -14.5 the frequency offset goes wild +/- 100,000 , then when the
SNR rises back above -13 it re-stabilizes at +/- 1500 Hz (two or three seconds )

Next experiment is the beam width of the bare lnb to keep the snr at -13. Initial pointing is a tolerance of 2 degrees … which is really the best ‘eye-ball’ accuracy … very tight requirement.

dschre · 2019-05-27T18:08:17.995Z

It has taken a couple of days to start testing the Universal LNB. Partly to blame are high winds over the past couple of days with gusts to 40 mph. I first tested the Universal on my 18 " DISH network antenna but had no signal at all. My newly purchased mounting bracket places the Universal about about 2-1/2" farther from the antenna than the original DISH LNB. Results after fiddling around for about an hour was a blank screen.

Then I mounted it outside with the horn constructed for the Maverick and matched the skew setting with what I use on the Maverick plus rotated to max in both directions. Briefly saw a SNR of about -16dB and a couple of packets but lock was for such a short time it wasn’t seend. Then back to a blank screen after further testing

I have been reading this thread carefully noting the tips about skew and the F-connector concern. Just to confirm my Tuner settings: For the Universal use the Americas satellite and Othernet Dual Band LNB settings. And, yes, I was sure to press apply, reboot and check that the settings had been retained.

So this morning I put the Maverick plus cone back to work outside and it performed as in the past - SNR around -10dB and RSSI -88.

I am baffled but not ready to give up yet. Will fabricate some sort of bracket for the DISH antenna and give that anther shot.

Richard

dschre · 2019-05-27T18:13:10.925Z

Meant to mention in last post that I continually checked the Diagnostics Log throughout testing and bottom 5 diagnostics were o.k.

wbrown · 2019-05-27T18:39:49.852Z

@Syed note in an earlier post where I plotted two mavericks at the same time. The other one had the sawtooth shape to the Fofs as seen in the Universal (I assume it’s being snap-corrected once it exceeds a certain threshold). For some reason the other Maverick seems to be a real golden unit.