Freeview Interference HD v SD

delback

Member
I'm getting occasional pixelation on Freeview HD channels. Most of the time the signal is fine - these are just brief moments of breakup. The signal strength is good (I live 4 miles from the transmitter). Freeview SD is rock solid - no problems at all.

Some questions for the experts here:
1. Given that the breakup is isolated brief moments, do you agree this is probably interference (eg, from passing traffic or other environmental factors) rather than just poor reception? Is Freeview HD more vulnerable to interference than Freeview SD?
2. I don't have an at800 filter. My HD broadcasts are on channel 29 (538.2MHz). Is this low enough that an at800 filter isn't going to help?
3. The aerial cable is RG6. Would changing it to CT100 or WF100 reduce susceptibility to interference?
4. Could it be that the signal is too strong? Would HD be less tolerant than SD of an overly strong signal?
5. My HD tuner is an Astrometa DVB-T2 USB stick. Are there better USB DVB-T2 tuners around that I should consider?
 

Clem_Dye

Distinguished Member
1. I don’t think so,
2. Not sure, but I don’t think it relevant here.
3. Probably not, but *100 cables are the norm.
4. What readings do you get from the in-built metering?
5. Not familiar with this kit. Could be software. Look at something like the Manhattan T3 box.
 

Rodders53

Distinguished Member
1. Probably unlikely.
DVB-T2 / HD in UK is 256QAM, 32k carriers, 2/3 Forward error correction
DVB-T / SD is 64QAM, 8k carriers, 2/3 FEC
So there is a difference in the two, but they are broadly received equally well.
BUT each RF frequency travels and is diffracted of ridges and reflected off hills and buildings differently en route from transmitter to aerial and may have a bigger effect than those transmission parameter differences.

2. Unlikely to help, but it will reduce signal slightly. AT800 would have sent postcard to you if likely to be affected Not sure if you received a postcard | Search a postcode | at800

3. Unlikely. RG6, while horrible stuff (most likely copper clad steel core and completely unsuited to outdoor use - steel rusts!) is well screened and has low losses at UHF if the copper cladding is intact.

4. Might be overload.
Depends on your location, transmitter erp, aerial gain (plus its location roof, loft, room etc.,.), plus any amplification and splitting in use.
I don't believe -T or -T2 is affected differently by overload - it's signal distortion = uncorrectable errors.

5. Such tuners are close to sources of RF interference from the computer/display/wifi card etc.,. Even the PSU feeding it may carry it in.
Unlikely another will be significantly different - but may be worth operating with a USB extension to see if that makes a difference?
What signal strength / quality does the tuner report on each and every multiplex frequency? That info might help with diagnosis.
 

delback

Member
What signal strength / quality does the tuner report on each and every multiplex frequency? That info might help with diagnosis.
Thanks for your comments. Measuring the signal strength & quality is tricky, because this is a MythTV system running on a RPi4, and it has limited feedback in those areas. But I think it does show something...

The DVB-T2 tuner (Panasonic MN88473 driver) doesn't report any signal strength - perhaps the driver doesn't support whatever API is used for that. The SNR hovers around the 60% mark on all channels (22, 23, 25, 26, 28 & 29).

The DVB-T tuner (Afatech AF9033 driver) reports between 70 and 90% signal strength, and between 96 and 100% SNR.

Also, these two tuners are fed from the aerial via a passive splitter. I used a cheap signal strength meter on the splitter and found that one side was down compared to the other. Interestingly, it was the stronger side that was feeding the DVB-T2 tuner, and when I swapped them round so that the DVB-T2 tuner had the weaker signal, it wasn't able to get lock. So that suggests the problem isn't due to overload.

I'm beginning to think that the DVB-T2 tuner just isn't good enough. It is just a cheap Chinese thing I got from eBay. (This one). Perhaps I should try something like the Hauppauge WinTV-SoloHD.
 

Rodders53

Distinguished Member
SPLITTER!!!! and two separate tuners!!
Do drip feed it makes my life fun.

Pic of the splitter and cables will confirm my suspicions...
 

delback

Member
SPLITTER!!!! and two separate tuners!!
Do drip feed it makes my life fun.

Pic of the splitter and cables will confirm my suspicions...
I am intruiged as to what your suspicions are.

Picture 1 showing just the splitter and the tuners:
s1.jpg

Picture showing the full horror:
s2.jpg
(Tuners fed via USB3 hub to Pi4. Also going via the hub is the hard disk for storing the recordings).
 

Rodders53

Distinguished Member
I am intruiged as to what your suspicions are.

Picture 1 showing just the splitter and the tuners:
View attachment 1504157
Ah not what I expected at all... (A plastic Y-splitter mis-connected as people often do with Belling Lee TV plugs and sockets).

THAT is a FILTERED splitter so no wonder it doesn't work properly!

SAT output passes 950-2400 MHz
TV out passes 5-864 MHz
as printed on the label.

The tuner fed from the SAT port (LHS) will have no UHF TV signal at all... other than direct cable pickup and/or something severely attenuated thro' the filter. Circa 25dB as it'll be similar to this Triax equivalent (combiners and splitters are the same device just go the other way). https://www.triax.uk/products/terrestrial/filters/mast-filters/triax-110-tv-sat-combiner-diecast

You need a simple full-band splitter e.g. PROception Wideband TV & Satellite Splitter 2 Way to replace that and all should work much more reliably.
 

delback

Member
THAT is a FILTERED splitter so no wonder it doesn't work properly!

SAT output passes 950-2400 MHz
TV out passes 5-864 MHz
as printed on the label.

The tuner fed from the SAT port (LHS) will have no UHF TV signal at all... other than direct cable pickup and/or something severely attenuated thro' the filter. Circa 25dB as it'll be similar to this Triax equivalent (combiners and splitters are the same device just go the other way). https://www.triax.uk/products/terrestrial/filters/mast-filters/triax-110-tv-sat-combiner-diecast

You need a simple full-band splitter e.g. PROception Wideband TV & Satellite Splitter 2 Way to replace that and all should work much more reliably.
Oops. A Homer Simpson moment from me, then. (TBH I never looked at the fine print on the splitter!)

What's interesting, though, is that the SAT output is feeding the DVB-T tuner, which is making rock solid recordings of Freeview SD channels (group A). It's the unfiltered TV output that's feeding the DVB-T2 tuner, which is making generally good recordings, but with occasional brief breakup.

I did say I have a strong signal - perhaps the enough of the lower frequencies are getting past the SAT output filter to allow the DVB-T tuner to work.

But getting a full-band splitter obviously makes sense, so that's the first thing I'll try. If I still have problems, I'll try replacing the cheap DVB-T2 tuner. Many thanks for your help.
 

Rodders53

Distinguished Member
Location? (approx) to estimate signal level from Wolfbane.
Aerial in use? any amplification on it?

A splitter is -4 dB (-3 dB is half).
The filtered unit is -22 dB on the sat out at UHF I think, -2dB on the TV out. (With higher losses at the higher sat frequencies.)

You probably need to lose any amplification and/or get a few attenuators to reduce the signal in to both USB tuners. -12dB and -6dB can be placed in series to give -18dB and with a -4dB splitter would be -22dB. Variable attenuators are available, too - but I'm not keen on them for permanent use.
 

delback

Member
Location? (approx) to estimate signal level from Wolfbane.
Aerial in use? any amplification on it?

A splitter is -4 dB (-3 dB is half).
The filtered unit is -22 dB on the sat out at UHF I think, -2dB on the TV out. (With higher losses at the higher sat frequencies.)

You probably need to lose any amplification and/or get a few attenuators to reduce the signal in to both USB tuners. -12dB and -6dB can be placed in series to give -18dB and with a -4dB splitter would be -22dB. Variable attenuators are available, too - but I'm not keen on them for permanent use.
Wolfbane says I'm 3 miles from Stockland Hill and the signal should be 85dBuV. It recommends a set-top aerial.
My actual aerial is a Blake BLA-LP36T 36-element log periodic in the loft, with no amplification.
The cheap signal strength meter I have indicates 70dBuV at the input to the splitter - but who knows how accurate it is?

I don't think overload is the problem. The splitter was only inserted when I added the DVB-T2 tuner. Before that, I was getting perfect recordings (SD) on a direct feed from the aerial. But then again, perhaps the DVB-T2 tuner is less tolerant of being overloaded. I have an attenuator kicking around somewhere I could try. But the first thing to do is clearly to change the splitter to an unfiltered one.

I suspect I may end up having to replace the DVB-T2 tuner with something a bit more reputable. Any recommendations? I only need a single tuner, as MythTV supports multirec.
 

Rodders53

Distinguished Member
Based on those numbers:

85 dB outside -10dB roof materials + 10dB antenna gain - 2dB cable and termination loss = 83 dBuV at the set.
Via the filter-splitter perhaps say 61 (sat) and 81 (tv).

But those are theoretical figures. Practical measurement in real life may well be different. (Roof may have more or less loss for example).

Cheap meters do some funny 'averaging' of the RF signal across all received frequencies. Pro meters tune to each multiplex in turn to measure signal, noise and bi-error rates (and best have a spectrum analyser display to see the waveforms, too).

Recommended ideal range is 45 to 65 dBuV. Below that = lack of signal and problems. Above that the potential to overload tuners. Some tuners can and do cope better than others.

That the -T device has coped with full: if you meter is accurate 70dBuV and 48 dB (via sat port) is reassuring. If the theoretical is reality almost amazing it doesn't overload!

Acid test is to direct connect the -T2 stick to the aerial and see how it copes with full output from the aerial on its own? and also via an attenuator at progressively lower signal levels. That the -T2 is exhibiting overload issues at 68 and doesn't work at 48 may just mean it's very 'particular' about its input level needs?

Note also it may be influenced by interference from items nearby if it's poorly shielded, say, so it may be worth experimenting with its location vs other items, and such like stuff before condemning it.

Afraid I have no direct experience with such USB tuner devices. MythTV and similar boards may have uses with such expertise?
 

delback

Member
Latest:
I replaced the splitter with an unfiltered one, and have experimented with feeding the DVB-T2 tuner via the splitter and direct from the aerial supply, both with and without a 12dB attenuator. Results are pretty much the same regardless of configuration: good recordings with occasional brief moments of breakup.

One interesting observation (and I have no objective measurement to back it up, it's just a gut feeling) is that the incidence of breakup seems to be more frequent on BBC channels compared to ITV/C4/C5. All of these are broadcast on the same MUX, so it can't be anything to do with different frequencies. BBC broadcasts have a higher bitrate - does this tell us anything?
 

Rodders53

Distinguished Member
BBC Free to View Limited operate the Coding and Multiplexing of the BBC B HD multiplex(es). There is more than one (Scotland, NI, Wales and a few in England supporting the different ITV HD regions, with more promised for BBC English Regional news).

They are all statistical muxes so the available data is shared between all TV channels. It's most unlikely that any one channel always gets more data rate than any other by design - it's all allocated according to need (with some limitations of minimum and maximum in the programming of the C&M). Fast-moving items need more data than static captions/talking heads - so the data needy steal from those in data surfeit.

DVBT DVBT2 Bitrate Monitoring BBCB-PSB3 DTT UK London / National to me doesn't show BBC rates as especially different to the other PSBs in HD? (Note these are snapshots of a moment in time: 13:16 on 3rd May when I post this reply).

AFAIK, the full multiplex datastream will be tuned in by the stick and all that data (40Mbit/s) passed over the USB link for the appropriate video, sound, subtitle etc.,. streams to be extracted for the TV channel(s) wanted; either for storing or live decoding in the computer.

Within the computer / decoding software is probably where I'd be looking for the glitching/bottlenecks? If the tuner isn't in an overload/signal distortion/error-making mode?

70 dB -4dB splitter = 66dB - 12dB = 54 dB
83 dB -4dB = 79 dB -12dB = 67 dB
I'd like to see how it works with even more attenuation / find where it starts to break up from lack of signal? If you have anything to do that extra attenuation?
 

delback

Member
BBC Free to View Limited operate the Coding and Multiplexing of the BBC B HD multiplex(es). There is more than one (Scotland, NI, Wales and a few in England supporting the different ITV HD regions, with more promised for BBC English Regional news).

They are all statistical muxes so the available data is shared between all TV channels. It's most unlikely that any one channel always gets more data rate than any other by design
Thanks for that. I was basing my assumption on the relative sizes of the recorded programmes. I had noticed that BBC recordings were significantly bigger than ITV/C4/C5 for the same length of program. However, that was a memory from the size of SD recordings. Looking at the HD recordings, I can see they are much closer in size. So my apologies for jumping to the wrong conclusion.

I will reiterate that my impression that BBC channels get worse breakup than ITV is just a gut feeling, and could just be coincidence.
AFAIK, the full multiplex datastream will be tuned in by the stick and all that data (40Mbit/s) passed over the USB link for the appropriate video, sound, subtitle etc.,. streams to be extracted for the TV channel(s) wanted; either for storing or live decoding in the computer.
Yes, that's my understanding. The MythTV system uses LinuxTV and supports multirec - it can extract multiple streams from a single MUX, so it's possible to record multiple programmes simultaneously from a single tuner.
Within the computer / decoding software is probably where I'd be looking for the glitching/bottlenecks? If the tuner isn't in an overload/signal distortion/error-making mode?

70 dB -4dB splitter = 66dB - 12dB = 54 dB
83 dB -4dB = 79 dB -12dB = 67 dB
I'd like to see how it works with even more attenuation / find where it starts to break up from lack of signal? If you have anything to do that extra attenuation?
I might try adding in the unfiltered leg of the second splitter to get a few more dB of attenuation to see what happens.

Thanks again for all your advice.
 

delback

Member
Just wanted to report what I've discovered.

1. Adding various amounts of attenuation appears to make little difference. If anything, adding 12dB of attenuation just made things worse.

2. Being suspicious of the cheap Chinese DVB-T2 tuner, I bought a Hauppauge WinTV-DualHD. This has two nominally identical tuners (Silicon Labs Si2157), fed via a single aerial cable. I set it up to simultaneously record a program in HD and SD using the two tuners. The HD recording had interference and the SD recording did not. This suggests to me that something about the signal being delivered by the aerial is less suitable for HD than SD, despite Rodders' statement that they should be received equally well.

The only other difference that occurs to me is that the HD broadcast is on channel 29 while the SD is on channel 26. I can't imagine that such a small difference in broadcast frequency could possibly be significant.

Although the signal strength is good (according to my cheap meter), I have no idea about its quality (eg. if the aerial isn't perfectly aligned, or its location inside the loft might affect it). Can a poor quality signal make HD reception less reliable than SD?

My next step is going to be to replace the RG6 downlead with CT100, and move the recording setup to a different location further away from various mains power leads (that I guess might possibly cause interference). If that fails, I'll get a local aerial installer in to take a look.
 

The latest video from AVForums

Samsung Micro LED, Mini LED, NEO QLED TVs and More: AVForums Podcast Interview
Subscribe to our YouTube channel

Full fat HDMI teeshirts

Support AVForums with Patreon

Top Bottom