DS Forums

 
 

BBC DAB reception issue


Reply
Thread Tools Search this Thread
Old 28-01-2016, 09:53
Darcy72
Forum Member
 
Join Date: Jul 2011
Location: South Yorkshire
Posts: 52

I live on the Worksop side of Sheffield and for the past two days the reception of the BBC National Ensemble has been virtually non existent indoors.
All other stations are fine and the BBC reception page says no reported problems with the three local transmitters - Sheffield, Clifton and Emley Moor.
Reception on this ensemble is usually the best anywhere in the house so has anybody else locally noticed a change ?
Darcy72 is offline   Reply With Quote
Please sign in or register to remove this advertisement.
Old 28-01-2016, 13:25
vinnielo
Forum Member
 
Join Date: Jan 2002
Location: South East England
Posts: 5,697
The BBC are aware of the problem.
vinnielo is offline Follow this poster on Twitter   Reply With Quote
Old 28-01-2016, 14:25
Darcy72
Forum Member
 
Join Date: Jul 2011
Location: South Yorkshire
Posts: 52
Aha!
Transmitter information finally updated, I suspected reduced power from Clifton, thanks for the prompt Vinnielo.
Darcy72 is offline   Reply With Quote
Old 28-01-2016, 23:34
Andy_Burns
Forum Member
 
Join Date: Jan 2016
Posts: 3
ON Wednesday 27/01/16 I was working in the M18 area of south Manchester and BBC radio 5/world service kept cutting out, and at several places even parked up no BBC DAB, D1, Manchester mux and the local nio cast perfect. I`m back in the area tomorrow and will check again.
Andy_Burns is offline   Reply With Quote
Old 29-01-2016, 07:40
Mark C
Forum Member
 
Join Date: Dec 2001
Location: Hampshire
Posts: 7,954
ON Wednesday 27/01/16 I was working in the M18 area of south Manchester and BBC radio 5/world service kept cutting out, and at several places even parked up no BBC DAB, D1, Manchester mux and the local nio cast perfect. I`m back in the area tomorrow and will check again.
I had the same problem in Surrey on Wednesday morning. Between 09:15hrs and when I arrived at my destination at 10:10.

BBC Mux kept cutting out, this was from the M3 at Bagshot, and round the M25 to Reigate.

All other muxes were perfect, as is BBC reception normally on that route
Mark C is offline   Reply With Quote
Old 29-01-2016, 09:09
Darcy72
Forum Member
 
Join Date: Jul 2011
Location: South Yorkshire
Posts: 52
ON Wednesday 27/01/16 I was working in the M18 area of south Manchester and BBC radio 5/world service kept cutting out, and at several places even parked up no BBC DAB, D1, Manchester mux and the local nio cast perfect. I`m back in the area tomorrow and will check again.
Careful! - you'll upset the natives! - the M18 isn't and will NEVER be part of Manchester despite the alleged "Northern power house"

Full signal now restored by the way.
Darcy72 is offline   Reply With Quote
Old 29-01-2016, 19:18
Andy_Burns
Forum Member
 
Join Date: Jan 2016
Posts: 3
I was referring to post codes, and I was back there today and BBC was perfect, they must have been doing engineering work.
Andy_Burns is offline   Reply With Quote
Old 30-01-2016, 00:03
WellHiddenMark
Forum Member
 
Join Date: Jun 2009
Location: Not-London
Posts: 1,654
Careful! - you'll upset the natives! - the M18 isn't and will NEVER be part of Manchester despite the alleged "Northern power house"

Full signal now restored by the way.
But M18 is most definitely east Manchester and not south.
WellHiddenMark is offline   Reply With Quote
Old 30-01-2016, 09:08
Darcy72
Forum Member
 
Join Date: Jul 2011
Location: South Yorkshire
Posts: 52
I was referring to post codes, and I was back there today and BBC was perfect, they must have been doing engineering work.
Apologies - I misread your reply and assumed you were confirming my problem.

I now have an official answer from the BBC:

Dear Audience Member

Thank you for contacting the BBC Reception Advice service.

We understand that you were having issues with your DAB reception earlier this week. This happened due to disturbances of the time reference equipment which resulted in national DAB suffering outages since late Tuesday evening. The outages were caused by a rogue GPS satellite (SVN23), which was taken out of service in the evening of 26 January.

This issue has now been resolved and you should be able to enjoy service as normal on the full list of BBC stations. We do apologise for any inconvenience this caused you.

Thanks again for taking the time to contact us.

Kind Regards

BBC Audience Services
Darcy72 is offline   Reply With Quote
Old 30-01-2016, 11:50
Gerry1
Forum Member
 
Join Date: Sep 2005
Posts: 2,411
This happened due to disturbances of the time reference equipment which resulted in national DAB suffering outages since late Tuesday evening. The outages were caused by a rogue GPS satellite (SVN23), which was taken out of service in the evening of 26 January.
Whoops !

So in a national emergency if a GPS satellite fails or the GPS system is shut down by the US Government (or shot down by Another Government) then the UK loses its emergency broadcasting network?

Scrapping Long Wave doesn't seem a good idea...
Gerry1 is offline   Reply With Quote
Old 30-01-2016, 13:24
Mark C
Forum Member
 
Join Date: Dec 2001
Location: Hampshire
Posts: 7,954
Whoops !

So in a national emergency if a GPS satellite fails or the GPS system is shut down by the US Government (or shot down by Another Government) then the UK loses its emergency broadcasting network?

Scrapping Long Wave doesn't seem a good idea...
Well, I'm not surpised the problems experienced were the result of some sort of GPS sync error, but hang on, all DAB networks rely on GPS along with countless other services, how come they were seemingly unaffected, I do detect the possibe aroma of BS here
Mark C is offline   Reply With Quote
Old 30-01-2016, 13:33
moox
Forum Member
 
Join Date: Oct 2004
Posts: 14,377
Well, I'm not surpised the problems experienced were the result of some sort of GPS sync error, but hang on, all DAB networks rely on GPS along with countless other services, how come they were seemingly unaffected, I do detect the possibe aroma of BS here
Can't comment on the BS-ness of it, but http://www.navcen.uscg.gov/?Do=gpsShowNanu&num=2016008 would suggest an issue with the satellite (it's a shame that the US military is still rather "concise", we're not using teletypes anymore)

Not being a broadcast engineer, would there be any way to integrate alternate time sources (e.g. Galileo when that comes online) into the network?
moox is offline   Reply With Quote
Old 30-01-2016, 13:53
lundavra
Forum Member
 
Join Date: Dec 2009
Posts: 25,074
Whoops !

So in a national emergency if a GPS satellite fails or the GPS system is shut down by the US Government (or shot down by Another Government) then the UK loses its emergency broadcasting network?

Scrapping Long Wave doesn't seem a good idea...
Though the networks distributing the programme for the Long Wave service will quite likely be using GPS as time and frequency standards. I presume the drives at the transmitter itself have other references but would not be surprised if GPS failure caused a problem there as well.
lundavra is offline   Reply With Quote
Old 30-01-2016, 15:45
Mark C
Forum Member
 
Join Date: Dec 2001
Location: Hampshire
Posts: 7,954
Can't comment on the BS-ness of it, but http://www.navcen.uscg.gov/?Do=gpsShowNanu&num=2016008 would suggest an issue with the satellite (it's a shame that the US military is still rather "concise", we're not using teletypes anymore)

Not being a broadcast engineer, would there be any way to integrate alternate time sources (e.g. Galileo when that comes online) into the network?
Well, that's an excellent find ! The report says the satellite was taken out of action Mon 25th at 22:00hrs ZULU (which is more or less GMT), so well before Wednesday morning.

And anyway, GPS satellites are in medium earth orbit so at any given place they pop up above the horizon and and move across the sky, only being visible in one location for a few hours, and not reappearing potentially for a few days

http://gpstracklog.com/2013/09/good-satellite-days.html

My guess is it was the Beeb's satellite distribution that was having problems on Wednesday morning. All Txs are fed by satellite, but the 'heritage' DAB sites (CP, SC, Emley, Winter Hill etc) have a line back up. Any DAB Tx that loses its programme feed should shut down, to avoid punching a hole in the SFN. Perhaps the sat feed was coming on and off, and the network reverting to core sites, and back again. That would explain why I was having problems in Surrey, and the others in Manchester and Yorkshire at the same time ? If that was it, it'll give some idea what D2 reception will be like

SFNs require a time reference, any will do, but GPS is the only easy one available at present
Mark C is offline   Reply With Quote
Old 02-02-2016, 16:00
Darcy72
Forum Member
 
Join Date: Jul 2011
Location: South Yorkshire
Posts: 52

So finally the BBC get round to an article:

http://www.bbc.co.uk/news/technology-35463347

And the USAF had this to say:

http://www.itnews.com.au/news/satell...anomaly-414237

I too share Gerry1's concerns:
So in a national emergency if a GPS satellite fails or the GPS system is shut down by the US Government (or shot down by Another Government) then the UK loses its emergency broadcasting network?
Scrapping Long Wave doesn't seem a good idea...
Darcy72 is offline   Reply With Quote
Old 02-02-2016, 16:15
Mark C
Forum Member
 
Join Date: Dec 2001
Location: Hampshire
Posts: 7,954
So finally the BBC get round to an article:

http://www.bbc.co.uk/news/technology-35463347

And the USAF had this to say:

http://www.itnews.com.au/news/satell...anomaly-414237

I too share Gerry1's concerns:
So in a national emergency if a GPS satellite fails or the GPS system is shut down by the US Government (or shot down by Another Government) then the UK loses its emergency broadcasting network?
Scrapping Long Wave doesn't seem a good idea...
I'm still sceptical. Are there any reports of other users having problems, GPS is a very widely used timing platform. No other DAB network was affected. Mobile phone networks use it to lock the timing slots I think, did any mobile phone networks go down ? Did ambulances and delivery drivers turn up 4km in the wrong places.

I don't doubt the GPS satellite had a wobbly, but I'd have expected a torrent of reports
of no end of affected systems, rather than a few of us being unable to listen to Ken Bruce last week !
Mark C is offline   Reply With Quote
Old 02-02-2016, 20:21
Vectorsum
Forum Member
 
Join Date: Apr 2011
Location: Future EU Scottish Republic
Posts: 808
I'm still sceptical. Are there any reports of other users having problems, GPS is a very widely used timing platform. No other DAB network was affected. Mobile phone networks use it to lock the timing slots I think, did any mobile phone networks go down ? Did ambulances and delivery drivers turn up 4km in the wrong places....
The linked BBC article is wrong, there was no "4km error" in derived positioning as all the GPS birds, even the affected bloc, were still broadcasting the correct almanac and ephemeris.

GPS receivers could therefore all still solve correctly for 'x', 'y' and 'z', it was 't' that wasn't being solved correctly, and this only because all birds in the bloc that were affected all of a sudden picked up effectively an incorrect backup timing reference when SVN23 dropped out of the constellation. GPS receivers are designed to 'over-solve' and remove satellites from the solution(s) that they find have a 't' which is different from two or more other satellites - they can't do this if 't' is wrong over a number of visible birds.

It was an error in the ground-segment timing software that couldn't have been either anticipated or tested as SVN23, by far the oldest NAVSTAR bird, was effectively 'pinning' the constellation at the correct legacy timing; when it was removed the default timing reference changed to the modern, erroneous one for a large bloc of birds a generation or two on from SVN23.

Investigation about what went wrong at the Beeb DAB TXs is probably still ongoing but my guess would be that their high-quality GPS timing references all of sudden noticed an abrupt 13 microsecond step change in UTC, thought "sod this" and shut down. The TX timing reference would have been left to 'freewheel', which it can quite happily do for a few seconds, not for the hours/days of the effective error. The affected TXs should really have shut down.

As to the vulnerability of the GPS birds, forget it. Recall that their main mission is nuclear launch detection, transmitting positioning information is an 'add-on'. If daesh turned up in the Hebrides with a jeep full of rocket-propelled grenades, I'd quite happily hide behind one of these NAVSTAR thingies, and expect it still to be working afterwards. In any case, GLONASS and Galileo are interchangeable with GPS, so with a firmware mod a receiver can use any or all of the constellations.

By the way UMTS, although being a wideband CDMA system, is designed as an asynchronous system as opposed to cdmaOne which is synchronous and slaved to GPS timing, for precisely the purpose of being invulnerable to this sort of thing. It also makes it far easier to put micro and nano cells in indoor locations without having to poke a GPS antenna through the roof...
Vectorsum is offline   Reply With Quote
Old 03-02-2016, 08:00
Mark C
Forum Member
 
Join Date: Dec 2001
Location: Hampshire
Posts: 7,954
Very interesting, thanks Vectorsum
Mark C is offline   Reply With Quote
Old 03-02-2016, 10:41
commseng
Forum Member
 
Join Date: Jul 2009
Location: London
Posts: 695
A quick question, probably for Vectorsum, whose knowledge about GPS is quite impressive - if a whole constellation are using an incorrect reference, would that mean that they are all sending timing data that is consistent but wrong?

Why then would the GPS receivers not lock to this incorrect data, after a suitable period?

As long as all the transmitters in the same area had the same data would there be a problem?
Presumably the same incorrect data would be time stamped onto the mux at BH,?
However if that was different - presumably working off the atomic clock rather than GPS, could that have been the issue?

Please feel free to point out the errors in my thinking!
commseng is offline   Reply With Quote
Old 03-02-2016, 17:33
Vectorsum
Forum Member
 
Join Date: Apr 2011
Location: Future EU Scottish Republic
Posts: 808
A quick question, probably for Vectorsum, whose knowledge about GPS is quite impressive - if a whole constellation are using an incorrect reference, would that mean that they are all sending timing data that is consistent but wrong?
Yes, and that's exactly what happened. The 'constellation' in this case being all birds still taking the legacy L1(C/A) timing feed so presumably at least the Block IIR sats and maybe also some of the Block IIR-M sats where L1(C/A) was notionally replaced with L2C for open, civilian use.

SVN23 was the last of the sats from the Selective Availability days of L1(C/A) so my simplified guess is that someone plugged the remainder of the non SA sats directly into the timing reference rather than via the software that used to 'wobble' time, not realising that there was a delay of several microseconds transiting the old, dormant SA software relative to the direct path. I can't quite believe someone would be that slapdash, but we're talking about jarheads here.....

The US military are being surprisingly coy about the exact reason and failure mode, probably because if they gave a full explanation of how time is used and routed in the ground sector, they'd leak information about how relative sat timings are used as part of the 'toolkit' to deny GPS availability to potential aggressors on a regional basis.

Why then would the GPS receivers not lock to this incorrect data, after a suitable period?
My guess is as above, that what happened in the Beeb's and other organisations that 'lost' timing is not that the receivers all of a sudden were unable to solve for UTC at the WGS84 reference geoid surface, just that as they were happily doing this all of a sudden it shifted by several microseconds relative to their internal precision timebase. They shouted foul and at best marked any time output of theirs as 'suspect' if not shut down and refused to output anything altogether until the user intervened.

Depending on the model and vintage of the time receiver in use, it may also be using L2C timing signal which all of a sudden would have been different from the L1(C/A) on the same satellite. Long story short these civilian signals are on different L-band frequencies for the same reason that the encrypted military P(Y) signal is on both, so that the path distortion on transiting the ionosphere/troposphere can be inferred. An L1/2 timing receiver experiencing an abrupt path difference of 4km between the two signals might have concluded that the ionosphere had turned into pea soup. It's not unreasonable for it to think, "hmm, better not output anything" in such circumstances.

As long as all the transmitters in the same area had the same data would there be a problem?
In theory no, if they're still all splurting out correct almanac and ephemeris, in fact a receiver turned on just after the step error in time was introduced would have no way of knowing anything had happened. Rather I'm guessing that the problem with timing was as above, that the step change in calculated UTC caused sophisticated receivers to blow chunks, as they would probably be designed to do.

Presumably the same incorrect data would be time stamped onto the mux at BH,? However if that was different - presumably working off the atomic clock rather than GPS, could that have been the issue?
There's nothing stopping the Beeb doing that, and in fact the post-mortem from all of this may be that incorporating a relative timing facility onto the sat feed may be a Good Thing.

The main problem is that geostationary satellites, as used by the Beeb for the ETI feed, are anything but stationary. In fact, they wander around in a notional cube of about 100km on a side under the command of the ground station, which means that the path length to each of the DAB TXs is constantly varying.

OTOH, if it were somehow possible to get a feed from the satellite operator of exactly where in the sky over the orbital position the relay bird was instantaneously located, it would be possible for the receiver, at the fixed location of the TX on the ground, to calculate how much delay to impose on the outgoing signal in order to comply with any absolute or relative transmission timings.
Vectorsum is offline   Reply With Quote
Old 03-02-2016, 17:49
commseng
Forum Member
 
Join Date: Jul 2009
Location: London
Posts: 695
Thanks for such a detailed answer, I think I am going to have to do a bit of reading up on GPS to get my head around some of that.
I guess it will not be helped by GPS having a military primary use as you say, and therefore some details being kept secret.
commseng is offline   Reply With Quote
Old 03-02-2016, 18:34
Vectorsum
Forum Member
 
Join Date: Apr 2011
Location: Future EU Scottish Republic
Posts: 808
Thanks for such a detailed answer, I think I am going to have to do a bit of reading up on GPS to get my head around some of that.
I guess it will not be helped by GPS having a military primary use as you say, and therefore some details being kept secret.
It's actually surprisingly open, considering that the main mission is nuclear oversight and giving Polaris boats an initial reference for missile launch. You have Al Gore and the Clinton Administration of the nineties to thank for that and, a few years earlier, Reagan for kicking the whole openness initiative off. To think that such a colossal prick could have been capable of anything worthwhile stretches the imagination but it's true, nevertheless.

After Korean Airlines 007 the Americans, who were sitting largely in secret on what would have been a perfect navigation solution that would have prevented such loss of life, realised that they'd better proliferate it. Think of it as the USA's gift to the world, a bit like Blighty's gift to the world was association football, cricket and decent legal systems.

Back in the day the standard text used to be Kaplan's 'Understanding GPS: Principles and Applications ' from Artech House, but I've just checked the IET online library and there is a book there, 'Global Positioning Systems, Inertial Navigation, and Integration, Second Edition' available for download if you can't get your hands on Kaplan. The IET text will also have the low-down on the new L3/4/5 'Safety of Life' services of latter years.
Vectorsum is offline   Reply With Quote
 
Reply



Thread Tools Search this Thread
Search this Thread:

Advanced Search

 
Forum Jump


All times are GMT. The time now is 06:01.