Every so often, with live TV on my LG 43" 15 year old LCD TV, half the screen becomes garbled. Pic here:
Could this be the aerial, or something else? How best to test the aerial?
faulty RAM? faulty ribbon connector to display panel?
I don't see how an aerial could cause that.
I think the TV is gagging on Victoria's impressive mammaries :-)
TBH, joking apart, RF interference/ low signal wouldn't be just part of the screen. I suspect there is a hardware fault in the TV
That does not look like a typical aerial signal issue - more likely a TV fault - dodgy TCon board, or perhaps a poor flex connection to the screen.
On the other hand, if no other sources are implicated, then it can hardly be display related either. It sounds like its in the part of the set that processes the signals from off air to the place where the source switching occurs
Have you found a pattern in which channels are affected? I mean, have a check to see which channels are on each multiplex and see if it aligns with when it happens. Brian
My though. My only question was this intermittent, worse in cold or hot. Is it temporarily resolved with a good old fashioned tap to the case?
If it is only live TV, I’d check the antenna and coax first. Has a tree grown in the path? Could the antenna have been moved? Coax damaged? Is the plug connection good?
Those are ‘easy’ things (almost) any DIYer can check.
(I’d not rely too much on the signal measurements due the problem being intermittent. )
Don’t assume the coax can’t be damaged just because it is in the loft. Years ago, we had a squirrel chew a coax in our loft.
Don't think that's RF problems.
Have you Googled using the set make and model and this fault? In general, few faults are unique to just one example. If common, you may even find a repairer for this exact fault on Ebay.
Signal really needs to be 100%. We only see problems when there is extremely heavy rain and on the borderline signal channels.
It is a low signal failure where the MPEG stream has a bad block that passes the CRC and results in insane decoding. It typically flings the image sideways and turns the picture into psychedelic colours or pastel.
It should correct itself when the next iframe comes along but talking heads are very prone to being screwed up for extended periods. There is a similar low signal fault where the iframe itself is corrupted and repaints only part of the screen leaving a hybrid of two scenes.
Yes it will. The MPEG stream is corrupt. The image up to the point where decoding failed is fine and the parts after it have wrong colours and some random positional shift relative to the rest of the image. Once a bad block has been interpreted the image decoding presses on regardless doing its best to make sense of what is by then a random bitstream.
The fault will persist until the next iframe comes along to repaint the entire screen.
We haven't had sufficient description of the fault from the owner to know what's happening, if it occasionally glitches like that, it could be corrupt p-frames "mangling" the output, normally it looks a lot more random than that with corruopt streams though, but as you say, it would all reset again at the next i-frame.
But if it stays with that "cut in quarters" appearance for more than a couple of seconds, I'd say it's faulty hardware or connections inside the TV.
The way to tell would be to have a close up of the pixels at the transition between good picture and bad. Armed with that pixel level info I can quickly determine if it is an MPEG decoding error. Resolution at full frame isn't up to seeing that level of detail.
The damage will show as one 16x16 block where an 8x8 sub-block is only partially right and then the next block right or down is some crazy colour. It invariably throws the picture sideways by a random shift.
If the fail always occurs in the same place then I'd agree it is most likely hardware memory fault but I have seen quite few variants of the same thing on marginal satellite TV during extreme thunderstorms with thick cloud preventing an adequate signal from reaching the decoder.
The failure is suspiciously close to half way across the screen so either interpretation is possible at the moment.
The more common decoding failure mode is a few blocks go haywire and the decoder bales out leaving correctly coloured cubist artefacts and maybe the odd ultrasonic click or gap on the sound track.
I had a fault looking very similar on a computer monitor. No MPEG involved
I'm not saying it is definitely a decoding fault, but that is pretty much what an MPEG decoding fault looks like. The sideways throw and colour shift can vary enormously when this happens.
A hardware fault in the memory addressing is also possible. Made much more likely if it always occurs in the same place on the screen.
That is not a reception or tuner problem. DVB coding uses interleaved data, you cannot get errors segregated in that way across a image
I haven't left it in that stae for too long - maybe a minute or two. As I say, changing channel solves it.
I'll make a note of the citrcumstances next time. And have a look at the aerial/cables. For now, though, it's not much more than a minor inconvenience.
Thanks for all the pointers.
<panto mode>Oh yes you can!</panto mode>
I grant you that it is rare and requires the signal to be very close to dropping out entirely but a TV on a digital stream can be provoked to do exactly that and a couple of other weird variants if the MPEG stream for a key frame is incorrect after the decoder has finished with it.
Usually you get 16x16 blocks that are wrong but sometimes it will do almost exactly that linear shift and wild crazy colour change.
It could still be a memory fault though.
If it always occurs in the same place then it *is* a memory fault or a defective connector on the display itself. If changing channel fixes it or it only ever appears on one channel then it is a firmware bug in the decoder (there are plenty of those too). Unplayable media content.
Several older sets have some of the p frame coefficients in the wrong way around. If you look very closely when the news is on and they pan across a table with a shallow diagonal line it breaks into a small sawtooth.
Running people also shed more bit error smoke in the sky than they ought to. Those are the only symptoms of that particular bug.
Its not RF based. Its a hardware fault.
I might have hoped by now you would have dismissed areas such as the aerial and cables from being the cause!
Yes yes, but Martin Brown seems to be making a comprehensive case for a possible aerial issue.
HomeOwnersHub website is not affiliated with any of the manufacturers or service providers discussed here. All logos and trade names are the property of their respective owners.