HDMI cable interference

[deeptho]

The last couple of days, we started loosing TV reception on cable TV: from time to time
heavy pixellation occurred on many channels on two muxes. This lasted for several minutes
and then the problem disappeared for a while.

I finally found out that the problem was due to interference from a 3 meter long HDMI cable
in a room in a completely different part of the house. The Aha-momemnt came when I discovered that
the problem went away when the computer activated the screen saver. Apparently sometimes
the screen woke up because the mouse reacted to small vibrations, but because the monitor is
mostly off and the computer mostly on, I did not notice.

I then replaced the HDMI cable by another one, and the interference became even stronger.
All TV pictures went away fully. The PC is not even connected to the cable system but a cable coax comes
close to the hdmi cable at some point.

I noticed the strongest interference on 147MHz, 294 MHz (the former only when the PCs screen is ot
blanked) and 300 MHz (constant even if screen is blanked). The 294MHz caused the picture problems.
Interestingly I can change the center frequencies of the interference by changing
the PC's screen resolution and I can spread it around by playing a video.

So here is the question: are these HDMI cables really that badly shielded? The first one was a regular round cable and
was installed many years ago. I never noticed problems. There could have been interference already but perhaps
not yet strong enough to notice.

I did disconnect the pc recently and reinstalled it in the same place. So perhaps some minor damage to the connector
could cause problems like this?

The second cable is a flat ribbon cable (and brand new).

 

[Terryl]

One thought, is the cable TV coax properly grounded? If not then there could be leakage or cross talk between the two.

Also take a look on google for some snap on ferrite beads of flat ones, they can cut down on external RF on the shielding.

 

[rolfw]

Have you thought about perhaps it not being the HDMI lead, but the monitor?

 

[Channel Hopper]

or an issue with the ring main ?

 

[deeptho]

One thought, is the cable TV coax properly grounded? If not then there could be leakage or cross talk between the two.

Also take a look on google for some snap on ferrite beads of flat ones, they can cut down on external RF on the shielding.

It depends on the definition of "properly". The cable system has been installed by the cable company and the cable runs underground. The modem is not too far from the ground rond but is only connected to that through the electricity earth.

In any case, I know that ground loops can be a big problem, but it is probably not that. I will post some
results below that will probably prove this.

 

[deeptho]

Have you thought about perhaps it not being the HDMI lead, but the monitor?

Of course, but the interference is also present with the monitor off. It is a flat panel dell monitor.

 

[deeptho]

Today I did some experiments with 2 other hdmi cables. One of them reduced the interference
level buy 9 dB. The second one was a bit shorter (2m instead of 3m) and thicker and reduced
the interference by 20dB. This was measured with a rod antenna connected to a usrp and with sdrpp
(only relative measurements possible).

With that second cable, the interference when measured on the coax cable became invisible on the spectrum,
but I could still here in the am demodulation that was on the switching on and off of the hdmi signal.

With all three cables I also tried installing a ferrite mantle. That made no difference whatsoever, no matter where I placed it. That is not what I would expect for common mode signals.

In any case it shows a big difference between cables.

By the way: a long time ago when I bought my tv, the seller tried to sell me a really expensive hdmi cable
with gold (coloured) connectors. He claimed it was much better, but I did not believe it. Perhaps he was right after all,
but not regarding "better picture quality" (obviously nonsense) but less interference.

Of course, he did lie to me about the TV: I got it for a reduced price because it was a "shop display model". I asked how many hours it had been playing and he gave a low number. At home, after discovering the secret remote code
to enter service mode, I discovered that the TV had been used for at least 1000 hours. I did not complain afterwards.

The Philips TV is still working fine today. It has DVB-T but not DVB-T2. This show its age. So it has been a good
purchase anyway.

It was also one of the last "non-smart" TVs that you could buy, which is a huge bonus as well: fast channel changing, no nagging about updates, no "phoning home to the company" and no other crap such as the samsung TV that I had and that destroyed itself after pressing accidentally a button in service mode.